Hello fellow builders and lurkers. I've been knocking around these forums for a few months while deciding which router design to go with for my first build. I have previously built my own redesigned version of a manual duplicator based loosely on the Copy Carver. A photo of it is included below.

However, I also wanted to build a CNC router - even as I was working on the previous machine. The machines have distinctively different applications and similarities of course.

For various reasons I decided to build David Steele's 25x37 dual lead screw design. One of the reasons is the ability to raise the machine easily for carving on thick objects. I chose red oak for the wood parts mostly for the stiffness and for the looks, being somewhat of a wannabe woodworker as well. The side rail and y axis boards are laminated 1x4 boards to arrive at the equivalent of the 2x4 boards specified in the Solsylva book.

Current status and changes from the plans book - The table frame is completed and the y axis gantry is completed except for addition of the end plates and trucks. All of the oak blocks for the Z axis have been cut but not drilled. I made the stepper mount plates from 1/4" aluminum plate. I'm currently working on the z axis side plates, which will be made from 1/4" aluminum plate. I have cut the x axis bearing support blocks from Purpleheart but this may change to 3/8" aluminum if it is necessary. The lead screws are going to be 1/2" everywhere possible. I'm not sure at this point if the leadscrews will be allthread or acme thread. I made some 3/4" x 1/2" thick round nuts to go inside the EMT tubing. The nuts were installed in the tubing and a length of 1/4-20 allthread was blue Locktited in place in the nuts. This allows me to hold the tubing and channel in place on the side rail and gantry with a nut and the assembly is easy to remove if necessary. I installed a piece of 5/16" thick by 7/16" wide oak to go inside the channel so that spreading of the channel walls won't be an issue if the mounting nuts have to be torqued a lot.

I hope you find the changes to be entertaining and worth following. Your comments, suggestions, and questions are welcome.

CarveOne

Hey carveone, looks great so far, I seen your thinking on which lead screw to use, Dont even think about it, get the acme 2 or more start, youll only want to change it right away if you use the allthread, take it from me and many others that did the same, I only have the 1 start but my next machine will be more, davids plans are awsome and can be adapted into other designs. Have fun with the build. Graham

Thanks Graham,

I already have the 1/2" allthread and couplers on hand and have determined from email discussions with rc_flyer that acme is the way I want to go. He is in the middle of upgrading his machine now, and I'm continuing to build my remaining machine parts until I can get some money together with sources for an order.

I found that the 1/2-13 allthread from Lowes builders centers (US) is close to the correct diameter and has a rough surface, but the couplers they sell for the rods are too tight. Running one of the rods through a threading die cleans it up a little, pulling out small chips and smoothing the surfaces but running a 1/2-13 tap through the coupler pulls out long strings of metal. Running the recut coupler on the recut rod is now pretty smooth but it has too much freeplay.

Running the recut coupler on a non-recut rod is also smooth and has less freeplay. 13TPI is going to be too slow with my 425 oz/in stepper kit from Xylotex anyway, so I'll go with the acme rods, find the bearings that are needed, and get the anti-backlash nuts from DumpsterCNC.

Oh well, maybe the 1/2-13 rods will make good tomato stakes next spring or find some use in my next project......


CarveOne

I have the same setup as you with the 425's I can get good speed with the y and z axis but the x is not good, 30 imp which I believe the problem is with the belt. I had the y and z axis going 90 imp and could have gone faster but the machine shakes due to whip, so when you get your acme make sure there straight. I think the pull from the belt causes the stepper to go off tilter or somthing, It might be a good idea to use another gear or somthing to drive the pully that drives the belt if you know what I mean lol, think I lost my self on that one lol. What will you be doing with the machine once its done?Oh make sure you use a good computer, over 1 ghz, I used a 667 mghz and it ran slow so I hooked up the pent4 and wow what a diff. Graham

Hey guys..

just put up a thread to deal with the 1/2" acme upgrade on my year old Solsylva router.. Figured a seperate thread made sense since it will help answer a bunch of upgrade issues..

ALSO.. Does anybody out there either know anywhere that sells 12" lengths of 1/2"- 4 or 5 start acme lead screw, without a $60 minimum??

OR does anybody have about 12" left over from a build or upgrade?? Budget for the router is about empty, since we are shopping for a new printing vinyl cutter for my wife's biz...

Don
www.eurekaaircraft.com

Graham,

I'm concerned with the whipping also, and will very likely go with rc_flyer's upgrades that he is doing as we speak. (There'a a few true leaders and there's lots of followers, eh?) Meanwhile, I'm just making the other parts I need anyway.

Speaking of which, I didn't make nearly as much progress as I had hoped this weekend. Making just eight parts took the few hours I had available. See photo below.

I will be using it to make birch plywood and liteply parts for wind tunnel models and possibly some grips for 1911A pistols. Maybe some 3D carvings for woodcraft projects. Anything else that might repay the expense of building it.

CarveOne

Hey CarveOne,
Wow looks pretty spectacular! Even though you say you work slowly it's a lot faster than me....and your workspace is neater/bigger than mine too... looks like I'm losing haha.

I was interested in the upgrades you were talking about, seems like you're going to 1/2" leadscrews haha, and it seems like you must grow a lot of tomatoes haha, (I ended up using the Conduit to bar my windows at home from possible home invasion, the tomatoes will have to wait I guess). I'd like to upgrade to ACME as well, probably 2 start as Graham was suggesting, but I guess it'll be a project for later (CNC + Uni = no money/debt). Is it true though that the torque would be a lot less than with the 1/4"? I guess the ACME ones will be a lot more efficient though, that might make up for some of the power transmission losses.

I had also looked around for anti-Backlash nuts too, and on DumpsterCNC. I noticed somewhere that an anti-Backlash nut doesn't look that complicated to make if you have a CNC already.... It just seemed like you needed to cut 2 pieces of plastic, make them interlocking, then drill a hole through them and tap it. After that, have a spring in the middle that pushes them apart, compress the assembly, and thread it onto your leadscrew. Sure it all "sounds pretty easy" but I'm probably missing a critical detail that makes it really hard to do.... Might try it out later this year in the University machine shop, might need to use a mill to do some more precision stuff than what my CNC will be able to do (seriously worried about slop in the assembly...some shotty incompetent university student was building my machine so I dunno.... ha. ha. ha.).

I liked your round nuts, they look pretty kickass. Is that steel? Should be pretty good and won't rip out of the EMT, unlike those threaded holes, looks sketchy. I also experienced a problem with the rails spreading when I tightened the nuts a lot...is the oak piece used as a spacer or as a reinforcement? If it's a reinforcement how are you attaching it to the channel?

What software are you using by the way, for interface to the controller card (I have the HobbyCNC one)? Mach3? I also got my hands on MasterCAM V9 tutorials and MasterCAM X2 but I have yet to try that out. How will you be going from Solidmodel to G code?

pumelloman,

Don't worry about the tapped holes in the conduit, or the spreading channels. Mine has been running for about a year with no problems with either. I did use steel tubing for the rail and gantry supports, which might have made a difference. I also used allen head bolts instead if hex head bolts to hold the conduit in place.

It seems by some tests I have done here, that people having problems with the above are tightening the bolts a lot tighter than needed, since very few people will be using a torgue wrench to tighten them down. When tightened properly, not 'tightened till the bolt won't turn no more', and using lock-tite, mine have not budged or moved in a year. Tighten them till they are firm, and they will not go anywhere. My test showed that when using a wood frame, it feels looser at the same torgue settings than it does using metal supports. Flattening out the top of the holes, so the washer doesn't have to bend will also help tremendously when tightening them down. That give as the washer is bending is very deceiving on how tight it is.

You will find that David's designs are the best out there, from my experience. They work well when built as described, but lend themselves to upgrades VERY well. His design would make a great all metal machine, with 4 or 5 start lead screws, solid rod instead of conduit, etc. is just depends on what you want to spend, and your building skills. I have been slowly upgrading mine as it makes me money. As time goes on, I'm sure it will be an all metal, kickass machine, that grew as my uses (and spending) dictated. Most other designs wouldn't do that very well. I'm sure if money were no object, A shop-bot or equiv would be great, but knowing me I would have went the Mechmate way then, for the building experience.

Good luck..

pumelloman,

rc_flyer is correct about the threads in the conduit holding the rails just fine as David designed it. I always find some excuse to overbuild everything I do. Bad habit I learned over the years.

The round nuts are made from 3/4" hot rolled steel bar cut into ~ 1/2" thick sections. The 1/4-20 threaded holes go all the way through. They are very easy to install by sliding them into the conduit with a dowel while holding the conduit with the 1/4" clearance holes facing upward. The nuts are wide enough that they won't spin around while running a section of Locktited 1/4-20 allthread into it. If the allthread is not tightened against the opposite wall of the conduit the allthread is easy to align with the holes in the wood side rail.

The 5/16" x 7/16" wood strip is just spot glued into the bottom of the aluminum channel with Tightbond II adhesive. It is there to limit over squashing of the channel walls if I have any trouble with conduit flexing sideways from router bit side loading. It's not necessary, as Don says, but overbuilding it now is better than having problems later and having to mess with realigning the x axis rails again.

I will be buying Mach3 and have TurboCAD 12 for drawing parts. Hopefully the dxf file export will import ok in Mach3. If not, then something else will be sought. Can't afford any high priced CAD packages. Will have to deal with the software issues when the machine is ready to run.

CarveOne

Made a few more parts for the gantry last evening and this morning. Working with the red oak is nice. The blocks are are not showing any tendency to cup after three weeks since I cut them to length, and there has been no finish coats applied yet.

Pardon the wierd looking photo. It's a top down view with a black rubberized mat for a background.

CarveOne

Hey carve one, as long as your wood is nice and dry before cutting every thing then it wont cup,warp or bend,just dont let any moister get in .
Ive noticed a few ppl are a bit worried about losing torque using 1/2" acme with this machine, dont be, theres tons of torque, couldn't stop the machine if I tried. Forgot to turn the router on once and it almost ripped the machine right off the table which is all screwed down, Ive stalled the router a few times, its a PC laminate trimer. So lots of torque. Graham

grahamshere,

This premium grade oak 1x4 lumber is expensive, very dry, and I bought it for the additional rigidity and good looks, and have not had it warp on other home projects I've used it on before. Where I normally have to buy pine lumber the stuff is not at all stable and I wouldn't use it for building one of these machines. Making the frame and gantry out of steel might have been less expensive.

I don't think the 1/2" leadscrew concern is valid unless it happens with rough poor grade allthread rods and/or threaded couplers that are cut too close to the minimum diameter. I could believe that properly sized 1/2" althread and couplers could have a little more frictional drag than the 1/4" or 5/16" allthread and couplers due to the larger mating surface area. This is one reason I ordered the 425oz/in steppers. They will have the additional torque at lower speeds and still have enough torque at the higher speeds (I hope). I tried running an allthread rod through a 1/2-13 threading die to smooth out the rod, and running a 1/2-13 tap through a coupler to eliminate the tight spots, eliminate binding, and it reduced the drag to a very tolerable level. With the anti-backlash nuts this rod will work fine except that the rod has a small serpintine spiral in it that I can't tweak out of it by hand.

rc_flyer is using his machine again now that he has completed his upgrade effort (is it ever really completed?). He seems pretty happy with his new 1/2" rods so far. I don't recall him complaining about x axis rod whipping either.

Gotta get back to making some more parts. Later guys.

CarveOne

Two more gantry parts completed, slotted holes are looking a bit rough though. I'll clean them up a little tomorrow with the 3/16" end mill I should have used in the first place. These were just drilled and filed.

CarveOne

More parts drilled this morning. I didn't counterbore and drill the bearing hole yet due to not knowing for sure what the bearing's OD will be yet. Just drilled a 1/8" pilot hole for now. Drilling the end grain holes rarely goes as planned on wood parts. I used my milling machine and a new, stiff drill bit and it still wandered off center a little even when drilling very slowly to let the drill "find center" before going a little deeper. There is enough clearance for the 1/4-20 allthread that it shouldn't matter. When cutting the blocks a few weeks ago I cut a spare piece just in case this didn't go well enough on the first try.

One more wood piece to drill this evening.

CarveOne

When you drill the tool holder wood center on the Z axis, Don't do any counter boring till you have the router and have it all mounted. With the front recess in the bottom hole for the lead screw mount, you couldn't get to the nut for adjustment, once the router is in place. I fixed that by filling in the recess, and redrilling, (washers would do the same thing) and spacing the router away from the mount enough to get to the 2 adjustment nuts. A shorter router may not have that problem. Till I changed that, I had to remove the router to make adjustments, and the adjustments will change, if even ever so slightly, when the weight of the router is added.

Not a biggie, but be forwarned.

Forewarning is noted Don. I haven't made this part yet.

I just checked for more things to make this evening and it looks like I must have the 1/2" acme leadscrews, anti-backlash nuts, all of the bearings, pulleys and belt on hand before I can make any remaining brackets and wood parts that are affected by these items. Looks like I need to spend some serious moolah online this weekend.

CarveOne

My machine is finally starting to look like something other than a pile of parts. Nice weather finally arrived here this weekend so I built a new work bench outside yesterday. This bench will be used for running the cnc router outside the workshop when weather allows it and can be used for anything else that needs repairing or for building other home projects.

Last evening and this morning I assembled the end plates to the gantry beam. Wiped some golden oak stain on the end plates and sprayed a couple of coats of clear satin polyurethane on them. The intense sunlight darkened the photos too much and I'm not too hot with doing photo corrections to lighten them up, so here it is anyway.

Edit: Added three more photos.

CarveOne

That looks so good...

The Oak really does look amazing, I must say. So does the bench! I'm going to be building a table to put my CNC on too, but our weather is pretty craptacular, ie, Vancouver weather. The sun looks so nice...too bad I won't be seeing that for another 5 months...

It seems like you and I have tackled the project from different ends, and now we're meeting in the middle! I just finished my Gantry Endplates today too, and the bearing assemblies. On to the frame tomorrow!

By the way, what kind of bearings are you using for your table? I had a really difficult time today trying to bend those 120 degree pieces of flatbar...hope you have better luck/ideas :D

The weather here in eastern NC has been mid 90s to 102 F range and near drought conditions since June. Really hard to work on anything outside very long when you're used to constant air conditioning at work and at home. Sometimes the air is so hot and humid it's hard to breathe.

The 30" x 96" treated lumber bench cost $50usd for the materials. Screwed together with 3" deck screws. The 4x4 post legs were notched 3/4" deep and the perimeter boards were mitered at the corners. It took a little over four hours to build it, and I worked kinda slow at times. The notching makes a very stiff bench and is worth the effort if you have a radial arm saw or a sliding miter saw - but it can be done pretty well by careful hand saw cuts and chisel if you are up to that. Power tools - just gotta have them.

I've been working on making the wood parts first from the table top upwards, then making the metal parts and assembling the wood parts. Last evening I ordered the bearings from VXB.com, three 6' x 1/2-10 acme rods from Enco.com, and three 1/2-10 1 start antibacklash lead nuts and two 1/2-10 acme couplers from DumpsterCNC.com. I looked at sdp-si.com for the pulleys and belt but they show out of stock on the pulleys. Will look for them again tonight on alternate source sites.

I ordered 1/2" bore x 1-1/8" OD bearings (code R8ZZ) for the 1/2" acme leadscrews, and the 5/16" bore x 7/8" OD bearings (code Kit209) from vxb.com.If you use 5/16" allthread leadscrews per David's plans you just need four of the kit209 and one additional bearing. These aren't the high grade bearings but I'll use them until they fail and replace them with better bearings eventually.

I like working with the red oak for woodworking projects as it looks good and has much better characteristics than pine for many home projects. The rear frame rail even has some nice curly figure in it.

The flat aluminum bar stock wasn't particularly hard to bend for me. I put it in a 6" vise between two scrap pieces of steel bar stock, clamped it tight and whomped it hard with a 2 lb hand maul until it complied with my "coaxing".

CarveOne

Shipping confirmations received for all parts orders. Now I'm twiddling my thumbs while waiting for the parts. Still need to order belt and pulleys.

More photos by weeks end.

CarveOne

I like working with the red oak for woodworking projects as it looks good and has much better characteristics than pine for many home projects. The rear frame rail even has some nice curly figure in it.
CarveOne

I've been working all summer with RED OAK for the first time. I really like it a lot. So much so that I may redo my kitchen in red oak. And I like the fact that I can take my wood scraps and smoke some mean@$$ pork ribs!:D

At $1.65 a bd ft, its worth it! I may even use some for parts for my own CNC machine project.

Keep up the good work!

I have enough purpleheart to build a router table frame but at $9 a board foot I said "Nah!, let's use the red oak." It would be as stiff as steel tubing though. I did cut the bearing blocks out of the purpleheart though. It looks nice in combo with the golden oak stained red oak. I have some bloodwood and padauk (very red woods) that would look good also.

EDIT - Oh yeah, all hail to ribs on the grill...... Make some for me and tell me how they were please.

CarveOne

I received the 1/2-10 acme rods and bearings today. I hope to get the ABNs and acme stepper couplers Thursday or Friday. Looks like some job related work will stall the router build progress this weekend.

All three acme rods look very straight.

CarveOne

The DumpsterCNC AB nuts and acme couplers arrived today. They look very well made. Looks like I have some fun time coming up this weekend.

CarveOne

Made some unexpected progress this evening. Installed bearings on the trucks.

I ordered the 5/16" ID and 1/2" ID inch size bearings instead of the 22mm x 8mm size that was recommended. Hopefully, that won't make any difference in the adjustment range.

CarveOne

That looks so good.... I love how all your holes look like they're in the right positions lol. Did that on the mill?

Have any pics of the Dumpster AB Nuts? I was looking at the ones on their website, and yeah, I agree, they do look really well made. Expensive though! $15 a nut...but they look really good haha. You're using the 1/2"-10 single start, right? not double start?

pumelloman,

Yes, 1/2-10 1 start. The Use-Enco rods I ordered were $6.99usd each. MSC has the 2 start for $50.99usd each. I would rather have used 2 or more start though. Not a small difference in my limited build allowance.

Here's a not-too-well-focused photo of an anti-backlash nut (left) and the acme stepper coupler (right). The coupler is optional, and only eliminates having to turn the end of the acme rod to fit the hose type coupler shown on the plans.

Yeah, kinda pricey, as I bought four AB nuts and two couplers. If I end up going to 2 start (or more) rods on a future upgrade I'll have to buy them again. I don't think I would try to make them myself to save the money though I have the machinery needed to do it. They look simple enough but the effort wouldn't be worth it unless going into small production.

Yes, drilled them on the mill, mostly because drilling the corner of the angle stock would be easier on the mill. I have a cheap x-y vise on my Delta floor standing drill press that would have worked ok too. I noticed that your z axis block has two end grain holes that are well centered. Drilling mine on the mill wasn't so accurate on one of the holes. The hard red oak end grain caused the drill to wander off center a little by the time it was half way through the block. I should have used a brad point drill instead of a regular drill. (chair)If looking at it aggravates me enough I will re-drill it. Don't hold your breath while waiting to see if I do that. :D

CarveOne

More progress this evening as I made some time available to assemble the x axis bearing trucks to the gantry end plates. There has been no attempt at this point to adjust the bearings to fit to the x axis pipe rails yet.

I think CarveOne's Area 51 Skunk Works, LLC is in serious need of a thorough cleaning up.

CarveOne

EDIT - Added two close up views of x axis bearing assemblies.

The weekend started off well, but slowly deteriorated thereafter when I started work on the z axis construction.

As I was assembling the wood blocks I discovered that I had mis-measured the lower block and it was a little too short. I made another one with the remailing short piece of left over 1x6 board. Then I discovered that the new block was cut on the wrong side of the cut line. I made a 20 mile round trip to buy another piece of 1x6. This time I cut it correctly, and I proceeded to glue the upper and lower blocks to the vertical block and went outside to do much need yard work.

Later in the evening I discovered that the lower block is supposed to be removeable so the x axis can be installed on the gantry. So I managed to break off the lower block with not much damage at the joint and cleaned it up. Then I stained the wood assembly and left it to dry overnight.

This morning I sprayed a coat of satin polyurethane on the assembly, then noticed that the top and bottom plates have the counterbored 5/16" hole and the counterbored bearing hole swapped.
Oh well, back to the drawing board and start over..................(chair)

CarveOne

FYI,

While at at the local Lowes home center buying the remaining small bolts and nuts I found some chromed pipe plugs that fit the 3/4" EMT tubing. The "fingers" fit loosely until you bend them outward enough that the fingers need to be compressed a little to insert them into the tubing. The plug OD is a perfect match to the tubing. Gives the tubing a "finished" appearance and keeps unwelcome critters from building homes in the tubing. (eeeeekkkk!!)

Hillman Metal Hole Plugs # H881267 costs a little over $1.00 usd each in this area. They also had 1/2" plugs but I didn't get any yet.

CarveOne

Whoa! EMT endcaps? I think I'll have to pick some of those up next day....I was wondering how I would seal those. I've seen some other people do some designs to hold the 1/2" parallel near the top (especially when the cutter is lowered), but I'm not sure if I will need this kind of fixture ....yet.

The z axis is assembled and installed on the gantry. The 1/4" polycarbonate plastic side plates look good but if they crack under stress I'll make 1/4" aluminum replacements.

All of the steppers are mounted and DumpsterCNC acme couplers are installed on the y and z axis stepper shafts.

The belt idler is assembled and installed.

Neat trick -Two US nickels were used to seat one of the bearings a little over 1/8" deep into the PVC coupler. They are the perfect diameter that covers the outer bearing race but doesn't get stuck in the coupler. I couldn't find any washers around here that fit as well as the coins.

Tomorrow I'll work on modifying and mounting the DumpsterCNC anti-backlash nuts and make new bearing support blocks for the larger 1/2" ID / 1-1/8" OD bearings. I'll also take some outdoor pics without the junky work room clutter for a background.

CarveOne

Neat trick -Two US nickels were used to seat one of the bearings a little over 1/8" deep into the PVC coupler. They are the perfect diameter that covers the outer bearing race but doesn't get stuck in the coupler. I couldn't find any washers around here that fit as well as the coins.

CarveOne

We (my girlfriend and I) worked on my machine today a lot too, we made the stepper motor mounting plates, finished the bearing mounts, and started assembly ^^

Ah....Nickel....today I tried using a washer, but it was exactly the same diameter as the bearing... My girlfriend said "won't the washer get stuck in the PVC tubing?" but I was too excited about how close I was to assembly/finishing and I pressed the washer in anyways. Then it got stuck... and it was super hard to get out... sigh....

Your machine looks so good... I love the polycarbonate!!!!! My uncle brought me some clear Acrylic or something from the insides of LCD monitors to make my windows for my enclosure, when and if I make it. Are you worried much about dust for your machine?

If you don't have any of those old (recently devalued) US nickels lying around maybe the Canadian version is about right also. Last time I saw one many years ago when I was on TDY training duty at Selfridge AFB in Michigan they were slightly smaller and had small facets around the edge. Are they still like that?

Otherwise, just grind off a little around the washer's perimeter until it fits correctly. A US nickel is 0.836" (21.24mm) diameter and you can go to 0.855" (21.72mm) according to my cheap HF digital cow pliers (calipers).

I thought about mounting my idler assembly in the lathe and skim off just enough plastic to clean off the two small humps and true up the surface, but I'll wait and see it it needs it once the router table is running. It's more or less running nearly wobble free. I'll set the belt to run right in the shallow valley between the bearings.

After seating the bearings I warmed up my toaster oven a little and turned it off. Then I put the idler assembly in and let it warm up for a few minutes to relieve some of the stress.

I'm going to make two 3/4" thick by 3-1/4" wide by 6-1/2" long red oak router mounts so that I can use a $20 HF trim router or a $49 refurb Ryobi 1.5 hp router. The z slider will have 1/4-20 tee-nuts mounted from the backside. 1-1/2" bolts will attach the mounts to the z slider assembly. The routers will be held in place on the mounts with oak or 1/4" aluminum clamps.

Dust? The Ryobi blows air downward with lots of force. That means the chips and dust will go everywhere, probably even with a ShopVac connected to a dust shoe. If the shoe has a removeable solid bottom plate that has an opening just slightly larger than the cutter I may not need a ShopVac, just an exhaust hose.

ps - If your girlfriend voluntarily helps with the router table construction then she's a keeper. Don't let her get away. :)

CarveOne

Yeah, the nickels are still pretty much like that, they haven't really changed the coin design in ages... (but they're worth more now! I'm using this opportunity to buy all things eBay haha).

Hmm I have been thinking about doing the same thing for a router, I have a craftsman router that's pretty old that I was thinking about using to cut cabinet doors or something. It's only a 1/4" collect (I think) so I'm not sure if there's any advantage over the 1/4" trim router (Porter Cable) which I received yesterday. Although the craftsman is 6.5 amps....hmmmm...

Yeahhhhh she said she really found it fun to work on the CNC... I'm so happy..

I found out also that my table is TOO SMALL (!!!!!) it's not wide enough! oh my god...how am I going to build an enclosure if the router doesn't even fit on the table...

pumelloman,

You can use a half sheet of MDF on top of the table to make it wider and it will also act as waste material - unless you add a second layer of 1/4" hardboard as the waste material and preserve the 3/4" MDF layer as just support for the cnc router table.

I roughed out the trim router mount today and installed it. The $20 trim router's plastic housing is not a perfect cylinder but the removable mount seems to work ok. It was easier to make this mount than the wider mount for the Ryobi. This one will work for "air cutting" and 2D parts cutting purposes.

CarveOne

I picked up some 1/2 - 10 acme nuts from the local Fastenal store today and will start installing the x axis lead screws tonight. It looks like the only problem I will need to correct is that the 1-1/8" OD ball bearings will have an interference fit with the wood side rails in four places, but a little creative wood removal with a Dremel tool will make room enough that the 1/2" rod will remain in the original 5/16" rod hole centers.

The DumpsterCNC anti-backlash nuts will also work if the round flange is cut off on one side and a new bracket is made to tie the ABNs to the wood block on the gantry end plates.

I'll post some close up pics this weekend once I'm happy with the results.

CarveOne

Can't wait to see the DumpsterCNC AB nuts in action, hope to see some pictures!

I don't know how long the Delrin ABNs will last but I think that they will "seat" on the threads during operation kinda like piston rings seat in an engine's cylinder. A little light lubrication once in a while may slow down the wear.

Here's a quickie look at one. The challenge is to make a right angle bracket to allow two screws to mount the ABN flange to the wood block. It is shown on the rear edge of the block in this photo, but it may end up facing the other way and mount to the front edge of the block.

CarveOne

This is the bracket that I settled on for my x axis ABN mountings. The left side bracket is a mirror of this bracket. Putting it in this position and with the spring end forward does not limit the gantry travel in either direction. Putting it on the rear side of the gantry block would cause the ABN to run into the bearing block at the rear leg. I could remove the gantry block and cut it shorter so that the spring end of the ABN could be under the gantry block but that requires a larger slot in the bracket. So I'll reserve that for later if it becomes necessary. (I'm starting to feel thay way about a lot of these little details. I'm about ready to see this thing run, even before it's finished.)

CarveOne

Moving right along this weekend.

I made the purpleheart bearing blocks and have assembled the left side of the x axis lead screw. The right side is ready for final assembly.

The pulleys and belt are scheduled for UPS delivery from sdp-si.com on Monday.

The y and z axis lead screws have not been worked on yet, nor have I made the three stepper cables.

I removed the circuit board from a dead pc power supply with the intent of using the case for mounting the stepper control board. The case has a large (5-1/4" square) 12vdc fan that is in good condition, and an ac power switch. The power supply is 24vdc though. I'm thinking of making a case that can contain the 24vdc power supply and the controller board with two fans for cooling.

CarveOne

The pulleys and belt arrived as expected today. Those are the last components that I needed to complete the assembly of the hardware.

Also as expected, the two x axis pulleys do not have enough aluminum hub to bore a 1/2" hole through them so that they fit on the 1/2" acme rods. I can turn the ends of the rods down to fit the existing bore holes in the pulleys and thread the end long enough for two 5/16-18 nuts. I may be able to find another set of larger pulleys with the same ratio and proper bore size. Oh well, there had to be a serious hitch in my git-along at some point in my build, and this seems to be my usual luck with home projects.

Time to go assemble the right side x axis leadscrew hardware.......

CarveOne

Looks like I'm in luck with the 1/2" bore 15 tooth XL pulleys. sdp-si.com shipped a pair today even though their website said that they were out of stock last evening. Amazing. I should be finished with the x axis hardware by this weekend.

Now, if I could just win the state lottery..... :) I would start on my own bigger, faster, fancier cnc router machine build.


CarveOne

Now, if I could just win the state lottery..... :) I would start on my own bigger, faster, fancier cnc router machine build.


CarveOne

Looks like everyone's the same, hahaha

Pumelloman,

Wishing is free, buying lottery tickets costs money. I'll probably be spending any potential lottery ticket money on my next cnc build.

Made some unexpected progress this evening. The 1/2" bore pulleys arrived today and I completed the left and right x axis lead screw assemblies and installed the belt assembly. Only the right side anti-ballistic nut, errrr anti-backlash nut, needs to be installed on the aluminum bracket.

Because of this good fortune maybe I can complete the y axis lead screw assembly this weekend and start figuring out how to install the ABN and lead screw on the z axis.

The http://www.sdp-si.com part number for the 1/2" bore 15 tooth aluminum XL pulley is:

A 6A 3-15DF03716 (currently priced at usd $9.84 each)

Added photo of belt drive installation.
CarveOne

Spent the whole day working on the router machine today. The final assembly of both sides of the x axis is completed.

I made a bracket for the upper ends of the z axis tube rails from 3/4" x 1/8" aluminum angle, two lathe turned steel plugs, and two 12-24 screws. This will be used for mounting a carved purple heart plate when the machine is capable of making one. Not sure what will be on it yet. Sorry about the fuzzy photo.

The ABN will be easy to install on the y axis. A 1" right angle aluminum bracket for it will be finished tomorrow. The bearing blocks have been made but have not been bored and drilled yet. Will do that tomorrow as well, and I hope to complete the y axis assembly tomorrow.

The ABNs put a lot of drag on the lead screws due to spring pressure. If I compress the spring a bit the drag lessens considerably. Seems that the spring needs to be a little lighter design, but maybe the ABNs need some run-in time on them to fully seat in the rod threads. Will ask Tim at DumpsterCNC about this.

CarveOne

See photo below for the y axis ABN mounting bracket details.

While working on the y axis lead 1/2-10 acme screw and ABN installation yesterday I found that there is too much flexing and lost motion in the small mounting block that the ABN is mounted to. The block moves laterally when rolling the lead screw a little in both directions. Also, when moving the z axis assembly by hand there is too much movement at this block also. The long 1/4-20 all-thread rods that run through the oak ABN mounting block also have some flex.

In order to eliminate the freeplay problem I'm going to remove the small oak ABN mounting block and long 1/4-20 rods and replace the small ABN mounting block with a longer piece of 1x6 oak that is about 5-1/4" long. Instead of drilling through-holes for the 1/4-20 rods, I'll drill two holes at each end of the block about 1-1/2" deep and spaced the same as they are on the plans. In these four holes I'll install 1/4-20 brass threaded inserts with some epoxy on the outer threads. The long rods will become four shorter pieces that install through the top and bottom plates and screw into the threaded inserts in the new ABN mounting block. The ABN mounting bracket will have obround holes for mounting to the block and for minor adjustment of the ABN position.

This will make the oak ABN mounting bracket fixed in a position where it clears the ball bearing bolt heads and will be much stiffer due to the shorter 1/4-20 rods. The nuts on the 1/4-20 rods can still be used to adjust tension on the y axis roller trucks. The z axis assembly can still be removed when necessary by removing the four short rods from the brass inserts. (Maybe just the upper two rods will allow removal.) Another photo will be posted when this is completed.

The ABN mounting bracket is a 2-1/2" piece of 1/8" x 1" aluminum angle with a 2-1/2" piece of 1/8" aluminum flat stock as a spacer. About 1/4" of the ABN flange was ground off for clearance to the wood block.

CarveOne

Edited: I added two photos of the new ABN mounting block which shows the larger block from both sides. The new block is 5-1/2" x 5-1/2" x 3/4". It should be made 5" tall, as I had to relieve the backside of the block where there is an interference where the bearing bolts are located. I drilled into the block 3/4" deep in four places to match the holes in the top and bottom plates. I then drilled for 1/4-20 x 1/2" long brass threaded inserts and screwed them in place using a T wrench and washer. The washer on the end of the T wrench made sure that the inserts were pulled flush with the end of the block and forced the insert into alignment. (The T wrench is sold at woodworking stores for installing these inserts.)

The four shorter 1/4-20 allthread rods are cut 3" long and I cut a slot in one end of each rod with a Dremel tool and heavy duty cutoff wheel. The slot allows a screwdriver to be used to screw the rods into the block until it seats against the bottom of the holes. Nuts and washers are used on the rods to tension the z trucks as before this modification. The aluminum angle has slotted holes now, and the aluminum spacer under it is made from 1/16" sheet.

This is now a very rigid assembly. There is no noticeable free play or flexing in this connection now. I highly recommend doing this change from the plans even with the original 5/16" allthread lead screws.

CarveOne

Last evening I reassembled the y axis ABN mount and edited the last post to include the new photos and description. I continued messing with the adjustments and looking for any new problems with the changes.

I'm really happy with the y axis movements and it seems like a rock solid connection from the leadscrew to the z axis assembly now. All four short 1/4-20 rods need to be removed to take the mounting plate off. There is no slop in the 1/4" holes in the top and bottom plates that the short rods mount through. If there were, additional nuts can be added to the rods so that there are locking nuts on the rods that can be tightened after the tensioning has been completed. I will add them on my machine just so nothing loosens up with use.

CarveOne

Last evening I reassembled the y axis ABN mount and edited the last post to include the new photos and description. I continued messing with the adjustments and looking for any new problems with the changes.

I'm really happy with the y axis movements and it seems like a rock solid connection from the leadscrew to the z axis assembly now. All four short 1/4-20 rods need to be removed to take the mounting plate off. There is no slop in the 1/4" holes in the top and bottom plates that the short rods mount through. If there were, additional nuts can be added to the rods so that there are locking nuts on the rods that can be tightened after the tensioning has been completed. I will add them on my machine just so nothing loosens up with use.

CarveOne

Just a thought, wondering why you didn't use a solid piece of wood and not use the long 1/4 20 threaded rod on the backside? Or was ot for adjustment?
Good looking job.
Hager

Mr.Chips,

The two long 1/4-20 rods are for tensioning the y axis trucks against the two pipe rails on the gantry. The Solsylva design calls for 1/4" holes through the small mounting block for the rods to pass through from the top block to the bottom block of the z axis. The bottom block is screwed, but not glued, to the center vertical block of the z axis assembly. The top block is glued and screwed. The bottom block is able to pivot a little when the two rods are tightened. The holes in the small block that the rods pass through have some free play and could be epoxied to the rods. The long rods would then flex under heavy side loads on the cutter. The small block "floats" on the rods and self centers to some degree with the position of the y axis lead screw threaded coupler (of the original design).


My first thought was to replace the long rods with 3/8" rods, but instead of drilling larger holes in the top and bottom plates plus the small block,
I chose to shorten the 1/4-20 rods and use a longer block to minimize all of the free play and flexing. I put obround slots in the ABN bracket to allow adjustment of the ABN mounting bracket to align with the y axis 1/2" acme lead screw. The mounting block could have been narrower, but adding the brass threaded couplers to the original smaller block required a slightly wider block so I just used a full width piece of 1x6 thinking that it also just looks better than the small block.

CarveOne

I finished making the y axis bearing blocks this afternoon and installed them. That completes the y axis work.

Tomorrow I'll take the z axis off the gantry (might not be necessary) and install the leadscrew and anti-backlash nut. It looks like there will be enough space for the ABN if I cut it off on each side of the ABN flange. There is 1" available space and I can cut the flange to 0.850" with no problems. The 1" right angle bracket for mounting the ABN will need to be cut to about 7/8" so it doesn't rub the main vertical plate as the ABN mounting bracket goes up and down with the router mounting plate. Might take a couple of hours to complete the z axis work.

Remaining things to do are:

1. Make three stepper motor cables,
2. Mount the stepper motor control board and power supply in some sort of fan cooled case.
3. Make a breakout board for the limit switches and zero setter plate.
4. Buy and mount some limit switches and wire them to the controller board.

5. Make a zero setter plate and wire it to the breakout board.

6. Make any necessary adjustments to the lead screw and stepper motor alignments.
7. Learn to use Mach3.

TIP #2 -

I bored 5/8" clearance holes for the 1/2" acme rods in my frame, gantry end plates, and bearing blocks. To keep the acme rods centered in the holes while I drilled screw holes for the bearing block mounting I cut 3/16" wide x 1" long strips of 1/8" balsa and inserted four of them equally spaced in the 1/8" gap around each end of the acme rods. This will be a snug fit, but will keep the rod centered in the hole and even allow the rod to be moved through the holes enough to install the bearings and bearing blocks. After installing the bearing blocks remove the balsa pieces. Now slide the rod back and forth to see if there is any drag on the bearings due to the holes not being aligned or something is not square. If it slides easily, install the stepper motor and coupler. If any minor alignment is needed, it can now be done using the bearing block screws.


When the z axis leadscrew is completed tomorrow that will make the mechanical assembly 100% completed. :banana:

CarveOne

After taking a break late today I turned my attention back to the idler pulley issue.

I mounted the idler in the lathe and put a live center into the inner bearing race to align the pulley and snugged the chuck jaws onto the plastic coupler just enough to cut it down to the bottom of the little valley between the pressed in bearing locations. When this was done I flipped the coupler end for end and cut down the other half to match. This made the surface nice and smooth, and true. It should work nicely now, and it still tensions the belt within the range of the mounting bolt slots. It shows no signs of splitting and still has plenty of plastic thickness at the bearings.

When I get some spare bearings I will still replace this idler with a couple of bearings and make some flanges for each side.

CarveOne

After taking a break late today I turned my attention back to the idler pulley issue.

I mounted the idler in the lathe and put a live center into the inner bearing race to align the pulley and snugged the chuck jaws onto the plastic coupler just enough to cut it down to the bottom of the little valley between the pressed in bearing locations. When this was done I flipped the coupler end for end and cut down the other half to match. This made the surface nice and smooth, and true. It should work nicely now, and it still tensions the belt within the range of the mounting bolt slots. It shows no signs of splitting and still has plenty of plastic thickness at the bearings.

When I get some spare bearings I will still replace this idler with a couple of bearings and make some flanges for each side.

CarveOne

What are the dim. of the bearings you need?
Hager

I'm using vxb.com's kit209 7/8" OD x 5/16" ID skate bearings in the existing coupler style idler and all of the bearing trucks.

I used vxb.com's R8ZZ 1-1/8" OD x 1/2" ID bearings for the 1/2" x 10 acme lead screws on my x/y/z axis. I will also try them for the idler when I get some more of them for spares. On the same order I'll get some more of the kit209 7/8" OD x 5/16" ID skate bearings.

The coupler idler seems to be working well enough that I may just keep using it. It won't be much longer before I'll be able to run the steppers.

I finished the z axis lead screw installation this morning, so now I'll start wiring up the steppers.

CarveOne

Carve One

I see a potential problem on your Y axis.. It also noticed the same possible problem with the x's, but took a wait and see apporoach.

#1 My Y axis mounting plate (the wood piece) is tight against the bottom plate, with nuts on the top of it holding it there as per the plans. It is firmly against bottom plate, not floating, and has NO movement problems in over a year. It is very solid. The slight angle change as the bottom bplate moves is why the holes are slightly oversized. The difference is probably because of the way you mounted the bracket for the ABN, which leads to problem #2

#2 On both your X and Y axis, I only see one axis for adjustment, not the 2 required. That is why my upgrade brackets were made up of 2 parts, to mimmick the original design, giving horizontal AND vertical movement. Check the pictures on my upgrade post to see what I mean.

You will need both directions of adjustment to get the ABN to align perfectly. The perfect alignment of the nuts is critical to the rods not binding.

Email me if you have any questions.

Carve One,

Don't worry about the flanges, once adjusted properly, the belt rides very nicely in the middle of the tensioner. No need for flanges. If it doesn't ride centered, there is an alignment problem. I turned the inside of the plastic piece for the bearings, instead of trying to force fit them. Kept it nice and straight. Ran a skim pass over the outside with the lathe just to be sure.

rc_flyer,

Thanks for the heads up. Warning flags are welcomed.

I thought of turning the inside of the coupler also, and bought a second coupler just in case I decided to do that. Then the double or triple bearings idea was mentioned and I wanted to use just two bearings with flange and went off on a tangent thinking about how it could be done. The initial belt setup can be done by eye and measurements but if the idler bolt is not square the belt may still try to run off to one side. Wherever it seems to stay will be the place to set the bearings.

You are correct about the y lead nut support block, as I didn't look at page 1-33 in the plans book when I assembled it. I forgot the nuts that hold it down against the bottom plate and the washers under the plate. No wonder there was a lot of free play. Sheesh!

On my newer version of the y lead nut support block, there is vertical positioning of the ABN by tightening the top nuts and loosening the bottom nuts to compensate. This would move the whole block vertically along with the ABN bracket. There is additional vertical adjustment with obround slots at the ABN bracket mounting screws. Front to back horizontal adjustment is done by changing the spacer plate thickness between the ABN angle bracket and the y lead nut support block. The ABN was installed after the y bearing trucks were tensioned and the y lead screw was installed and centered. The ABN mounting plate mounting holes were drilled after the ABN was screwed to the ABN angle plate and installed on the y axis lead screw. The mounting holes were then made obround for extra movement.

Everything on the x and y axis has been checked for alignment by measurement and machinist square. The ABNs and their brackets were also installed after the lead screws were installed and adjusted for alignment. The vertical angle brackets have obround mounting holes for vertical adjustments. They were also checked for alignment with the lead screws with a machinist square and measurements. The wood blocks on the gantry end plates were square with the lead screws so I didn't think any other adjustment would be needed.

Once the steppers are running and I see that something needs tweaking I will make any necessary corrections. I expected that some things will need to be changed or readjusted. I've made considerable effort to build this thing as squarely as possible so that things fit with little need for major adjustments. However, disasters happen in spite of the best engineering efforts.........

CarveOne

rc_flyer,

Thanks for the heads up. Warning flags are welcomed.

CarveOne

One other note. The plastic bearing assembly. Coogrrr did the same thing and when the temp dropped, it trashed most of them. So hopefully your router is in a temperature stable environment.

See his video:

http://www.youtube.com/watch?v=iw1Aq3MTm8k

OzarkCNC,

Thanks for the link. I have it inside a temperature controlled room in my house and it should be ok until I replace it with a pair of stacked ball bearings.

I'm not surprised by the plastic coupler shattering since it gets brittle in cold temps and and the bearings were forced into the coupler which causes it to stretch unnaturally. Machining out the coupler is easy for me to do with a lathe, but builders without much more than hand tools to work with is what these plans were written for. I just take that in stride and make allowances for it to end up with a better machine.

CarveOne

Didn't make much progress on the router build this weekend. I added enough wire to the stepper motor leads to connect them to the control board and twisted the wires. Can't do much else until I buy or make an enclosure for the control board and power supply. The Xylotex system kit power supply came with the 110v/220v switch set for 220v so I set it for 110v and connected the power cord and dc lead to it and set it aside for now.

I'll be busy with getting a 30' x 40' workshop foundation and concrete pad constructed this week so I'll have no time to work on the router during the week. Need more space for building my home projects than the small room I use for it now.

CarveOne

Last evening I reformated my Fedora 6 Linux computer's hard drive and installed Ubuntu Linux 6.06 LTS, then installed all 170 available updates, and then downloaded and installed the EMC2 precompiled package from http://www.linuxcnc.org.

It took a few hours but it all went smoothly and seems to work ok as I am posting this with Firefox from the newly installed Unbuntu distro. This computer is a 1.7GHz P4 (266MHz FSB) 256Mb SDRAM, 60 Gb HDD, 1.4Mb FDD, 18x8x12 DVD, USB2 mid-tower with 19" monitor. I'll add more SDRAM soon as I can find some.

It's using an ethernet connection initially, but I'll see if it recognizes my Linksys WMP54GS Wireless-G PCI card next. Never did get it to work with Fedora 6. No driver was available at the time. Would be nice to unhook the LAN "chain" off of it so it is more easily moved to the room where the cnc router is being built and tested.

CarveOne

Using Ubuntu 6.06? I'm using 7.10 Gutsy at home, I like it a lot. I was using 7.04 Feisty, but Gutsy has way better support for wireless devices...I had a Belkin USB wireless card and that was horrible to set up in Feisty, had to recompile the Kernel and everything...
My roommate installed Xubuntu on a kindof POS pentium 2 at our house, and it works! I was pretty impressed, the computer barely works in windows (although it is damn slow in linux too). I don't think Xubuntu has very good support though for any drivers, period. The computer I'm going to be using for my controller is an AMD Athlon 1400+ or something like that, so I'm not too worried about it being too slow with full Ubuntu on it, instead of Xubuntu.

EMC2? I think in time I'll give that a shot, but for now I'm going to stick with Mach3, and probably switch over to that in time (just like how I went to Ubuntu from windows in the end...)

Carveone - nice build and thread. Thanks for posting all of this, esp. the learning curves part.

As far as the Lynksys cards - goog luck. I refuse to use them anymore due to historical challenges. It will be interesting to see how it goes.

Thanks harryn, I hope to have a decently useful cnc router when it's finished and try to entertain the readers a little along the way.

Ubuntu 6.06 LTS recognizes the wireless card and the Ubuntu troubleshooting site shows some things to try that have not worked so far. The Linksys setup program in my Win XP main computer doesn't seem to allow turning off the encryption once I enabled encryption during the initial installation of the wireless router. I'll probably have to log into the wireless router as "admin" to disable encryption. I gave up on it for the time being and will get back to it later. I have bigger turkeys to fry this week.

pumelloman, as for Ubuntu, I like it so far and installed this version to avoid issues with EMC2 compatibility. I will download the newer "Gutsy" version and see how it works with EMC2 and the Linksys card. I'm no guru with Linux but have been playing around with it since before Red Hat desktop v5 or so. Suse, Mandrake, Lycoris, and a few others have been on my computers also. I should know more about the inner workings but just haven't delved into it that far.

CarveOne

Oh yeah, I bumped the ram in the Linux computer last Friday evening to 784Mb so that EMC2 has a bit more ram available to use. A friend scrounged up a 512 Mb module from his spare parts bin for me.

CarveOne

Not much work done on the cnc router in the past week but I've been working on the software side of things. I drew a sample 2D part drawing in TurboCAD 12 and converted it to a gcode file with 30 day trial converter Gcode2000. The gcode file will import into Mach3 and EMC2 ok now, but I don't know what to do with it yet.

I was having a hard time getting the TurboCAD .dxf output file to load into Mach3 and EMC2 so I went looking for a dedicated converter program for WinXP.

Seems that the software is harder to "put together" than the hardware was. Too many parameters and meaningless numbers. :confused: Maybe it will all start to mean something if I keep at it.

CarveOne

Yeah, I'm also confused at how to get going from Mach3, but I've mostly been messing around with toolpaths etc.

Right now I'm waiting to make my 24VDC power source, before I can actually test my circuitry. I'm kind of scared of electronics (even though I'm in mech-electrical engineering...), have had a few bad near-death experiences.... very scared of electricity now. Made my box though, it looks pretty awesome :D will post a pic tonight.

pumelloman,

Looks like I have some unscheduled time off while the boss is beating the bushes for some new work projects for us to work on. Contract work is like that. I might have a few days or a few weeks off. It should help me finish the cnc router and most of the new workshop build though.

I have been looking for breakout boards and i/o cards today. CandCNC.com has what I need for my Xylotex steppers and controller board that mounts in a plug together case that I may order this week.

I have a local cnc knowledgeable acquaintance who is helping me sort out the software stuff. He has a homebuilt cnc foam cutter and laser cutter.

Don't play around with a homebuilt power supply if you don't have the proper electrical knowledge. Buy one that has a regulated output. The line input side of the transformer can kill you if you touch it while it's plugged in. Filter capacitors on the line input side can remain charged to near line voltage if there is no bleeder resistor in the circuit. Buy one that has a regulated output. CandCNC.com has one for $49usd. Many others are out there.

CarveOne

pumelloman,
I have been looking for breakout boards and i/o cards today. CandCNC.com has what I need for my Xylotex steppers and controller board that mounts in a plug together case that I may order this week.
CarveOne

I started out with Xylotex driver baords and through stupidity fried them. You will fry them if they become disconnected from the stepper or shorted (mixed wires at stepper) That is their weakness.


My real electronic breakthrough came when I found Probotix, They have a very user friendly web site with a wealth of illustrations, showing color drawings of wiring connections, not the typical encrypted text that I have so frequently seen, and was confused by. It’s just the ticket for the electronically challenged. I studied the drivers and sent emails and found out that my six wire Xylotex steppers could be connected up either as Bipolar or Uni-Polar, Probotix has a driver called Probostep Stepper Motor Driver that supported the Uni-Polar stepper, this driver has some terrific attributes in that it has Short/Open Circuit Protection and Thermal Overload Shutdown protection. (an Open was the demise of my last drivers) It supports Full, 1/4, 1/2, 8th & 16th Microstepping Resolution. Check out the details it is well documented.
Their driver can be connected to their PBX-RF Isolated Breakout Board by simply connecting their IDC cable between the two, this makes connecting everything up even easier, and there is less chance of miss-wiring and frying expensive electronic parts.

Driver: http://www.probotix.com/index.php?vi...&product_id=12
Breakout Board: http://www.probotix.com/breakout_boa...reakout_board/
Main Web Site: http://www.probotix.com/

I have been using them a couple of months now and they are working fine.

Thanks Mr.Chips,

I'll check those links before buying anything this week. I have the Xylotex 3 axis board and 425 oz/in steppers on hand now. I have 40 years experience in electronics behind me so the wiring and soldering part is easy for me. That's assuming that I get the correct information to work from. The instruction sheet that came with the system kit has the wrong colors relative to the motors I received, so I questioned it in an email to them one evening. A reply was received by noon the following day. Without the factory datasheet for these particular motors I'm still not 100% comfortable that these new instructions will not fry my motors and driver board. I've read of too many such incidences on this site already. I've been trained over the years to question everything before applying power.

CarveOne

My little 116 oz/in Xylotex motors were wired Bipolar but with Probotix help I rewired them to be Uni-Polar which is what is required for the Probostep Stepper Motor Drivers.

I like the feature of having seperate driver boards and the safety of Short/Open Circuit Protection and Thermal Overload Shutdown protection for newbees like myself.

Some of the features of Geckos but at a fraction of the cost.

Good Luck

Nice link Chips!

very cool - glad to see you trying emc2. It keeps getting better and better. For simple stuff I use ace converter.

sam

CarveOne - I've had good luck using CamBam. Draw up your 2D part, save it as a DXF, then import into CamBam. Once there you can choose the order of operations, have it do 2.5D milling, pockets etc., then have it create the gcode. It's a few more steps than ace, but it gives you a little more flexibility if you need it.

Mr.Chips,
I spent some time on the Probotix site this morning. I'll need to check with them (Len) to see if either of their drivers can handle my 425 oz/in motors. They are 5a/phase bipolar parallel. I'm not sure if they will work with either of the drivers they sell.

samco,

I plan to work with both control programs, at least initially, so that I become somewhat dangerously inefficient with both of them. I'm semi-literate with Linux as long as I don't have to tread very deeply into the pits of a terminal window too often. :) I actually have become kinda fond of Ubuntu. I downloaded a trial copy of gcode2000 and ttftoDXF. Maybe the EMC2 team will add a gcode generator screen in the near future. Let's all gang up on them and push the issue. LOL

ron39,

I will give them both a try, soon. Thank you all for the help.

CarveOne

Made some progress with the drawing file I created in TurboCAD v12 yesterday. Simple enough 2D file, but there is a lot I have to learn about this stuff. I thought I had closed the polyline outline but it wasn't. and it was causing complaints from the gcode converters I am trying to use.

Thanks to my local acquaintance, the gcode is working now and I have been able to run it in EMC2 and Mach3. So, when the hardware is all connected up I have a real part cutting file that I can use to "further improve my ignorance of the cnc routing world".

I downloaded and installed CamBam last evening and will try it with the newly corrected drawing file later today.

I will now go back to working on ordering and installing the limit switches, relay board, opto-isolated interface board, and making a box for the electronics that will functionally complete the machine.

If this wasn't so much fun I would have bought a plug and pray commercial machine. (chair) The ShopBot factory is just 100 miles due west of here.

CarveOne

Hey Carve....

You are just getting into the fun part........

I'll second ron39's comments on Cambam. Once you get the hang of the interface, it works great. Very accurate, and pretty forgiving on dxf import. It imports acad 2004 dxf as splines and lets you convert them to polylines in the program. Most of the rest I have tried are afraid of splines, or handle them terribly. You can also import polylines you forgot to join, and it joins them very nicely. I have switched to it as my only CAM program. Oh yeah, the drilling holes to shapes and the height maps are pretty cool too:)

Be sure to get on the forum. It is a quick place to get answers till Andy gets the docs re-done. And the stuff Arie does with it will blow you away. :cool: I loved the mold for the 1/12 scale tie wraps

Don
www.eurekaaircraft.com

rc_flyer,

I'll join the CamBam forum for sure then. I'm sure this will all get easier once I get more familiar with the software and have the hardware running correctly. Kinda overwhelming at the moment though.

I hope everything is going well with you, and your machine is running happily after the recent upgrades.

CarveOne

My cnc router build will be in a holding pattern during the Christmas and New Year holiday period. I thought I would get a work break for most of December but it hasn't happened so far.

I moved my Linux/EMC computer near the CNC router and will try to run the steppers as soon as I get a suitable parallel cable (or make one up by hand). I also need a longer LAN cable for the new computer location.

CarveOne

I made up a parallel ribbon cable yesterday from parts I have accumulated over the years. 26 pin female (socket) IDC header, 7' 25 conductor ribbon cable, and a 25 pin male (plug) d-sub connector. The ribbon cable wires were split out 1-1/2" and individually soldered to the cup pins of the d-sub connector. Everything behind the d-sub connector has been strain relieved with tape and heat shrink for about 2-1/2".

Next thing to do is to follow the Xylotex instructions to get the software ready to talk to the controller.

CarveOne

Good work CarveOne, glad to see that one of us is making progress ^^

I went back to Saskatchewan (where I'm originally from) for X-mas, but then I caught pneumonia, and a bad case of it too. So I'm writing this from the hospital now (been in here since the 24th), hopefully I'll get discharged today?

Don't know when I'll be back in Vancouver to work on the CNC but we'll see.

So other than your Xylotex, are you pretty much good to go?

Hi pumelloman!

I just happened to be poking around in the joe2006 forum when you posted.

Sorry to hear about your pneumonia problems. Hope you get over it soon. So far I have avoided the latest rash of flu and other stuff that is going on around this area.

I've been in the doldrums with respect to the cnc machine for all of December. I still have to purchase and install limit switches and make a box for the controller and power supply stuff. It's ready to run mechanically. I also need to set up the Xylotex driver stuff in EMC2 before running the steppers. Can't seem to get enthused enough to "just go in there and get it over with" so the fun can continue.

Take care, be kind to the nurses, and happy New Year....

CarveOne

Well, I got enthused enough today to work on the enclosure for the controller, power supply, and fan. It is 2/3 completed now, and I hope to finish it tomorrow. Photos will be posted tomorrow evening finished or not. With that done, there is no real reason to delay seeing if it all runs fine, or smokes. (Yes, as promised, there is plenty of clear polycarbonate on the top and three sides to see where any smoke originates from.)

Limit switches not installed yet - (Ha!! who needs 'em. Just crutches for the timid!) :):):)

CarveOne

Ok, photo time again. The controller enclosure parts are completed, board and power supply are mounted, but I need to find some Weld-On #16 gap filling adhesive for polycarbonate since my cuts were not done on a CNC router machine. (I wonder why?) If I don't find some locally I will attempt to drill and tap the 1/4" polycarbonate sheet for #4-40 FH screws and use some silicone sealer/adhesive made for Lexan that I already have.

I need to fabricate a cover for the AC input screw terminals on the power supply after I cut the opening wider for cable access, then decided to run the wiring through the air intake opening instead. The fan will be pulling air out of the box through three main openings.

The 1/8" thick aluminum on the rear of the power supply is an additional thermal mass for the power supply. It needs thermal compound added and maybe some fins bolted to it if it runs too warm.

CarveOne

Fantastic looking machine. Great job on this. Should be very strong.

I buy the glue you mention from McMaster Carr.
$3.62 a tube. It is some strong stuff as well.
http://www.mcmaster.com/catalog/114/gfx/small/7515ap1s.gif

Here's that part number.
7515A11

Keep up the great work. I'll keep checking back on your progress.

Thanks Lee.

I found a number of places that have it online . Once it's glued together it doesn't have any stress on the joints and it will be just a removeable cover.

Five minute epoxy seems to be good enough as long as the mating surfaces are scratch sanded. I'm giving it a try on some scrap pieces as I write this. The epoxy is hiding the scratches in the edge joints to some degree. I used a very small amount on both of the surfaces. Cleaning off the excess with acetone after it's partially cured isn't too bad so far. Seems to be plenty strong after curing, so I will go with the epoxy joints tomorrow.

CarveOne

I went through most of the Xylotex setup procedure for their controller board today. The reference voltages were set per their instructions for the 425 oz/in steppers I have and then I proceeded to connect the motors one at a time to see if they "lock up". Everything went according to plan with no visible smoke and fire. I'm really happy with that.
:cheers:

Before jog testing the motors I want to check out the logic level stuff for any I/O pins that need to be set to a particular logic level.

CarveOne

Sweet, sounds like your build is well on it's way again and you're out of that rut.
Seems like you really know what you're doing with the controls, what kind of things do you need to do to the Xylotek to hook it up? With the HobbyCNC I think the procedure is totally different.

Btw your enclosure looks fantastic. So professional looking!!! I was very impressed.

Thanks for the complements pumelloman.

On the regulated power supply that comes with the Xylotex system kit there is a small potentiometer that needs to be adjusted to the correct voltage range of 24vdc - 27vdc.

On the Xylotex control board are three small potentiometers that must be set to the correct Vref voltage for each stepper motor. For the 435 oz/in steppers it is 3.35vdc at a test pad on the controller board for each axis. These came preset to 0v, presumably to help minimize damage to the board from a bad wiring job. From what I've read on the Zone, the driver chips don't have open or short circuit protection built into the chips or board. Your HobbyCNC board has this protection.

All that needs to be done now is to connect the parallel cable to the computer and controller board and try to jog the steppers manually. I have temporarily disconnected the stepper shafts from the lead screws in case the software tries to run the steppers past the physical limits and there is no limit switch to prevent a runaway event. (I believe in the "belt and suspenders" theory.) There's a good chance that I may do that later this afternoon when I finish another project for my sister that I promised to complete this weekend.

Did you make any determinations yet about your stepper problems?

CarveOne

Well, sort of.... :)

While setting up a new configuration in EMC2 Stepconf Wizard and with the controller connected to the computer and power applied, the EMC2 Stepconfig Wizard jog tests run all three steppers just fine.

But, in EMC2/Axis I get some error messages about Joint 1, joint 2, and joint 3 and I think it's related to not having the limit switch connections installed. Anyone surprised? No? Thought so. I have lots to learn on the software side now. (Pray for me.)

CarveOne

It works now! All three axes are moving under computer command with no error messages.

CarveOne

Congrats CarveOne! What is the first thing you carve?
L

Hi Larry,

It will be a sheet of 1/4" aircraft birch plywood and my first cut file is for a pair of tank mount brackets for something that I currently make by hand with a bandsaw, disk sander, and Forstner bits. The cut file has a g-code error that I need to resolve (unrecognized g-command) and some cut area limit settings need to corrected before I can try cutting with it.

I have been running the machine today with the jog speed maxed out at 60 ipm and making preliminary adjustments of the lead screws to remove binding. Seems to be just minor adjustment needed since I paid close attention to hole location and squareness of the wood parts when I assembled the machine. I'm not sure what a reliable working speed will be for this machine but 40+ ipm is probably about right. That's pretty slow. I can see that new 5 start acme screws and anti-backlash nuts will definitely be in my next upgrade.

I located a 36"x49" piece of flat MDF at work today that is just the right size to slide under the legs to support the machine on my too narrow work table better. The boss was happy to donate it to my cnc project. Once I have him trained on how to use this thing doncha know that he'll probably run off and buy a large commercial machine of his own to play with. (chair)

CarveOne

Very nice CarveOne!
I agree about the speed thing, but for me I will probably remake a lot of my machine since I definitely didn't build it as nice as yours ;) I would say though that 40+ipm is a good estimate for the feed rate, that's around what I will be running mine at too.
I haven't figured out what's wrong with my stepper as of yet, pretty busy this week but I will go in on Thursday and see what happens. So far I'm convinced it's the gcode.

Be sure to upload some pics of what you cut!!!

pumelloman,

Pictures, yes, definitely. A video also. But it may be a while yet. Dunno how long though. The 231 page EMC2 manual is quite a bit of technical reading. I can read the words, but I don't necessarily "hear the music". :confused:

CarveOne

pumelloman,

Pictures, yes, definitely. A video also. But it may be a while yet. Dunno how long though. The 231 page EMC2 manual is quite a bit of technical reading. I can read the words, but I don't necessarily "hear the music". :confused:

CarveOne

Mach3????? you definitely have all the support you need on this forum and theirs.

Hi Joe,

Yes, I know. There is support for EMC2 also, but I won't learn it if I get too much hand-holding. I have Mach3 demo version on another computer and will buy it soon. I'm using EMC2 on the router computer just because I don't want to get WinXP for the router computer, and I don't want to risk my main computer near the router's dusty environment.

Truth is, I'm stalling a bit due to not having limit and home switches installed yet. I have the parallel I/O pins pulled low temporarily to eliminate error messages. I can try to run one of the example files and see what happens.

CarveOne

I've cleaned up the junk from around the machine table. Time for another photo. The room it's set up in hasn't lost any nearby junk though.

CarveOne

Hey CarveOne,

Man, SO NICE. Definitely win the appearance award for your machine ^^

Good news on my end too, I tried drawing a circle using G-code generated by Mastercam, and it drew a oval!!!! (although it was supposed to draw a circle, that's ok, it just means my steps per unit on one axis is incorrect). No more pen crashing into the Mdf and it was exactly how thkoutsidthebox said to zero the Z-axis.

Now it's time to show off what we can make using our CNC's ;)

Well, I knew it was going too well for too long.

Today I snagged a stepper wire and caused my controller enclosure to take an 8" tumble onto the floor from the top of a small box I had it sitting on. :eek:!!! The Z axis green stepper wire pulled partially out of the control board connector and now the Z axis stepper doesn't run. It doesn't "lock up" any more. Y and Z still work ok. Looks like I'm done for a while pumelloman. Be careful with yours.

I don't know much at this point other than the phase windings still measure 2.4 Ohms as before they did before, and the Vref voltage for the Y axis is still the value I set it for. I saw no smoke, heard no pop, and smelled no fumes.

CarveOne

I am not sure what drivers you have but in many cases disconnecting a motor while powered up will kill a driver.

Garry

Garry,

Yes, I know, and I assure you it was not done intentionally. I could have been a little more careful though. The question now is whether the stepper is ok or will the stepper kill the other driver circuits if I try swapping it to either the X or Y driver connector. I'll discuss it with Xylotex tomorrow.

They really need to add some open wire protection to their control board design. This is just too easy to do, and I also don't like the green connectors that they provide for attachment of the wires to the board. I think I'll mount the enclosure to a larger piece of plywood tomorrow so that I can use cable clamps to strain relieve the stepper wiring.

CarveOne

CarveOne

Ok

It wasn't done intentionally! But the driver doesn't know that....


I thought you were asking what probably went wrong. It usually doesn't hurt the stepper motor only the driver.

Hope you didn't hurt much...
Garry

PS
Thats why I like this board. Still cheap at $73 per axies and offers some protection for these accidents.

http://www.pmdx.com/PMDX-150/

Thanks Garry,

Looking at the feature list that looks like the driver I should be using. Three of those would cost more than a new Xylotex board, but then maybe one driver wouldn't have died during my little mishap either. In this case I can't blame anyone but myself for the problem.

Maybe it will not be so expensive to have my board repaired this time since it's just one driver chip to be replaced. I'll strain relieve the wiring this time.

CarveOne

great machine:) I just want to see the performance of this machine..pls show some parts:)

I know how you must feel. That sinking feeling down in your stomach. It can be tremendously disappointing. I did something ignorant with my first drivers and it was a Xylotex, but brand probably didn't matter in my case. While under power, I brushed the twisted end of a motor wire that wasn't taped very well up against the case. Killed one axis instantly. Like you, I saw no smoke and heard no noise. I could not see any physical damage, but it was there.
Luckily, Jeff fixed this one for me at no cost.
I ran it for a couple years without any trouble after that.

Things will get better for you and you still have some very nice gut feelings to look forward to along with a little pride when all is said and done.
That is one cnc speed bump that is never easy. :)

Oh my god CarveOne, I feel so bad for you. I will definitely take your advice and be a lot more careful with my electronics too.

Fortunately you don't have to solder an entire new board, and Xylotek might help out a bit. So things might not be so bad. As for the wire protection, are there rubber grommets to hold the wire in place like in the HobbyCNC package? I'm not sure if it makes much difference but it appears to help in my case.

I hope things work out with Xylotek, let us know how it goes.

pumelloman, It's ok. I brought in a member of the local clergy to help out today. Actually, he and I spent some time setting up the EMC2 software configuration for the steppers, and the x and y axis run considerably faster now. I definitely need to add inertial dampers to the 425 oz/in steppers as resonance is slowing down the speeds that they could attain. This a known issue with these steppers and the dampers can be home made. This clergyman has built his own CNC foam cutter and laser cutter, so he knows how to run this thing. We just had to ignore the z axis testing for the time being. I would post a photo of the cable tie-down I added but the Zone won't let me do that for some reason. It has been flaky all day for me. It won't let me post or won't let me log out or won't let me connect to the site at all. CarveOne

I received a very prompt email reply from Xylotex (Thanks very much Jeff). I swapped the z and y stepper cables and now the z stepper works ok and the y doesn't. I'll be sending the controller board back soon.

Here is a photo showing the added enclosure base plate and the strain relief method I used to secure the cables. The plate allows clamping to the bench or other means of attaching it to the bench. I removed the nails from the electrical cable staples, drilled the holes a little larger, sanded off the plastic bottoms enough to clamp the cables more tightly, and use flat head screws to fasten the staples to the particle board base. The base plate is just glued to the enclosure base plate with Titebond III.

CarveOne

That looks good. This still looks a little vulnerable though. It was a minor accident that caused this initially, so we know they can happen. ;)
The way it is sitting now, it looks like a prime candidate for some accidental shearing action should a board or anything manage to slip and fall near the control box. I would like to see a conduit or a three sided box covering the wires where the enter he control box. This would let things bounce if they are to slip and land right around there. Especially so now that you have them locked down.

Those are critical wires and should be treated with the same respect you would give a high voltage line. Not necessarily for your safety, but for the drives. :)

Lee,

I agree with your concerns. The enclosure is just sitting there for the photo. It will be solidly mounted under the table when repaired. The CNC router machine will not be on this workbench bench for the long term. I'm building it at home for use at my work location. Once it is moved the wiring will be sleeved and the cables dressed more correctly. The little 9' x 10' room that it is in now is a very cramped working area even without the machine taking up much of the walking space. Unfortunately, the computer and monitor can't be any closer to the machine and cables have to be where I don't really want them. At my work location the machine and the computer stuff will get its own a 4' x 8' workbench.

CarveOne

I ordered the limit and estop switches from Arturo @ CNC4PC today. My controller board will be mailed today or tomorrow morning.

The next thing to work on while waiting on parts to arrive is to make the caster wheel style stepper motor inertial dampers shown on the Solsylva web site in hopes these will help smooth out the resonances and improve the results of the stepper tuning efforts in software.

http://www.solsylva.com/cnc/damper.html


CarveOne

I started out with Xylotex driver baords and through stupidity fried them. You will fry them if they become disconnected from the stepper or shorted (mixed wires at stepper) That is their weakness.


My real electronic breakthrough came when I found Probotix, They have a very user friendly web site with a wealth of illustrations, showing color drawings of wiring connections, not the typical encrypted text that I have so frequently seen, and was confused by. It’s just the ticket for the electronically challenged. I studied the drivers and sent emails and found out that my six wire Xylotex steppers could be connected up either as Bipolar or Uni-Polar, Probotix has a driver called Probostep Stepper Motor Driver that supported the Uni-Polar stepper, this driver has some terrific attributes in that it has Short/Open Circuit Protection and Thermal Overload Shutdown protection. (an Open was the demise of my last drivers) It supports Full, 1/4, 1/2, 8th & 16th Microstepping Resolution. Check out the details it is well documented.
Their driver can be connected to their PBX-RF Isolated Breakout Board by simply connecting their IDC cable between the two, this makes connecting everything up even easier, and there is less chance of miss-wiring and frying expensive electronic parts.

Driver: http://www.probotix.com/index.php?vi...&product_id=12
Breakout Board: http://www.probotix.com/breakout_boa...reakout_board/
Main Web Site: http://www.probotix.com/

I have been using them a couple of months now and they are working fine.


Nice work on the machine and have to say www.probotix.com have some very nice prices for hobbyist another website worth saving to my favourites

cheers

FPV_GTp,

I think that all of the stepper driver boards will eventually need this protection embedded in the driver chips or these boards will just stop selling in favor of the newer and better design boards. Dead driver chips are really an untenable situation whether it is accidental or just due to flaky connector pins and/or over flexed cables.

pumelloman,

I picked up the casters and other hardware today. The local Lowes did not have the proper spring for sale separately but I found a bi-fold door guide rod (Prime-Line #N-6729) for $2.67usd that has a 2-1/2" (63.5mm) long compression spring that may work ok if I cut it into three equal length pieces. It's just a little larger inside diameter than a 5/16" (8mm) bolt.

The lowest cost 2-1/2" (50.8mm) wheel caster ($3.83usd) has a 2" (50.8mm) long 5/16" axle bolt with a nylon insert locknut. The special axle bolt is not usable for the inertial dampers so a 2-1/2" hex bolt and three 5/16" flat washers will be needed per inertial damper. A suitable size set screw for each inertial damper will also be needed. I expect to add a wing nut as described in the Solsylva instructions or some other type of small knob.

CarveOne

Here's a shot of the assembled inertial damper. I don't know if the spring tension is enough to tune it for best slippage/performance on the 425 oz/in stepper motors. It may need a spring with a larger gauge spring wire. When the machine is running again I'll post some subjective results here. For this photo I used the original axle bolt. I'll drill and tap the hex head bolts tomorrow and post a photo of one installed on a stepper motor.

CarveOne

pumelloman,

The only problem I ran into while making the modified bolts is that the stepper shaft is 0.250" and all of my 1/4" drills make a 0.249" hole. I finally found a 1/4" four flute end mill that will clean it out to a 0.251" by 1/2" deep hole for a nice slip fit without wobble. I drilled and tapped a 6-32 hole in the hex head for a set screw. I borrowed them from R/C model airplane wheel collars.

I already had 3" bolts on hand so I cut 1/2" off of the threaded end of the bolt. I also needed to add two threads more at the unthreaded part of the bolt so I can completely compress the spring. If you buy 2-1/2" bolts the additional threads won't be necessary because the unthreaded portion of the bolt will be proportionally shorter. I left the assembly almost fully tensioned with about 1/32" of movement of the wheel on the bolt. You can't see it in the photo but I cut a piece of brass tubing to fit between the bolt and the spring as a gap filler so that the spring remains aligned better on the bolt.

I'll install them on the steppers and see what the results are when my control board returns from Xylotex. Meanwhile, the limit switches and Estop switch are on their way here from CNC4PC and should arrive Monday or Tuesday.

CarveOne

Hey CarveOne,

Looks really good, can't wait to see if it works well. Hopefully you'll get your Xylotek back in working order soon. Thanks for all the tips on building the vibrational dampeners, I will likely pick up some supplies today and then drill/ream a hole in the bolts this week at school.

pumelloman,

If the spring doesn't have enough tension against the wheel I think I can make some neoprene washers to help add some drag to the wheel in addition to the spring. We also have a couple of well stocked hardware stores locally that may have some more suitable springs. I didn't think to check the plumbing section at Lowes when I bought the casters. Check valve springs may be a possibility.

Measure your stepper shafts and you can check the hole sizes before you take the bolts home. A smooth hole that is 0.001" oversize is just right.

BTW, the wing nut is there just for turning the stepper shaft by hand.

CarveOne

I received the limit switches and Estop switch from Arturo at CNC4PC today.

These limit switches have a small plastic roller at the end of a 1" long lever on a snap action switch. The switch has NC, NO, and COM terminals. Total travel of the lever is 1/4". I'll have to figure out how to mount these so that the lever isn't bent if the moving parts don't stop soon enough. Any suggestions from the CNC Zone experts are welcome.

The Estop switch mounts easily through a knock-out hole in a standard steel electrical box. Two screws are used to tighten up the assembly from inside the box. It has NC and COM terminals only.


CarveOne

I received the limit switches and Estop switch from Arturo at CNC4PC today.

These limit switches have a small plastic roller at the end of a 1" long lever on a snap action switch. The switch has NC, NO, and COM terminals. Total travel of the lever is 1/4". I'll have to figure out how to mount these so that the lever isn't bent if the moving parts don't stop soon enough. Any suggestions from the CNC Zone experts are welcome.
CarveOne

I used the same kind of Limit/Home Switches and mounted them as shown below.
I used some Alum angle made a pivot hole and a curved slot for vertical adjustment.
Used the same Alum angle on the the actuator, two slotted holes and beveled it so it would ramp up on the SW. didn't set the switch so high that it would be damaged if it still continued on, just to actuate about half way up the ramp.

Here's a shot of the assembled inertial damper. I don't know if the spring tension is enough to tune it for best slippage/performance on the 425 oz/in stepper motors. It may need a spring with a larger gauge spring wire. When the machine is running again I'll post some subjective results here. For this photo I used the original axle bolt. I'll drill and tap the hex head bolts tomorrow and post a photo of one installed on a stepper motor.

CarveOne

Uhhhh what does a inertial damper do? Haven't seen them on motors.

They dampen resonance vibrations that cause the steppers to miss steps or have complete stalling if it is severe enough. Dampening out the resonances allows running at higher IPMs (that's the theory anyway). The spring tension or caster size can be adjusted to tune for best operation.

There are commercial neoprene, silica jell filled, and ferrofluid filled versions available but the Solsylva mechanical version is cheap to make at home.

They basically allow the inertial mass (the caster wheel) to slip a little with each step to resist the step vibrations. If they don't work, one set screw will easily remove them. I'll at least have three casters available for use on some other project.

CarveOne

Mr.Chips,

Thanks for the pics. That looks simple enough to implement easily. I think I already have the angle stock on hand.

CarveOne

My repaired Xylotex controller arrived today. I mailed it on 1/31/2008. Less than two weeks turnaround from the east coast to the west coast (USA). I'm impressed with both Xylotex and USPS. It must have cycled through Xylotex in one day.

I'll set the driver voltages again tomorrow and see if I can be somewhat less of a klutz around the cables this time.

edit/update - I found some time to reinstall the board and it's working fine again.

pumelloman,

I'll do some more tuning without the inertial dampers and then see if installing the dampers allow improving the speed without resonances. I won't get a chance to do it until Wednesday at the earliest. As soon as I know something positive or negative I'll post the results here. If it's really positive, I'll let you know in your build log also. :D

CarveOne

I mounted four of the limit switches today. Not knowing for sure that what I'm doing is correct, I arbitrarily set the switches to trip 1" from the physical collision point. 1/2" would probably be more than enough stopping distance, as this thing traverses slowly enough that it doesn't skid to a stop at all. The switches can be rotated a little and the ramps can be bent a little to give any adjustments needed.

The ramp is 1/16" aluminum and seems plenty stiff enough to not flex under normal roller pressure. The ramp will pass by the switch without switch damage if the software doesn't see the switch closure and crashes the gantry into the physical collision point.

I epoxied the limit switches to a 1/16" alumimum plate that is roughly 1" square. I roughed up the limit switch surface and the plate's surface with 80 grit sandpaper to give better adhesion of the epoxy.

The Z axis switches may be a bit more tricky to mount than the X and Y.

CarveOne

Wow, that was actually pretty smart (or at least I thought so). I hadn't thought about what could happen if the software doesn't recognize the switch. I especially like how the gantry doesn't smash into the limit switch to stop, it gradually pushes it down. Very nice!

I have run my gantry into the side rails on purpose to see what hits first and what happens. The collision is not violent at the speeds that we will be running these at. The steppers just stall like they do during resonance. I don't care if the switch arm gets bent a little but I would rather not have to replace broken switches. Not recognizing the switches can happen during power glitches if everything is not on a UPS backup supply or during Windoze lockups. I have not seen any lockups in Ubuntu Linux and EMC since I've been using it. It's still worth installing your switches to save you effort later.

I'm also trying to keep most of the switch wiring on one side of the machine as much as possible for ease of wire routing.

I'm still not sure how I will fit the switches into the Z axis yet. Doesn't look like it will be as easy as the X and Y. I'll have to study on that a while. I don't think they will be so elegant when finished.

CarveOne

I am reading reading reading your design and work u posted here... and i am really happy with your neat work on wood... I wish to see this machine working... I will post details when i read all these interesting pages...

Wish you best of luck..:)
Regards
KFK

Kahlid,

Since deciding to build this machine with red oak I have found that it seems to be in more use than I had first thought. When I glued the 1x4 oak boards together to form the 2x4 side and end rails I oriented the annular rings in the boards so that they oppose each other and will, hopefully, produce very little warping and more stiffness. So far, it has been inside my house and everything has remained stable. The wood parts all have five coats of satin polyurethane sprayed on them and I don't expect to have moisture problems. It's not steel, but it's as close as I can reasonably get with wood. (I hope)

Thanks for taking a look at my humble efforts.

CarveOne

I like the way they look. did you cut them by hand? i have some 1/4 inch laying around and would love to make a set. If you cut them on another cnc do you have the files used so i can make a couple for myself? Thanks and congrats on a sweet looking setup.

Jim

Thanks Jim,

I don't have the CAD files or gcode for these parts. Even if I did I doubt that David Steele would be happy that I gave them away. I'm sure that he prefers that you buy the plans from him if you don't already have them. These were easy enough to mark out and cut.

The 1/4" Lexan that I used came from Lowes. As this stuff melts at low temperatures you will need to use a coarse tooth bandsaw blade to cut it. The holes were cut with a hand held coping saw. A power scroll or jig saw will melt the plastic and it will fuse back together behind the blade.

The Lexan seems to be holding up pretty well so far. I was going to make them from 3/16" aluminum plate until I found the Lexan sheet I had left over from another project was large enough to make these plates.

CarveOne

Thanks Jim,

I don't have the CAD files or gcode for these parts. Even if I did I doubt that David Steele would be happy that I gave them away. CarveOne


Yea. I have the book. I will attempt to cut them on a veriable speed scroll saw. I have enough to test and see if it will work. Thanks for the info.

Jim

I tried my variable speed scroll saw first and had to switch to the coping saw because even though the scroll saw could slow down to a very low speed it would still overheat the Lexan. Maybe yours will be better. Though you normally don't need to do this with a scroll saw, I probably should have tried using some sort of light oil to lower the blade friction. When the Lexan melts it will fuse back together behind and around the blade, trapping the blade.

CarveOne

Water or air works pretty well to cool most plastics. Sometimes a couple of drops of detergent helps the water.

Garry

I'll have to try your suggestions next time I cut som Lexan Garry.

Would need to be sure that water doesn't get to the motor though. I thought about putting a thin trail of dish washing liquid along the cut line to act as a blade lubricant, and it wouldn't easily get into the motor since it would tend to stay where it is placed.

CarveOne

If you can direct a stream of compressed air over the cutting zone you might find it works pretty well.
It doesn't really generate a lot of heat it just melts at a low temperature.

Garry

If it's actually lexan or polycarbonate, you shouldn't be having any melting issues. You will have those issues with a