FLA-300 4' x 4' R&P Router

[brtech]

The second shipment for my Fine Line Automation FLA-300 router arrived last Tuesday.

I believe I have the first of these, and being first is always a problem of patience, understanding, and willingness to deal with the inevitable "oops". This one was especially trying because Fine Line really is a very part time business, with a proprietor that has a full time job, is unable to predict when he will be able to do things, and occasionally stops communicating.

If you deal with Nate on a new or modified project, you can probably expect it to be late - his original estimate to me back in March was that it would be ready in May. I got it in September. You can expect him to say he will ship things on certain dates, and he will sometimes make it, but more often, he won't. He will occasionally start a UPS shipping process, you will get the tracking number, but then he won't actual ship it until days later. I've gone an entire week without any communication with him in the middle of one of these start ship with no actual shipment. You can expect some small problems with the parts (which I would have been surprised if I didn't have any on a first of a kind).

Having said that, the product is excellent, when he responds, he is helpful, and the value, I think, is very high.

If I had to do it again, with what I know now, I definitely would. It's that good. Just be prepared, be patient, and eventually, it will be worth it.

And with that out of the way, on with the assembly.

You get a bunch of parts and a SolidView model of the completed assembly and several subassemblies. There is NO other documentation. You are on your own figuring out what to assemble, how to assemble, how to align, etc. I made a lot of mistakes in this process, but it came out okay anyway. I'll try to point out things I managed to get wrong the first time.

When you assemble, in most cases you put fasteners into one side, then slide the other component in place, then tighten. For example, when you assemble 8020 extrusions with butt joints (a cut-off end of extrusion joins a side of another), you put the Button Head Cap Screws (BHCS) into the cut off end, and slide it on to the t-tracks of the side of the other extrusion, align the access holes in the side to the BHCS and tighten with a hex key. BTW - while I had a okay set of longer hex keys, I bought a 3/16" T handle hex wrench for these BHCS. This, and a 1/2" socket were the two tools within reach at all times (plus my squares).

Attaching fasteners and sliding is used a lot. The easiest way, for example, to assemble the rack, is to attach the BHCS and T nuts to the rack, then slide the rack onto the X or Y support extrusions. The only thing I found this didn't work on was the cold rolled steel rails for X and Y. These use carriage bolts in the extrusions and lock nuts. I found it too hard to hold up the rail while rotating the carriage bolts to slide into the support extrusion T slots. On those, I slid the carriage bolts on and then aligned the bolts with the holes in the rails.

You have to be aware of when there is another component that needs to be able to fasten to a T nut after you assemble the extrusions. In those cases, you have to remember to get the T nuts in place before you join the extrusions or other components. For example, the middle legs connect to the middle of the bottom bed frame sides with t nuts. You have to remember to put the nuts on the extrusion before you assemble the ends of the frame!

Another tip: align early, align often. Always checked for squareness at every step. I will say that even though I did that, I have, and still have, some alignment problems.

This is a big unit, and it comes in lots of parts. I got mine in two shipments. The first had most of the 3030 extrusions, plus the racks and rails.



The second had one 3030 (the Y axis support), the 1530 parts, and all of the small parts. The 8020 components come in long thin boxes with peanut packing, which gets everywhere as you unpack. One of the boxes was nearly cut through in the top where one long extrusion was packed with several smaller ones. A couple of dings, but no problems in the shipment.



The small parts box has a lot of parts



There is a packing list/Bill of Materials - check carefully, I eventually discovered that one bag of bolts was labeled correctly but had the wrong sized bolts in it! No big deal, but it was a surprise during assembly.

[brtech]

I started with the bed: a rectangle of 8020 extrusions that hold the spoilboard and form a brace for the frame. On my extrusions, the threading on the ends wasn't great. I had trouble starting the BHCS. I got into the habit of running a tap down the holes before assembling and that was pretty fast and easy, and I didn't have any more problems. The tap would very often come out with a lot of chips, so maybe what Nate was using wasn't cleaning the threads effectively enough. I used my square, and tape measure with a corner holder, and got it square within my reading ability of the tape.



The bolts coming up from the top of the extrusions will hold down my spoilboard.

I then attached the legs. You have to remember to put the middle legs in place before the end legs which block the T nut entry. I used an adjustable square to get the leg heights the same. It doesn't matter what the height above the bed is, but having them be the same is important. Locating the middle legs is tough, because there is no guideline on where they should be located. When you put the X support beams in place, they have access holes for the BHCS that connect them to the support beam. That is what determines where they go, but I put the beams on later in the assembly. You can measure from the model or the support beam to locate the middle legs. I found that these legs get in the way of assembly a lot, and I ended up letting them be loose until I got everything else lined up.



The frame is rotated in this picture: the middle legs are on the sides.

With the legs in place, the frame was put in about it's final location, and I started adding the X support beams. Again, put the BHCS screws in the top of the legs, and preassembled the corner support gussets with BHCS and T nuts. I slid the gussets on the leg and temporarily tightened one screw to hold it in place and slide the beam on to the legs and gussets. I needed to loosen the gussets to let them float as I slid them on but it was pretty easy to loosen the one bolt, get the T nuts lined up and slide, then repeat.



I thought that by making sure the legs were square to the bed, that getting them square to the support beams would be easy. Not so. In fact, by the time I got it all together, one leg was way out of square: I couldn't line the BHCS with the holes in the extrusions it was so far off. I ended up loosening the leg completely and squaring it. Then I was able to get the support beam in place. One of my extrusions had a rough edge, which prevented the beam from sitting flat. A file took care of that.

Then I added the rails and the racks. The racks slid on with the BHCS and T nuts preassembled. For the rails, as noted above, I slid the carriage bolts, roughly aligned them, laid the rail on top and then slid the bolts under the rail to match the holes. My T handle hex wrench was helpful. I found it easier to tilt the rail so I could get one row of bolts in first, and then leveled it out to get the other row in place. Then I added the nuts. I only tightened one bolt, so I could align the rails with the carriages on Y. On my router, I want to be able to route dovetails on the ends of boards. For that reason, I have slid the rails and racks forward a few inches so that I can clamp a board on end and have the router cuts the dovetails.

[brtech]

Then, on to the Y. Slid the rack on the same way as the X racks, and assembled the rails the same way.

The carriages were assembled. There was a machining problem on the extended carriages - there are two bolts that are threaded into the carriage which hold bearings. On mine, the threading wasn't deep enough and I couldn't get the bolts started. I ended up threading the bolt from the back, and using by socket wrench to force the steel bolt through the aluminum. It was effective enough, and I was able to easily back out the bolt and thread it in from the other side. I was unsure about how to assemble the carriages. The fixed bearings go in tight with lock washers. There are two other bearings that float - there is a captive nut and a set screw that adjusts the spacing between the bearings (against the top and bottom of the rails. Too tight, and you can't get the carriages on the rails. To loose and they don't sit straight. I ended up backing out the set screws, and leaving the bolt very loose. After assembly, I clamped the carriages together around the rails, and then tightened the bolt and the set screw. This is much easier said then done. My wrench had a hard time getting in between the carriage and the rail. I used a needle nose pliers to get the bolt down enough so I could get the wrench in there. I attached the carriage bolts to the carriages with the lock nuts, slid them down the Y support, clamped the carriages together and then attempted to align the carriages square with the Y support beam. Also easier said then done.



This picture was taken before I realized I had the Y on backwards - the rail goes towards the front!

You have to remember to put a T nut on each end of the Y support for the corner blocks that go on the inside carriages before you attach the outside carriage. After getting them assembled, the Y axis slid nicely up and down the X rails. I attached the corner blocks then. Aligning the Y carriages to be square to the Y support beam is tricky - parts in the way of getting a good square in there. One hint that you have them square is holding Y near the end of the support beams and measuring the positions of the carriages with respect to the beams. If the Y beams are aligned with each other (the frame is square), then the carriages should be the same distance from the end of the Y supports. Mine was way off. I had to loosen, skew and retighten.

Then I assembled the Z. I found the illustration on cncrouterparts.com to be very helpful in putting it together. I got stumped for a bit on the drill rod. Nate (I assume) had kindly pressed the rod into the bearing, and I assumed had set it to the right length. Alas, not correct, and after scratching my head for a while, I encouraged it to slide down some to fit properly in both the couplers to the motor and the screw. I'm still a bit unsure if there is supposed to be any play at all between the washer holding the needle bearing in place and the coupler or collar. I have them somewhat snug, but definitely not tight. At this point I found a missing part (the BHCS that attach the anti-backlash nut to the Z plate). I had some bolts (with the wrong head) that will do until Nate gets me the right parts. The assembly went fairly well except for those problems, and I got the whole Z up without much fuss. I can slide the Z up and down by rotating the screw by the hand. I had ordered a K2CNC mount for my (new) Hitachi MV12C router. Looks good.



The Z attached to the Y the same way the Y attached to the X, sliding the carriages down the extrusion, clamping them together, aligning and tightening the nuts. I was doing this myself and was really hoping that the clamps would hold things together well enough to let me get the Z mounted and aligned. It seemed to go well enough. Again, it's hard to measure how square the Z is to Y, but measuring distances top and bottom to an edge seems to be helpful. The Hitachi mounted easily. I adjusted the Z position so I had what I judged to be low enough that my shortest bit would just enter where my spoilboard (two layers of 3/4 MDF) will hit.



The R&P drives were next. I followed Ahren's video explanation, which was okay. I hadn't noticed the shims that were in a package with the belt, springs and turnbuckle. They weren't in the model, but a high collar lock washer was. The order is bolt, shim, plate, washer, carriage. I then found my second real parts problem - a 3/8 socket head cap screw was needed, but a 5-16" SHCS was in the package, and the length was wrong. Nate will send me the right parts. I'm going out to the local big box for something that will work in the mean time.



Another R&P issue I found was that the pinions are only engaged 2/3 of the rack. The high collar lock washers are spacing the pinion out pretty far from the carriage, and thus the rack. If they were narrower, more of the rack would be engaged.

But, modulo those screws, and thus the engagement of the pinions on the racks, it's mechanically assembled. Next up, the wiring to the Gecko and getting Mach to run on it!

[jemmyell]

Hi,

This looks really great! I SO want to build one of these. Did you order a complete kit or just the mechanicals? How much was it?

-James Leonard

[brtech]

I got a complete kit, with the motors, gecko, power supply, mach, etc. The only thing I'm adding is the computer, spoilboard, router and K2CNC mount. While I may have been able to save a few dollars getting parts from other places, I decided I wanted it to work right the first time and have a single finger to point at. I don't often do that, but I did on this purchase. YMMV.

I got a discount to be a guinea pig. I'm helping Nate get the model right, the BOM right, etc. You should contact him for a price.

I got the motors spinning. Something is wrong with the A axis slave to X in Mach (I changed the setting and restarted Mach), which I'm sure I can figure out soon.

[titaniumw41]

Wow, looks awesome! I have been designing a 4x4 R&P myself that looks very similar. I don't suppose Nate/Ahren would let you share the preliminary BOM and e-drawings with the community? I would love to see them to compare. Who knows, maybe if it is available soon enough (within 6 months or so) and the cost isn't too bad, I might end up ordering a kit too! I'll be watching your build very closely. Great job so far, very descriptive. Good luck!

[brtech]

It would be up to Nate to post the model/BOM. He has done so with the FLA-100, but not (so far) the -200 (2 x 4) or -300 (4 x 4). I think by looking at the FLA-100 model, the existing documentation on the R&P drives, racks an rails, and my picture, you could very straightfowardly rebuild the model. The X is really just the basic cncrouterparts drive, and the available FLA predrilled rack and rail. The Y is exactly the same but there is one of them instead of 2. The Z is exactly the same as the FLA-100 Z.

Everything is 3030 except the center supports in the bed, the center legs, and the Z, which are 1530. All the joints are BHCS into ends tapped 5-16 except the center legs (which are connected to the bed with BHCS into T nuts). The rails connect just the way they suggest: carriage bolts into holes on the rail with locknuts. The racks use BHCS into T-nuts and are countersunk (available as a stock part from FLA). The Y beam really just rides directly on the X carriages with carriage bolts and locknuts from the 3030 through the carriages. The Z works just like the FLA-100 Z, the carriages connect the same way (carriage bolts and lock nuts). The only flat plate aluminum is the R&P drive plates and the Z axis plates). You can see the gussets in the picture - they all connect with BHCS into T nuts on both sides, except the gussets on the Y beam to X carriages, where one side is BHCS into T nut, but the other side is BHCS into the carriage itself.

I admire the simplicity - by having the X support beam mounted atop the longer legs, with the bed suspended part way down, the Y becomes extremely simple - just the support beams riding on the carriages. The rack and the rail help stiffen the support beam, and the R&P drives are pretty beefy, so using 3030 as the Y support with a full span really does work. I'm out of town now, but I will measure and post the outer envelope for those who want to do a room plan. What I remember is that the bed is 54" deep by 64" wide, which makes the Y beam around 70".

I suspect that if you ordered a -300 from Nate today, it wouldn't take him too long to deliver it; the design is done, and the BOM is now good. The first one was the one that took so long - running another won't be nearly as hard.

[brtech]

Got all 3 axis moving. I had some trouble with a Windows 7 Mach install. On my first try, no errors, but Mach wasn't displaying bitmaps. After a check with the Mach3 support forum that others had successfully installed and were running Mach on Windows 7, I uninstalled it, deleted the Mach3 folder, and reinstalled it. Worked fine after that. No explanation of why the first install failed.

Couple of tips for those who come after me: You need to slave the A axis to the stepper on the other side. When you do that, you must reverse the direction of the Step signal! The two motors face each other, and therefore have to spin in opposite directions to move the gantry. Duh.

I have been calling the axis that moves the gantry "X". I think I've concluded that I should change it to Y. The way Mach works, and the way I am used to trigonometry, X is left/right, Y is in/out (or forward/backward). The way the FLA-300 is built, the gantry moves the Z left/right, and the side motors move the gantry in/out. That would make the double axis Y, not X. That means I switch the cables for X and Y and slave the A axis to Y instead of X. I'll bet if I go and look carefully, most of the routers out there are wired up that way. Silly me.

I copied the starter G540 XML file for Mach3 from the geckodrive website for my Mach profile. It has default values of 20000 steps per inch. The correct values for the R&P drives are 1909.861 (from the cncrouterparts.com website). The Z ought to be 200 full steps, 2000 microsteps per revolution, 10 revolutions per inch, so 20000 is probably right. My next step is verifying that I have these settings correct. I boosted the speed to 200 ipm on X and Y for now, but I should be able to get considerably higher.

I have to work out some temporary cable management until I get some echain rigged up. I'll probably just suspend the cables above the router for now, along with the dust collector hose. Mounting echain on this design is not so easy. The R&P drive wraps around the top and sides of the extrusions, so you can't put the e-chain there. I think I can cantilever it off the bottom side track extending it out past the drive. Gotta look through the McMaster Carr site and see if I can work up some parts for that. Maybe a carriage bolt, an inside threaded aluminum sleeve, and something to keep it sitting straight = thin angle iron or maybe just big washers. Suggestions welcome.

If you need to know, the outer envelope of the FLA-300 is 71" wide x 66" deep x 55" high.

If I get the cables hitched out of the way, and the steppers calibrated, I'll make some chips this week!

[arizonavideo]

I think this is one of the fastest builds I have sen yet.

It looks like a nice router.

Gook luck with the first chips.

[jemmyell]

Hi,

Can the router extend to beyond the end of the bed so you can mount sheet goods vertically and dovetail them?

-James

[brtech]

Yes it can. I've positioned the racks and rails so that the router bit extends 2" beyond the end of the bed so I can clamp boards vertically and route dovetails on the ends.

There is more than 48" of travel on both X and Y.

The bed is also clear enough so that you could cut a 4 x 8 in two parts if you wanted to. I have the router located against a wall, so I would not be able to push the sheet in farther; I would flip it end for end.

ArizonaVideo - It's a kit - all the metal is cut to length, drilled and tapped as needed. I only had to screw it together. There is basically no metal adjusting, although as I said, I had some threading issues that needed me to do some work, but not much. That's the advantage of getting a complete kit from FineLine.

Having done it once, I think I could put another one together in maybe 10 hours. It took me a lot longer than that on this one

[jemmyell]

Hi,

Do you think the current layout of the legs is stiff enough to put casters on all four corners?

If not, I could put stretchers between the legs I think.

-James