jgro Style CNC Build Journal

[ignatz]

Just a quick update to note my progress. Currently I am adjusting the machine. I have been able to get motion in all three directions but the Y axis binds a bit and the Z axis wobbles on the way up. Unfortunately, I blew 2 driver chips in the process so I'll have to wait for the weekend to make any more progress.

Does anyone have a suggestion for a finish? Since I used hardwood, I'd really like to stain and varnish it. I noticed that many seem to paint. Is there a specific reason for that choice?

[Dave's_Not_Here]

"Does anyone have a suggestion for a finish? Since I used hardwood, I'd really like to stain and varnish it. I noticed that many seem to paint. Is there a specific reason for that choice?"

But as a hardwood furniture and cabinet maker.... I gotta tell ya...

A lot of the DIY mills are made from MDF. It is a compressed composite of fibers and resins under extreme pressure, somewhere in 50,000 lbs. range. Therefore you get extremely high tolerances. However, you also get a tendency toward swelling if the MDF is not sealed properly.

To seal it you need to put Shellac on ALL surfaces, at least two coats on the flats and four or more on the edges. Some people put a slightly thinned mixture of yellow/white glue on the edges first. The edges are where the moisture in the air will penetrate first. After that, follow up with block sanding between coats of Shellac using 220 grit sandpaper. Follow the Shellac with several very light coats of high quality paint and a clear sealer coat. Shellac is used because it dries quickly, sands easily and makes a great base coat for paint.

Now before everyone thinks I missed Ignatz "hardwood" material useage, chill out…

All I did was set the stage for a comparison of the majority of DIY painted mills and Hardwood.

Hardwood by it's very nature, will ALWAYS absorb moisture into its cellular structure. It is trying to "equalize the to the relative humidity" in its current surroundings. In my area, the Relative Humidity (RH) is about 12%-14%. In my shop it is closer to 8% and in my house 6% to 7%.

As a furniture and cabinet builder, I have to let my hardwood acclimate to its surrounding and the RH before building anything. Why? Because the wood will expand and contract until it equalizes to the RH in its environment. Ever have a painted or otherwise sealed door or window stick at different times of the year? Guess what… its equalizing to the RH…

So, Ignats, to answer your question, you can stain and Shellac your mill but it will never remain constant dimensionally because hardwood has a cellular structure that will ALWAYS absorb or release moisture… it will NEVER be 100% moisture free.

MDF on the other hand, has no cellular structure and if sealed properly and kept away from direct and continuous moisture, will remain exponentially more stable than hardwood. In solid wood furniture, wide boards (like table tops and desk tops and raised panel doors, etc) are attached with additional space at the joints so the wood can expand and contract without exploding in the middle if the night! Conversely, more and more cabinets are being made out of MDF and painted…

Hope I didn't ruin your day or rain on your hardwood…

If you can keep your mill in as controlled an environment as you possibly can, you will still have movement, but hopefully not as much as in an uncontrolled environment. It is not unusual for some hardwoods to move more than 3/16 of an inch… in a CNC environment where we measure by the .001… 3/16" is like looking at the grand canyon…

[ignatz]

Wow, Dave.... Thanks so much for the information. I wish I would've known that before I began. Too late now. In my area(Central Florida), the RH is as close to 100% as it can get until it rains almost everyday. I've had major problems in the past using MDF so I thought that hardwood would be a better choice. Do you think that I could use a Deck Sealer to completely seal the wood? What about a latex paint? Given this situatoin, what's the best choice for a finish to help minimize material movement... or should I just use this machine to cut out MDF pieces for another machine?

[Dave's_Not_Here]

... a tough place for humidity.

I looked at your pictures late last night and was unclear if you have used plywood or solid hardwoods or a combination.

As for a deck sealer and latex paint, about all that is good for is mostly, is penetration of standing water. I am sure you have seen the gaps in deckboards change over time. Penetration of standing water will be minimized which will slow the rotting process but moisture will not be eliminated from the equation.

As for painting the wood, latex is more of a surface paint not much of a penetrating material. It cleans and thins with what you want to avoid... water
Oil based paints will have more penetration and protection, but will still allow moisture to move through.

You have no doubt seen the moisture vapor barriers (Tyvek Wrap) being put on the outside of new home construction prior to siding... that prevents moisture that will penetrate the siding from penetrating the walls of the home. Reason? to prevent mold and dry rot. It also means that the outer skin of the house, even though painted, stained, etc. is still "breathing" or allowing moist air to travel... which means humidity.

I guess you need to decide on the tolerances you can live with, keep an eye on dimensional changes, possibly make a set of "replacement" panels out of MDF so when the time comes you can disassemble and rebuild easily and not worry about making them "after" the current mill is out of wack.

Depending on your budget, you can, over time, make or acquire critical part replacements out of aluminum or other materials, like Phenolic, etc. It will be more expensive, but in the long run, more stable and predictable.

MDF, when finished correctly, will be reasonably stable for quite a long time. If it weren’t you would not see it going into so many new home's kitchens. The process of finishing these cabinets involves a process similar to powder coating. Heat actually bonds the paint to the surface. In some instances, there is a process where powder coating can actually be applied to wood, but then Aluminum would become cost effective then!

Lastly, as for RH in Florida… it is not so much the % of humidity… it's the change or fluctuation in humidity. If Florida was, say, 96% RH and fluctuated +/_ 2%, that's not such a big change therefore less movement in your wood. But if it changes by 30%, now you have issues. So, pay closer attention to degree of fluctuation in RH not so much actual RH.

If you decide to re-do some replacement parts from MDF for future use (contact me for finishing procedures), I would advise getting the 1" or 1.5" MDF. VERY heavy and more $$ but still pennies compared to Alunminum. I pay less then $20 for a sheet of 3/4" MDF at Home Depot and 1" is closer to $35-40 at specialty cabinet suppliers or sometimes lumber dealers. If you can keep your mill inside some place and control RH fluctuations, you will minimize your problems, not eliminate them, however, it you are relegated to a patio or semi-open structure… not a good thing.

Another choice is MDO. It's the same stuff the states use for road signs, Has a paper coating bonded to thick plywood. Again, finishing and especially edge treatments are critical. More expensive… sometimes easily over $100 for a 4x8 sheet…

All in all, you could get some 80/20 from their Ebay Garage Sale site
(tough to beat some of these prices…)...http://stores.ebay.com/8020-Inc-Gara...Q3amesstQQtZkm... and only use the thick MDF or MDO for replaceable tables… then replace them whenever they get bad… save a lot of money and time, and headaches in the long run… after factoring in repeated replacement costs, including cost of running down materials, MDF can become more expensive that extrusion over time… cheaper way to initially learn though…

[ignatz]

Dave - Thanks again for the thorough and speedy reply. I'm not using solid hardwood. It looks like several thin layers of wood glued together but I think it was labeled as (Bonewood or Blondewood) Hardwood. I *think* that makes it plywood, right? I can't quantify the RH change daily in this area but it has to be pretty high. I will have to talk to the Boss about moving the machine inside but its currently in my garage. My original goal for this machine was to have inexpensive starting point that would leave me with key components that I could reuse to build a stiffer machine later when/if my abilities demanded it. I've learned a lot from building this machine so far, but it looks like I should just save my money for aluminum and leave this one un-finished.

[Dave's_Not_Here]

I do not mean to imply that you are wasting your time. I would definitely keep going. Anything you learn can and will be applied to your upgraded machine and your learning curve will be used on inexpensive wood.

Once you have it working well, all you have to do is use the wooden structure as a template. Additionally, after you are finished and running for a while (forget about .001 accuracy right now) you will find a number of things you wished you did differently relative to the style and such. Better to find that our in wood rather then Billet Aluminum of Aluminum Sheet Goods.

As far a being in the garage goes, that is not a bad thing. It's much better than a carport or a patio. Plywood will not change nearly as much as solid hardwood. It will have some movement, more than properly sealed MDF, but not nearly as much as a stick of Maple, Oak, etc.

On your edges, where plywood wants to warp, you can use some aluminum channel and a bunch of self tapping screws to hold it in line, similar to a rail and stile constructed raised panel cabinet door. Just make sure you get at least 1/8" material that is at least 3/4" wide so it will be stiff enough to resist the plywood wanting to warp. It isn't all that expensive. And when you get your Aluminum mill built, the channel can be used to mount things to your table, so it won’t be a wasted investment.

By all means, finish the mill, I guarantee you will be glad you did, especially once you start cutting solid Aluminum parts for your upgraded machine. Buy solid Aluminum when you can, cut pieces, mill holes, etc, and set aside for the big tear down and rebuild! Get some Ebay 80/20 extrusion for your table base frame and maybe even your gantry y axis. Mill solid stock for your gantry upright supports and if you want, for your gantry y axis, in place of the 80/20. Mill your bearing supports from Billet Aluminum… Your Z axis mounting plates and router brackets...etc.

Get some 3/4" (16mm) to 1" round rails and bearings from http://cgi.ebay.com/ws/eBayISAPI.dll...521709535&rd=1 for your repalcement X axis guides (smaller ones are available for your Y axis), heck, I will even give you a set of Thomson rails and supports along with Super Pillow Blocks for your Z axis, but only if you commit to finishing the wooden mill! When you start to make replacement parts for your upgraed, e-mail me... I will set them aside and it can be your incentive to finish and move to the next level.

Then you can start making plates and router brackets to mount to the Z axis I will give you... Here http://www.onlinemetals.com/ is a place that has all kinds of metals at a good price and no quantity is too small...

Do it a piece at a time using your wooden router. Keep an eye on movement of the plywood, but if you brace it and cap the edges, it should be within tolerance long enough for you to jump on making parts for your upgrade. When you are done, you will reuse some of what you had in the wooden mill, so you won't have wasted anything and gained a huge amount of experience. If you stop, you will regret it and might not even finish an Aluminum beauty…

Don't make me have my Marine son pay you a visit….

[ignatz]

Dave you're the man. Thanks again for the support and the help with my project. Your information has made me decide to complete this router and start on its alum. replacement sooner than I had intended (all assuming I can figure this 3D CAD software out). Cheers!

Now for a quick update. The build phase of this machine is complete. I connected the table and support rails usings Larry's Suggestions. The only difference is that I bought my blind nuts from a local store called Woodcraft (Be careful, this place has lots of toys that are very expensive). I have to fine tune the alignment of the gas pipes and make sure that the ABNs are all adjusted properly. If anyone has any alignment/tuning tips I would welcome the input.

I got the new driver chips from HobbyCNC and replaced the two that had chips missing. All four axises work now and I can control the motion of the machine from within TCNC. I still need to fine tune the TCNC settings for the motors. Currently the machine is jogging at 10IPM. Does anyone have any suggestions for the following settings; Pulse Width, Accel, Start vel, Max vel, Slow jog, Fast jog? I don't really understand what the TCNC manual has to say for these options.


Here's a photo that shows the current state of things.

[Bubba]

Ignatz,
Basically, I would set up "reasonable" Low values. When I set up my machine, I tried to get real high rapids, but found that my start values were to high and was actually "snappng" small end mills as the head would move very rapidly initally and have to "slow down" for the cutting speed. I would have start values no more than my slowest cutting speed.
After that, select a top end values and try it. If your start missing steps, slow it down. Same goes for acceleration. Set a slow value and then start increasing it until you have problems and then back it off.
It is hard to give absolute values as each machine is different. Also, once you have done it and then later on find it is not suitable, don't be afraid to change it.
In my case, my base values are as follows (akk values imperial):
Default_Feed=10
Acceleration=16000
StartSpeed=334
MaxSpeed=15000
Fast_Jog=5000
Slow_Jog=175

Hope this helps, but also realize my system uses servos and YMMV

[ignatz]

Hi Bubba,

Thanks very much for the feedback. I played around with your values and they were too high for me. I did find the following information in the TurboCNC user group that helped quite a bit.

I wrote a fair number of posts covering setup. You should be able to find them around the same time frame that AccelErr.cnc in http://groups.yahoo.com/group/turboc...es/GCODE/REL4/ was posted. Basicly set Start Vel to an obnoxiously high value (1000 IPM), Accel to a low value (400 or less) and Max Vel to the same as the Start Vel. The high values are to override the limits so that you are not constantly working in the configuration menu. Write a short program to move the axes at controlled speeds, return speed should be about 1/3 of the test speed. Measure error on return. Preload the axis in the return direction first to avoid backlash from interfering with your measurements. Sample program: N100 #1000 = 1 ; 1 IPM default Feed Rate N200 ASK #1000 ; Test Feed Rate? N210 G01 X1 F#1000 ; Move at test Feed Rate N220 G01 X0 F[#1000/3] ; Return at 1/3 test Feed N230 M97 o200 ; goto line 200 Once you've found the Start Vel where you don't lose steps, set this value in the axis configuration. Now you can use the same program to determine the Max Vel. Ensure that you can accelerate to your chosen test feed rates in less than 1/2 the distance that is specified in the program. Remember that you need room both to accelerate to, and decelerate from the test speed, as well as travel at the test speed. Tony TurboCNC Development Team

Using this method, I found that my top speed was near 27 IPM on my X axis. I haven't ran the test on the Y and Z axis but this speed seems to work well on my machine.

Quick Update - It's Alive! (Almost) The machine moves in all three directions and now has the Rotozip mounted. The Rotozip is not square to the table so I might re-make the mount but I want to cut something (anything) first. I have to find a way to transfer files to the CNC machine... either network or USB.

I also need to figure out how to hold work pieces to the table. The JGRO plans call for aluminum hold down tracks. My problem with this solution is that I am concerned that I need to cut the slots with the machine itself so I can be sure that they are square... but I need to test the machine to make sure it is running properly so I'm hitting a Catch-22. Can anymore recommend a solution to this dilema?

[ignatz]

Here are some pictures of my first attempt at cutting with my machine. I am using a cheap HF bit and Steve Mazner's 12 Inch Ruler GCode. I think it turned out OK but I still need to address the clamping and tool height issues.

[js530]

Look great. How did you attach the ABN to the carriages? I'm building a machine myself, and I have no idea to avoid flexing.

[ignatz]

js530 - Thanks very much! I'm not sure I completely understand your question, but I think this picture shows how I connected my ABNs to the carriage. The image shows the X axis on the left hand side and the Y axis on the right. I didn't have many problems with leadscrew alignment on the Y and Z axises but the X was another story.
My X axis leadscrew is slightly bent and wobbles when it is spun. The wobble causes whipping. To deal with this I did two things, 1. I switched from the aluminum couplers that I made to LoveJoy couplers. 2. I changed the Anti-Whip block slightly. Instead of cutting two .281 holes in the Anti-Whip block, I drilled three slightly overlapping .281 holes to make a slot. Now, when the screw turns, the Anti-Whip block moves from side to side slightly instead of causing the gantry to whip. I hope that helps.