[FoxCNC1]

Good point. I have had one of mine fall out.

[pete1961]

I had used thread lock from the beginning. It was apparent to me that it was necessary when I noticed the motor shafts didn't have flats on them.

[knightsofnii]

i also used setscrews that were a pinch longer and had some knurling on the tip.
If your shafts dont have a flat section, perhaps check that this part hasnt weakened?

[FoxCNC1]

mine do.

[fortyeight]

not concerning your topic.
But i was wondering about these kits. i am fairly new to cnc i have a small home built kit and lately all i have been at is looking at this kit pro4828.
was wondering exactly what its cutting size is, is it 48 by 24. how was the build process and the how unit working out.
Would appreciate all i can find out.

[FoxCNC1]

The units works well. Just you have to be very methodical with the build etc - but that is the same for all machines.

[fortyeight]

thanks. i would imagine so, taking time and care with assembly will probably make life easier when up and running.
And regarding it cutting size, is it a full 48 by 24

[quinnkyle]

My CNC is doing the same thing with Mach3. Once calibrated, it will immediately throw itself out of calibration from cutting or even verifying calibration. This is affecting the X axis only.

I've ruled out all mechanical issues. Set screws are tight, I've measured precisely and over long distances (a 5'x10' table). A week ago I was cutting accurate parts on a new DIY CNC. I cut 12 sheets of plywood just fine. Suddenly I've ruined 4 sheets of plywood due to inaccurate cuts. The x axis is to blame.

I've eliminated all mechanical issues, and pinpointed a problem with Mach3: the g-codes and actual coordinates don't match as the part is being cut. In other words, the g-code will tell the router to go to X: 5 and Mach3 will stop at 4.59 on the actual readout of the router's position. I do suspect the x axis motor should be replaced, but after reading a previous post I'm rethinking that.

I'm thinking that n terms of software and chain-based cogs with nema motors, it's possible Mach3's interpretation of the g-code gets compromised. The g-code is right. The CNC coordinates are wrong as it cuts.

Since the X axis has no play when I attempt to move it when the motors are engaged, so there is no play in the x-axis. With motors off there is no resistance to movement.

Here's what's weird:

If you watch the g-code window as it cuts and compare line by line to the actual coordinates Mach3 cuts, you'll see the coordinates don't match the g-code on the affected axis. No amount of calibration will change that interpretation. In fact, I find once calibrated, this bug throws off calibration, and the CNC is at least 1/16 under cutting.

[ger21]

Are you saying the g-code says X5, but the Mach3 DRO says 4.59?

What electronics are you using?
Motors rarely go bad.

[neilw20]

If step rate or acceleration is too high stepper motor can miss steps.
I the step pulse width withing the specs of the driver. Set to 5uS should be safe.
Rapid reversals can sometimes make you lose steps. Is the step phase correct?
See the driver specifications for direction/step setup times. If direction changes, with step signal at incorrect level, steps might be lost.
You might need active high(or low) instead of low (or high) on the step signal.
Any of the above can cause your problem, not withstanding there are no mechanical issues.

[quinnkyle]

usually 4' x 8' 3/4" plywood (birch)

[quinnkyle]

I ended up putting a new motor in, which worked well. Noticed the set screw was in the wrong position (not the flat side of the nema motor) so that was probably the culprit. While I was that far, I just put the new motor in anyway.

[quinnkyle]

I researched these diy CNCs a long time before purchasing one. It's entirely made of wood - MDF and a wood base, and about $20k cheaper than leading brands. The thing I liked is that I could have a 5' x 10' table for larger jobs, so we have a big CNC for about $7k delivered (in unmarked parts). I will say, as all DIY machines, you learn a WHOLE lot about CNCs. I like that I can build another, now that I have a fairly solid knowledge of it, having built it from the ground up.

Previously we owned a ShopBot, and while it's a great CNC the downside - in addition to high cost - is that during a break-down you wait for weeks for replacement parts. With a DIY machine, most things can be purchased in town at a hardware store or radio electronics store. (I'm mourning the loss of Radio Shack nationwide!) I've been able to find limit switches, e-stop buttons and wiring easily.

That said, in my particular case I've had problems each step of the way with accuracy, and even have had some parts cut out of square last week. I'm told this hasn't happened before to other customers who purchase the kit, and I know of another company in Des Moines which cuts items requiring an even tighter tolerance than we need, and he hasn't had the same issues.

Since we cut a large number of wood parts, the machine requires frequent calibration and mechanical checking about 3x a week. For example, in regard to cutting out of square (our most recent problem) I suspect at this point there's some play in the spindle, so Monday I'll begin taking everything apart to check it out.

So I'd say if you don't mind doing your own research via message boards and youtube rather than calling a customer service hotline, and you enjoy investigative problem solving, a DIY CNC can be a rewarding thing to build.

Here's a look at our machine:

[quinnkyle]

Yes - the g-code said one thing and the Mach3 readout said something else - like the actual position was compromised. I put a new motor in, but believe this was caused by a slipping set screw that was not positioned on the flat side. Since it would slide on the round part of the motor, that's why we were getting inconsistent results each time.

[neilw20]

If the g-code says one thing and the readout, another, the steps are too coarse. You need micro-stepping drivers, or the micro-stepping needs to be set to a higher divide ratio.
Micro-stepping may not be the answer if the torque loads on the motor(s) is(are) high. A stepper motor between steps is just like a spring.
Look at the torque/speed graphs for the motor(s). Calibrate all you like. That won't fix a mechanical problem.
Mechanically, if you expect fine resolution, one step of the motor needs to be much smaller than the desired resolution.
Driving a belt? use smaller drive pulley.
Driving a screw, direct? Use a finer pitch screw or have a belt (no backlash!!) reduction.
Harmonic drive type speed reducers will give good results. Gears have backlash, unless compensated for with a spring loaded idler gear arrangement.
Spring loaded? Then the torque must not unseat the spring loaded parts.
You still have a mechanical/design problem.

[quinnkyle]

I currently have NEMA 34 Stepping Motors, and generally run about 18,500 rpm at a feed rate of 115 for the rough cut and 120 for the finish cut with a 1/4" 2 fluted compression bit. Generally cutting 2/4" birch plywood most of the time. Should I be shopping for different motors?

[neilw20]

How many steps per mm (or inches)? That is where I suspect your problem is, and acceleration is VERY important.

[quinnkyle]

I'll check tomorrow and get back to you. The curious thing is that Friday it was cutting out of square. I can't figure out how it's doing that! Thinking the spindle has some play in it.