And Mike I agree with you! Why does everything always need to be perfect?

Mike

That might be true, but never the less.
The Drives we opt to use, makes all the difference.

I used cheap Chinese Drives from Wantai motor, and those have a poor microstepping control I found out. My motors made heaps of resonance at low speeds.
Now after I mounted the Leadshine AM882 Drives, same configuration, same microstepping, and same 5% of maximum speed, the motors now have no resonance at all. That proves to me, that the quality of the drives do have something to say.

So it comes down to expense, how much do you wish to spend on getting a smoother driven motor on your CNC.

"Quality only costs the difference" is a Danish saying "I think", and it proves itself right this time atleast.

Kim,

Yes, it's always best to buy as much quality as possible.

I know nothing about your first set of drives or of the drives that you're using now; but, I have had the same experience with non-gecko vs gecko drives. The Geckos give much better performance than the older drives that I was using.

In my world of process control applications, the Gecko drives made everything easy and smooth. In the world of CNC routing, even the Gecko drives don't seem to be able to completely compensate for the stepper motor's characteristic resonance; but, because I have no other stepper drivers that work better than the Geckodrives, I would still use those drives if I built a MechMate, simply because I have those drives on hand.

My old drives were from longs-motor on ebay, the DM860A drives.
Apparrently they should be good, if you ask longs motor, and wantai motor, who are the manufacturer. But these drives are far from good, the microstepping or chopping in the drives are just horrible, they do make the motors move, but the motors resonated like hell.

The new drives are Leadshine AM882 Digital drives, which comes with a setup of 9 different types of motors from leadshine already programmed into it. If that's not good enough, then you can finetune the drives from the computer with a program that Leadshine applies free of charge.

I only hooked up the drives, and then set them for a Leadshine motor similar in size and frame as the ones I had. I powered up the system, and the motors ran ultra smooth and almost completely quiet. best buy ever. Easier to setup than the geckos, no need to fiddle with resistors for current control. All can be setup using dipswitches on the drive, both current setting, and misrostepping setting can be set using the dipswitches to... All in all, a good drive, at around the same price as a Gecko. They are somewhat larger than the geckos, but ever as good, and as far as I'm concerned, easier to setup...

http://www.leadshine.com/Product_Show.aspx?ID=44

This is the drive.

Here's a picture of the cut quality of my machine with only the new drives on the x axis. Notice that this is an eliptical small sign. I have only touched up the edges to remove the flossed sharp edges.

Hence my suggestion to use an encoder and measure the quality of a drive system.

Gerald,

I agree that running some tests with a good quality encoder and the other proper hardware and software devices would show whether or not the stepper driver was filtering out the resonance, but the tests would not change the way the cuts look.

I thought long and hard about building the things that you suggested, but when I put pencil to paper and added up the costs (minimum of $300 for parts) and the time (proabably 5 to 10 days), and then knowing that the cut quality would remain the same. I couldn't justify the expense or the time. That's the reason that I've written those posts about finding the limitations of the machine and then living with the limitations.

Only Gecko can change the resonance problem. Mariss has done an excellent job for the money that he charges. I talked with him on the telephone twice and visited his home in California once. He's as knowledgable a guy as I have ever met, but CNC machines are not the main market for his stepper drivers. He builds stepper drivers by the thousands for process control, where various axes move the same distance 24/7.

The least expensive test would be to put some Geckodrives on Kim's machine, tune the drives, make some cuts, and then repeat the process with the drives that he uses now.

If anyone has four G203v drives in Europe that they could send him for the test, it might help everyone decide which stepper driver to buy.

I don't notice much in the way of stepping on a 4:1 belt driven system (G540 and Keling 960 oz.in motors driven parallel) at any speed. I phoned Mariss years ago when I was building a Mechmate and he said to hook up a mirror to the shaft and map a laser pointer's position against a wall tracking the position of hte micro-steps.

The KFlop does stepper feedback and with 1000ppr encoders popping up on Ebay it's looking like a good option. The really juicy bit however is that you can compensate for the micro-stepping non-linearity of the steppers even with the active feedback present, but the holding torque of the stepper will still be strongest around the full step so for slow high load cutting or acceleration I could see why the position would deviate.

Hi Kim
I am very inspired with your builds.

I know it's been awhile since you posted this. But do you still offer this set?
Or have I read it all wrong.
Best
Mac

No, I do not supply these anymore.

They would probably be atleast twice the price if I were to order them again.