Need Help to Improve Surface Finish

[TomS]

If the diodes are blowing themselves off the board they are severely overstressed! The important parameters are the diode type (speed rating), breakdown voltage and current handling capability. I found a teardown log here: http://blog.bouni.de/blog/2015/02/03/taking-a-dq860ma-stepper-driver-apart/ that provides a link to the diode data sheet. The diodes are "ultra-fast" type with 2 amp/50 amp average/peak, 280/400V RMS/peak ratings. The key to finding a US replacement is the speed -- they MUST be fast-recovery for a PWM application. Of course, you want the current and voltage ratings to be equal to or better than the OEM.

BTW from the same teardown log I note there are some indications that the driver manufacturer may be using counterfeit chips -- the sanded-off surface of the controller IC either says it's counterfeit or the manufacturer did not want their board to be reverse-engineered. If it IS counterfeit, they also may be using diodes that were pulled from other boards. If so, they may or may not be fast-recovery. That could explain the early failure of the diodes.

Thanks for the info. Not inclined to try a repair as the cost for a replacement driver is only about $40.

 

[TomS]

I ran a multitude of tests this morning. The first test was mounting the DTI as jbolt suggested and running gcode. The DTI needle bounced around quite a bit. I read on another forum that you can run a Fanuc post processor with Mach3. So I tried it. No surface finish change using the standard PP settings. I then set the Fanuc PP Arc Center mode to Absolte and made the same change in Mach3 General Configuration. Again, no change. Made the same change to the Mach3 post processor with no change in surface finish quality. Out of desperation I changed Arc Output from Normal to Convert to Lines. Nada! All these tests were run with Arc Fit Tolerance set at .0004 and Spline to Polyline Tolerance set at .0001. The last test were test cuts made at three different Z heights to see if the faceting lined up. They didn't. See the pictures below.

Still have to run the drain wire disconnected test as jbolt suggested. If that doesn't show anything then I'm leaning towards a motor tuning issue.

 

[homebrewed]

Thanks for the info. Not inclined to try a repair as the cost for a replacement driver is only about $40.

Sometimes it's the down-time that is more important than waiting for a replacement driver. If the driver is that unreliable maybe you should get one or two backups? Depends on whether or not the down-time delays delivery to a customer vs. delaying a personal project.

I suppose your decision to replace or repair also depends on your ability to solder/de-solder surface mount parts without damaging the board or other components. I've done that at work but now, lacking a good stereo-zoom microscope, SMT would be a bit more challenging for these old eyes. I thought I might be able to use a cheap USB microscope but the frame rate is so low it's really difficult to place small components. This is a case of you really DO get what you pay for.

 

[TomS]

Sometimes it's the down-time that is more important than waiting for a replacement driver. If the driver is that unreliable maybe you should get one or two backups? Depends on whether or not the down-time delays delivery to a customer vs. delaying a personal project.

I suppose your decision to replace or repair also depends on your ability to solder/de-solder surface mount parts without damaging the board or other components. I've done that at work but now, lacking a good stereo-zoom microscope, SMT would be a bit more challenging for these old eyes. I thought I might be able to use a cheap USB microscope but the frame rate is so low it's really difficult to place small components. This is a case of you really DO get what you pay for.

I hope I didn't come across as flippant. My mill has been running for about three years now so if I have to replace a driver again in three years I'm OK with that. This driver failure was a first for me. Everything else has been running fine.

I'm with you on soldering small electronic components. My eyes aren't what they used to be and I don't have the proper equipment. I'd rather spend my money on tooling for the mill and lathe.

 

[jbolt]

My drivers are the ones that kept blowing up diodes, not the OP's. I believe this is the first driver failure for Tom and I have no idea what the failure is. I was just sharing my experience and why I moved away from them. As surface mount components go they are fairly large and easy to replace with a fine tipped variable heat soldering iron. I fixed mine a few time since it only takes a little time to do vs waiting. I'm not suggesting he abandon them but if it is a diode that failed to be aware.

Tom - if you exhaust all options with you current setup you might try reinstalling the old BOB and motion controller. Not fun but it might tell you if it is something in the 126, 107, ESS combo.

Grasping at straws here but I have seen issues (non CNC) with some cheap Ethernet cables. I make sure to use high quality shielded cables, keep them as short as possible and away from power cords.

 

[TomS]

My drivers are the ones that kept blowing up diodes, not the OP's. I believe this is the first driver failure for Tom and I have no idea what the failure is. I was just sharing my experience and why I moved away from them. As surface mount components go they are fairly large and easy to replace with a fine tipped variable heat soldering iron. I fixed mine a few time since it only takes a little time to do vs waiting. I'm not suggesting he abandon them but if it is a diode that failed to be aware.

Tom - if you exhaust all options with you current setup you might try reinstalling the old BOB and motion controller. Not fun but it might tell you if it is something in the 126, 107, ESS combo.

Grasping at straws here but I have seen issues (non CNC) with some cheap Ethernet cables. I make sure to use high quality shielded cables, keep them as short as possible and away from power cords.

I understand your logic but I really don't want to reinstall the old breakout board and the UC100. I've changed so much that it would be nearly a complete rewire job. I've got a couple of new things to try first. I'm going to lower the power supply output voltage and see what happens. Read that on Practical Machinist where the lower voltage altered motor resonance and improved the surface finish. I read somewhere else that raising the voltage did the same thing. Who knows? It's an easy thing to do.

I did run the drains disconnected test you suggested. Surface finish didn't change so my assumption is the grounding system is sound.

You bring up a good point about the ethernet cable. It's Cat6 but that's all I know. I'll check into it and make sure it's shielded.

 

[jbolt]

I understand your logic but I really don't want to reinstall the old breakout board and the UC100. I've changed so much that it would be nearly a complete rewire job. I've got a couple of new things to try first. I'm going to lower the power supply output voltage and see what happens. Read that on Practical Machinist where the lower voltage altered motor resonance and improved the surface finish. I read somewhere else that raising the voltage did the same thing. Who knows? It's an easy thing to do.

I did run the drains disconnected test you suggested. Surface finish didn't change so my assumption is the grounding system is sound.

You bring up a good point about the ethernet cable. It's Cat6 but that's all I know. I'll check into it and make sure it's shielded.

I totally get not wanting reinstall but it might narrow down the cause of your issue.

When you ran the test as I suggested how much did the Z jump around? Does it jump only when doing arcs or both arcs and straight moves?

 

[TomS]

I totally get not wanting reinstall but it might narrow down the cause of your issue.

When you ran the test as I suggested how much did the Z jump around? Does it jump only when doing arcs or both arcs and straight moves?

I'm going from memory here but as I recall the needle jumped about .007" to .010" and only during arc moves. You can hear the motors change sound when cutting arcs. During straight moves they are much quieter.

I tried a test run today with the power supply set at 55 vdc and 65vdc. Didn't make a difference in surface finish. I rerouted my ethernet cable so it was no where near a AC power cable. Again, no change in surface finish. Using a 2 flute 3/8" end mill I ran a 1/4" DOC by .005" WOC at 20 IPM then ran the same scenario at 5 IPM. Something is causing the motors to jump around corners but I don't why. It's got to be a harmonics/resonance issue. Wish I knew what's causing it.

 

[spumco]

I'm pretty stumped, too.

Shot in the dark here... Do you have one power supply for all three axis? I'm wondering - you mentioned motor noise changing - if the PS is acting wonky when there's a load from both X/Y at the same time. 'Cause you also mentioned that it's a problem on straight angle cuts, not just arcs, right? Any time you're cutting with both axis moving there's a problem?

Maybe a test doing an X/Z or Y/Z move might result in the same surface finish issue. Do a quickie CAD/CAM of a hill-shaped object and use a bullnose cutter with a parallel 3D contour.

 

[TomS]

I'm pretty stumped, too.

Shot in the dark here... Do you have one power supply for all three axis? I'm wondering - you mentioned motor noise changing - if the PS is acting wonky when there's a load from both X/Y at the same time. 'Cause you also mentioned that it's a problem on straight angle cuts, not just arcs, right? Any time you're cutting with both axis moving there's a problem?

Maybe a test doing an X/Z or Y/Z move might result in the same surface finish issue. Do a quickie CAD/CAM of a hill-shaped object and use a bullnose cutter with a parallel 3D contour.

I have a dedicated power supply for each axis drive. Straight line cuts are fine. Surface finish is good. So yes, problem/poor surface finish is on arc and circular cuts.

I'll try a sloped 3D cut and see what happens.