Is this normal for Gecko Drives?

[kevinkoons]

Hello Everyone!


When tuning the axis motors using the trim pot on the Gecko drive, the motor will suddenly stop when the trim pot is turned near the end of travel in the clockwise direction. When this happens the only way to get the motor to run again is to remove power from the gecko drive (power down the machine) and exit out of Mach 3, restart Mach 3 and then power the machine back up. Some times it is also necessary to reboot the computer in order to get the motor to function again.

I am also having this same problem with the motors suddenly stopping when running a program when the feed rate is 10 IPM or less. Running at 20 IPM for example, I can cut all day long with no issues at all!


I am running Mach 3, Version 2. I am using Gecko 212 drives. This is a Bridgeport retrofit Series 1 CNC machine with Sigma motors on the X and Y and a Superior on the Z. All axis motors have presented the same problem at one time or another.

When running a program I wrote specifically to trouble shoot the problem, the X Axis motor will quit at randomly. It has ran for as long as 4 hours and for as little as 10 minutes, maybe less.

I am at my whits end, just can't seem to figure this out. I know I am not the first person to use these old Bridgeport steppers, has anyone had a problem like this or something like this? Any ideas what it might be?

I configured the Gecko drive for NEMA 42 motors, and 10-micro step resolution. I have also tried 5-micro step, half step and full step, all with the same results.


Any help on this would be very much appreciated!


Kevin

[mike40272]

Hi,
I've had the same problem on 3 different Bridgeport series 1 Mills. The only
way I found to keep them running was to use a fan blowing directly on
them. The Gecko's (all 12) felt warm to the touch not Hot. I soon as I installed a fan they would run all day long. The steppers were sigma's
and superior's running at 5 amps 52 Volts. All three bridgeports are running
fine now. I am converting my own Bridgeport Series 1 right now. I will be
using Gecko 212's and Mach3 software.
Mike Carter

[mike40272]

Hi again,
one other thing, I never had to touch the trim pot. All the steppers ran fine
with the pot in the factory setting.
Later
Mike Carter

[kevinkoons]

Hi Mike, thanks for the reply!

I thought about heat but I ruled it out because the only time the problem occurs is when I am running at slow feed rates. Not only that, but the problem sill sometimes present in just a few minutes or after as long as 4 hours! One time it happened on the Z axis which saw very little action in that particular program.

I will give a fan a try, what I got to lose at this point? Not sure that this is the problem however and would appreciate any other suggestions.


Regards,

Kevin

[kevinkoons]

Hey Mike, how did you configure your drives? Are you set for full, half, 5-us or 10-us stepping?

Thanks,

Kevin

[mike40272]

Hi,
I hope the fan helps, on the drives I have them set for micro stepping (5). I tried Half step but they seem to run smoother using micro stepping. If I can
think of anything I give you a shout. Have you posted this question on the
Bridgeport fourm? That fourm has lots of real experts, not kidding , they
really know their stuff.
Mike

[Mariss Freimanis]

There is a possible reason for your experience assuming you are running the drive wide open (7A current set).

1) Depending on the trimpot adjustment, the trimpot adds a 0% to 20% of set current value to the motor phase current. This is to compensate or null an offset error introduced by your V / L (your supply voltage and your motor inductance).

2) The G202 / G202 short-circuit protect level is 8A. 20% of 7A is 1.4 Amps. 7A + 1.4A totals 8.4A which trips the drive's protection circuit latch. This latch is cleared (reset) by either recycling the power supply or momentarily hittting the DISABLE input.

3) The ADJUST full CW setting of 20% is to accomodate low inductance motors and have them run smoothly. Without that +20% offset they'd be at 5.6A at high supply voltages and run roughly.

Your Bridgeport motors are not low-inductance so in effect you are setting them to 8.4A with that trimpot wide open. The drive then trips at the 8A protection threshold and you get what you observe.

4) Setting ADJUST too high makes your Bridgeport motors run like crap as well below 10 IPM (if 1:1 or ungeared), never mind the protection trip problem. Here's how to set the ADJUST trimpot correctly:

4a) Return the ADJUST trimpot to the 11 o'clock position (at little CCW of center).

4b) Cause your machine to move at 5 IPM.

4c) Turn the ADJUST trimpot no more than +/- 1/8th of a turn CW/CCW. Listen to the motor. There will be a setting in that span where motor vibration dissapears. Tune it in back and forth a smidge a few times. Leave the ADJUST trimpot at that setting that nulls the vibration best. The drive won't ever trip the protection circuit again on its own when you do this.

4d) Don't want to bother with any of this? Just set the ADJUST trimpot to a little CCW of half-scale and call it a day. No more nuisance protection circuit trips that way either.

Mariss

[kreutz]

Originally Posted by Mariss Freimanis There is a possible reason for your experience assuming you are running the drive wide open (7A current set). 1) Depending on the trimpot adjustment, the trimpot adds a 0% to 20% of set current value to the motor phase current. This is to compensate or null an offset error introduced by your V / L (your supply voltage and your motor inductance)................ Mariss

Mariss;

Is this compensation similar to "slope compensation" used on current mode control?

Thanks,

Kreutz

[Mariss Freimanis]

No or maybe if I misunderstand your question.

Most switching drives depend on a reference voltage. Once the motor current sense reaches that reference, the drive reverses the direction of current change and the current revereses. The result is the motor current oscillates between the peak value set by the reference level as a peak and some value below that.

The motor responds only to the avarage value (0.5 peak to peak subtracted from the peak value). This introduces a negative offset equal to 0.5 peak to peak value. This in turn results in a crossover distortion equal to that value. Our means of compensating for this error is to introduce a constant positive adjustable offset via the ADJUST trimpot to null this V/L error. The result is a motor phase current that passes thru zero without any discontinuity.

Kreutz. You are a very smart guy. I have followed your posts and I am impressed with your thinking. Why don't you PM or e-mail me at geckohall@cox.net so I can meet you. We are not at cross-purposes at all. You are doing unpolar drives and I would like to take the oppertunity to talk you out of that.:-)

Mariss

[kreutz]

Originally Posted by Mariss Freimanis No or maybe if I misunderstand your question. Most switching drives depend on a reference voltage. Once the motor current sense reaches that reference, the drive reverses the direction of current change and the current revereses. The result is the motor current oscillates between the peak value set by the reference level as a peak and some value below that. The motor responds only to the average value (0.5 peak to peak subtracted from the peak value). This introduces a negative offset equal to 0.5 peak to peak value. This in turn results in a crossover distortion equal to that value. Our means of compensating for this error is to introduce a constant positive adjustable offset via the ADJUST trimpot to null this V/L error. The result is a motor phase current that passes thru zero without any discontinuity. Kreutz. You are a very smart guy. I have followed your posts and I am impressed with your thinking. Why don't you PM or e-mail me at geckohall@cox.net so I can meet you. We are not at cross-purposes at all. You are doing unpolar drives and I would like to take the oppertunity to talk you out of that.:-) Mariss

Well, your circuit might not be the same but it works in a similar way and for the same purpose the "slope compensation" works, basically, true current mode conversion should make the inductor current follow an error voltage, so the inductor becomes effectively a voltage controlled current source. By resistively adding a voltage (fraction of the PWM oscillator ramp) to the sensed current, it is possible to convert a peak current detector into an average current detector, allowing for a better current mode control. This system is used extensively on switching power supplies.

Thanks.

[kevinkoons]

Hello Mariss,

Thank you for your advise! When I first did the retrofit on this machine and began tuning the motors, I hadn't touched the trim pots at all. When everything was working as it should I began to further fine tune the machine. There was some vibration on the X & Y axis so I began fussing with the trim pots. When I realized that going too far clock-wise cause the motor to stop (and sound bad) I turned the pot back to the left a tad and left it. I think the problem was I had it too far CW.

I have taken your advise Mariss and set them to about 11 o-clock. From there I really didn't need to adjust anymore. Actually on the Y axis for some reason the motor causes a lot of vibration at 5 ipm and the trim pot will not adjust it out so I picked a "happy place" and left it for now. At this point I just want to know that the machine will run with out stopping!

I'm testing the machine now and will let everyone know how it turns out!


Thanks to all that responded, I appreciate it very much!!

Kevin

[kevinkoons]

Hello everyone,

After adjusting the trim pot on my 212 drives as suggested by Mariss I am still having a problem with the motors stopping or tripping the drive protection circuit.

Next I turned the trim pot all the way CCW (0%) and ran the machine, same thing after a little while the one of the axis motors will quit.

Last night I ran the machine (cutting air) for six hours. The X and Z axis motors both quit but the Y did not. In all of my attempts at solving this problem I had changed the micro step setting. The Z is set for 10 micro step and the X was set for full step. The Y on the other hand was (is) set to 5 micro step.

At one point I had all motors set for 10 micro step and ALL motors would quit at random when running at less than 20 imp.

Mike said he is running at 5 micro step so I'm giving this a shot in conjunction with a fan blowing on the drives and all drives have the trim pot full CCW or 0% position.

I'm still looking for answers at this point....


Update: After running almost 5 hours, same "sample" code I ran last night, everything seems to be working well! I have the drives set to 5 micro step and a fan blowing on the drive units.

It would seem the fan has helped, sorry to have doubted you Mike, thanks for the tip!!

Still doing some testing and will let you know how it all turns out.

Anyone using the Bridgeport steppers with Mach? What settings are you using on X & Y in motor tuning?

Thanks,

Kevin