[mc-motorsports]

Bridgeprort retrofit, Mach 3 controller, 20a 48v transformer, 3 servo's and 3 G320's.

I was cutting out a broken tap using a carbide endmill, I set the jog rate to .7ipm because I was trying to save the part, more or less just trying to grind out the tap and save the threads.

Well, the servo tripped, I reset it, tried again, it tripped again, I reset it again, one more time and then NOTHING... The Z axis won't move. I can reset the servo's and the X and Y work fine, but the Z won't move.

The other wierd thing is when I power up the power supply, the servo's automatically engage???

I'm 99% sure it's the drive because if I disconnect the encodor, the servo does not try to run away, there is still no movement. And with the servo and encodor disconnected, the other two drives engage and disengage fine.

So did I blow the servo drive running too much amperage on a servo at too slow of a speed? I've been using the servo's and drives with absolutely no problem for 18 months. Or is that just the life expectancy of the servo drive? I use the machine atleast 40 hours a week every week, no problems.

And I did try the Z axis servo and encodor on the X axis drive and the quill works fine.

Was it a freak thing? Did I move the servo too slow at max amperage? Or was it just the life expectancy?

I did order 2 new drives, one to replace the bad drive and another for a back-up from Gecko Drive, ordered at the last minute and they still over-night shipped it! Have to love thier customer service!

MC

[LUCKY13]

First guess would be yes, to much mojo going on at to slow a speed. Thats really loading things doing that with a servo system, you add the heavy weight of the head and it gets worse.


Of course there is always the chance of it just being a freak thing that the drive blow. ALso I would keep a eye on that servo because if it has a internal short thats trying to start acting up you might have problems again.


One thing I have always thought everyone should do is run temp reading equipment on the drives so you can see if things are getting in a strain. The drive will get hot if so. ALso some are using amp meters to help watch things.


All you can do is chaulk it up to gone for now and keep a eye on things. Be sure you use good thermal paste on the back side of the drive and good air flow/cooling for them. At 48v you would think you would not be pushing the drives that hard but its the amps that can get you the most and on a Bridgeport that can load things pretty quick. I plan to put temp reading equipment on each of my drives and maybe even amp meters. I am a little shy on the amp meters because of some info I was reading that was trying to claim it being a bad thing to put them in line but I feel like one with a good heavy shunt would be no problem. It would be pretty easy to build a little system that keep a eye on the temp with a PIC and some thermisters. It could even be setup to give warning signal or possibly even cause a fault or E-Stop if done right.


Make sure the wiring didn't take a hit and now has a lot of resistance built up in it. This could be worse than a bad motor. Connectors and power supply itself should be looked at close. Hopefully it was just a freak or a overload. The overload is something you see people do every now and then so I would guess that being the case. The servos like to run in the RPM not at low load. In fact any quick load that stops a servo is a bad thing and should be avoided.


Jess

[mc-motorsports]

Back up and running! I ordered the drives at 6pm EST and they were here at 12pm EST the next day, up and running again by 1pm. Thanks Gecko!

[mc-motorsports]

NEVERMIND!!! I just blew another drive!!!

how do I figure out if it's the servo or the encodor?

Please help!

[keebler303]

It sounds to me like you are overloading the drive too quickly and blowing the MOSFETs.

Mariss has said that the drive will die if you drive the machine into a brick wall. If you hit a hard stop, the drive does not have time to fold back the current to protect the power stage because the current rises to stall current (maybe 100A or more) almost instantaneously. A carbide endmill into a tap is pretty much a brick wall. It survives the first few times (when drive faults) then it finally gives up.

I would say .7ipm is too fast. Try something like .1 or as slow as mach will go. If I were you I would use the knee crank to advance it and leave the quill where it is. Also if you could find some way to lock the quill so the servo wasn't taking the load it would help. Go nice and easy.

Matt

[mc-motorsports]

Yes, I've done this many times, but I usually use the knee to feed the endmill, why I used the CNC to drive the quill, I don't know... Never thought it would be a problem, guess I got lazy last time.

But the problem is not the drive, I just blew another drive and I wasn't doing anything dumb this time, running the same programs I run every day and the new drive failed within 20 minutes. Telling me either the servo or the encodor was damaged or went bad. I just ordered another servo and encodor, I don't know which is bad, I'll just replace them both.

MC

[keebler303]

I don't think any problem with the encoder would kill the drive. Unless it somehow pumped excess voltage into the logic or its shorting the encoder supply.

Good Luck
Matt

[mc-motorsports]

I know almost nothing about DC brush servo's. The old servo measures 1.5 ohm static, no load, no RPM, and resistance goes up and changes with RPM, spinning by hand. The new servo does the same thing, so I'm gussing the servo is ok?

Pole to ground is infinate on a 20m scale, I only have a cheap VOM, I don't have access to a surge tester any more, anything else I can check? Probably not with only a VOM?

I'm replacing the drive, the servo and the encodor, hoping to solve the problem, there is no binding in the z axis, moves freely turning the ballscrew by hand. But I still don't have a definate answer on where the problem is or was...

[mc-motorsports]

I GIVE UP! I just blew another drive.... Works fine on the Y axis drive, works fine on the X axis drive, just swapped the encodor and power cables. I just burned up 2 brand new drives and I have no clue why. I checked all of the wiring, made sure every terminal was tight. I don't know what to do...................

[TOTALLYRC]

The other wierd thing is when I power up the power supply, the servo's automatically engage???
MC

Well if this is a change from the way it used to work, then it could be the culprit.

From your last post, it all works well on the x and y, but not the z axis.
I would check for continuity form the z motor wire to ground and each other when shaking the wires. sounds like you have a short between the 2 power wires or to ground. both of these will blow a drive almost imediately.
It would only take one stand of copper wire inside a connector or something stupid like that to pop the drive.

Mike

I would also consider talking directly with gecko for more suggestions before you blow anything else.

[mc-motorsports]

I can just unplug the power cable and encodor cable and plug it into the X or Y axis and it works fine. And yes, I did check the wiring, a visual of the outside looking for breaks or anything, and also checked it with a VOM before wiring in the new servo, everthing checked fine.

It's not the wiring, it's not the servo, it's not the encodor.......

BOB?

I don't know...

I'm ready to **** can everything and start over with brand new everything. I'm losing a lot of money and it's hard to just keep buying new parts when you have to turn away work.

[mc-motorsports]

One other thing that is weird and I don't understand. If the servo loses position, the drives don't fault??? I've had a couple of instances in the past where I did something dumb and plowed a tool into a part, and the servo would fault immediatly. I still have a feed handle on one side of the x axis, if you fight the servo and turn the handle, it does not fault out??????????

What the hell happened? Everything was working fine and now everything went to hell??? Dead serious this machine runs 40+ hours a week, has been for 18 months without 1 single problem, then the other day, everything just went to ****???????

I REALLY don't get it.

ANY help or advice is appreciated!

[TOTALLYRC]

I feel you frustration as I had problems with my setup when putting it together, but it has been working fine for a while now.

Borrowed from the gecko website,
here is how to see what is wrong with your blown drives usng an ohm meter.

"To check your drive using a digital multimeter (DMM), follow these steps:
Turn off power to your drive.
Set the DMM to Ohms and put the negative lead on terminal 1 on the drive.
If the drive is a stepper, put the positive lead on terminals 3, 4, 5, and 6. If any of these shows
0, there is a blown MOSFET. For a servo, do the same test but only put the positive across 3
and 4.
Now take the negative and put it on terminal 2 on the drive. If it is a stepper, put the positive on
pins 3, 4, 5, and 6 and follow the same rules as above. If it is a servo, only test pins 2 to 3 and 2
to 4.
If there is a blown MOSFET, the drive must be sent back to us for evaluation. If there is more
than one blown MOSFET, then it is not repairable."

If the mosfets are blown, then it is a short circut or overload somewhere. If the mosfets are not poped, then it could be related to your no need to power up reset the drives and you have bob or control input problems.

Hope this helps,
Mike

[mc-motorsports]

what would cause the working drives to not fault when the encodor loses position? That makes no sense to me what so ever. That's why I'm doubting it's the BOB or control input, the encodor feedback is looped to the driver only, and when I had the orignal failure, I was only using the Z axis...??? And now the Z axis is the one that I've blown 2 brand new drives on, even with a new servo and encodor.

Can 1 bad drive cause the other 2 to fail? Should I have just replaced all 3 drives? Did the 2 drives, X and Y axis go bad when I blew the Z, and are now blowing the new drives?

I know one thing, I'll be drinking a lot of beer for the next 2 days. Can't do anything until monday, may as well take advantage of it

[TOTALLYRC]

:cheers::cheers:Lots of beer and a few days off is always a good way to regroup and see it in a new light.:cheers::cheers:

Unless the three drives are tied together somehow, only one drive will fault out when its encoder is out of position. If you have all the error/reset terminals tied together, so that one encoderer out of position causes them to all fault, it could be a short or higher than expected voltage within that wiring.

Mike

:cheers:Cheers:cheers:

P.S. Now that I think about it, I will have a beer right now, since it after my normal lunch time and I am thirsty. Thanks for the suggestion.

[mc-motorsports]

:cheers:

All 3 drives are tied together so if 1 faults, they all fault. And they all run off of a common power supply. I'm no electrical engineer, but I can take down a 12 pack with the best of them :cheers:

Here's to better days! :cheers:

[TOTALLYRC]

Since they are all tied together so that when one faults, they all fault, and you no longer have to reset the drives upon power up as you appearently once did , I would consider disconnecting the wires from the err/res terminal and temporarilly use a push button switch to do the drive reset. This will at the very least eliminate one of the problems that you are having, before it pops any of the other drives. IIRC voltage much above 5volts on the control terminals will cook the drives.

Do you have a schematic of the drive reset circuit? It could be a bad component in it or depending on how it is wired, a bad circuit on the BOB that it is connected to. Assuming that you have it wired to estop the machine also.

:cheers:The beer is cold here in CT:cheers:

I can only have one as I will be running the machine soon and don't want to feed the bloodthirsty cutters with my own flesh.

Mike

[mc-motorsports]

Hell if I know how they are wired I bought a system from "low cost cnc", and I have nothing good to say about the guy who sold it to me... Long story... We'll just say after he takes payment, he doesn't care, and another $700 later, I fixed his BS and had everything running great since August 07. He originally sent me the drive unit with encodors which drew 3 times the amperage that Gecko drives can handle, 150ma I think? They were Renco encodors, drives failed on start up. 3 new drives and 3 new encodors later, everything was great, after a $200+ phone call to Australia for 45 minutes (lesson learned) that got me nowhere... I ended up calling Gecko and had everything running within days, they told me what was wrong, told me where to get the right encodors, really helped me out, everything was great until the other day...

What exactly happened?? I don't know. What is it going to take to get up and running again? I don't know that either. Right now I'm screwed. I'm just going to thank the beer gods :cheers: and maybe pray to the porcelin gods if I don't slow down I guess I have time for a hang-over though so it's ok!

[keebler303]

It sounds to me like the logic side of your wiring has a short or something. I think the only way the drive won't fault is if it is damaged. It would have to be in such a way that the control logic didn't get fried but the fault logic did. Have you tested the blown drives to see if it is the power stage or the logic that is blowing? I would look real close at ALL your wiring and be sure there is no abrasion on any wires, wires getting pinched in a cover, anything at all like that.

Definitely put in a call to Mariss on monday. Chances are he can narrow down the problem based on the symptoms you have.

Good Luck
Matt

[mc-motorsports]

I've attached a pic of how the drivers are wired for the engage/disengage and error resolution.

For some reason, it seems when 1 drive fails, the all fail, I mean BURN UP.

If something is wrong, please point it out.

And the 5v input for step/direction is daisy chained from the BOB, thru the X, then the Y, then the Z... The Z axis is what originally failed... Is that a clue?

Does it make sense that if one drive get's fried or mosfets burned, or catastrophic failure, the other 2 would be destroyed also?

I did not wire this, this is the way that I bought it from Logitrol, or "low cost cnc retrofit".

Please help,
MC