[Daniel Wilcox]

I have a hardinge HNC converted to run on EMC2. I kept the stock servos, amps, power supply, and resolvers. I am using a pico systems resolver to encoder converter for both axi.
The lathe was working great for a while, and eventually i started getting joint following errors for the x axis. I noticed it was only in one direction (x-), and when it did it, the movement was pretty rough. If i went x+ for a second and went back to x-, it was usually ok. I slowed down the velocity so i could still run parts and it helped for a while, but now it's to the point where it's almost unrunable. Even at like 10ipm, it'll still fault. It does not appear to be anything mechanical.. if i turn the leadscrew by hand, it works great. Also, i tried changing the belt and it did not help. I do not think it is the resolver or the pico board (i tried swapping the pico board with the z axis and it didnt follow the board). The machine is still holding a good tollerance, and it doesnt run away. I'm thinking either amp or servo. Since it's only in one direction, i'm wondering if maybe it's a brush? (brushed DC). Has anyone had this problem and figured out what it was? I know i can swap amps and see if it follows the amp, but i figured i'd ask first. I'm going to attempt to go pull the brushes to check them out.
Thanks in advance,
Dan

[Daniel Wilcox]

well, i pulled the brushes out, and they looked good. However, there was some black grease built up on them, so i wiped them off. It was only on one edge of each brush, so maybe it was enough to catch the next pole and make it run rough in that direction. Also, i noticed one of the brush caps wasnt tight. I put it back together, and it seemed rough in both directions and faulted in the same direction a couple of times, then it smoothed out. I cranked it up to 200ipm (the max) and i ran it back and forth a ton of times and it seems smooth and hasnt faulted. I'm wondering if the first faults were because i may not have put the brushes back in the same way they came out and they just needed some revs to seat properly, or maybe there was still some of that grease in there and it needed to make a few revs to wipe it off. Does any of this seem plausable?
I'm not going to call it fixed untill i run it for a day or two without a fault. If it is indeed a definite improvement like it seems, then at least i have a good place to start. I was tapped out at like 40ipm before and it was still faulting alot.

[Daniel Wilcox]

well i spoke too soon. it went to crap again right after i went out there after typing this.

[mike_Kilroy]

hmmm... no other takers so I'll give u my 2 cents worth... dc motor may mean u have a tachometer in the motor too. tach brushes are more apt to cause probs in 1 dir like u describe..... can u get to it and try carefully to clean its brushes/comm? what is brand and part no of motor and drive? from ur comments I would not think the drive is the issue but more likely the motor and more likely even the tach. let us know.

[Daniel Wilcox]

hmmm... no other takers so I'll give u my 2 cents worth... dc motor may mean u have a tachometer in the motor too. tach brushes are more apt to cause probs in 1 dir like u describe..... can u get to it and try carefully to clean its brushes/comm? what is brand and part no of motor and drive? from ur comments I would not think the drive is the issue but more likely the motor and more likely even the tach. let us know.

hmmm didnt even think about that! Thanks.

The motors are Electro Craft. Not sure on the tach, but it may be difficult to take apart. It's a pretty small unit, and kinda looks like a little rc car motor. I'll take a look at it though.

BTW, this is a pretty low mileage machine. The controller died 10+ years ago from what i heard, and before that, it was only used for prototyping. The ways are PERFECT, and the paint is still in pretty damn good shape. Basically what i'm getting at is i doubt it would be the motor, but i could definitely buy it being a bad tach.

[Al_The_Man]

If you have the original drives and they are using tach input then they could most likely be Westamp or Glentek, these often suffered component change with age, one item was the 10 turn pots for tuning.
I usually ended up fitting Advanced Motion drives, they can be run in the Torque mode and eliminate the tach feedback altogether.
Generally you also get tighter control.
They are usually plentiful on ebay and the BLDC models can be used just as well with DC brushed motors.
Al.

[Daniel Wilcox]

If you have the original drives and they are using tach input then they could most likely be Westamp or Glentek, these often suffered component change with age, one item was the 10 turn pots for tuning.
I usually ended up fitting Advanced Motion drives, they can be run in the Torque mode and eliminate the tach feedback altogether.
Generally you also get tighter control.
They are usually plentiful on ebay and the BLDC models can be used just as well with DC brushed motors.
Al.

the original amps are GE, but the tachs are nameless.

I was alwas told that velocity mode was best for EMC2, is this false?

I took the tach apart and cleaned the brushes and comm, but everythnig looked good. working on putting it back together now.

[Al_The_Man]

The drive manuf. I use such as AMC and controller such as Galil Motion recommend the torque mode in CNC usage, I found found quite an improvement when switching over from the old velocity mode drives.
Did you try swapping the drives?
Al.

[Daniel Wilcox]

The drive manuf. I use such as AMC and controller such as Galil Motion recommend the torque mode in CNC usage, I found found quite an improvement when switching over from the old velocity mode drives.
Did you try swapping the drives?
Al.

i havent swapped the drive yet, still trying to get the tach back in. I got it in, but now it throws a following error in both directions. I'm not sure if this is due to the fact that i cleaned the brushes or if it's because the tach is missaligned. I'll have to play with it a bit. At any rate, i cant afford to replace any drives or motors at this point.

[Daniel Wilcox]

well i played with the tuning a little, and it seems to be helping. I'm back up to 200ipm. I'm going to see if i can dial it in a little better and give her a run.

Could it be that it needed to be retuned simply because the brushes might have been dirty?

[Daniel Wilcox]

just ran a part, and it went smoothly. No problems whatsoever at max velocity. This is the first time in quite a few months that it has done that, and actually the first time in a couple weeks that it made it through without getting an error even at 30ipm.
I'm not calling it fixed until i run it for a week or two, but this is definitely progress.

The weird part is that it got ferrors in both directions after cleaning the tach, and it also started ocillating a little when stopped. Tuning got it smoothed out, but i find it weird that it needed it just from cleaning, and it didnt even look dirty in the first place.

Thanks for the help guys!

[Doctor Tsugami]

What you have is pwm drives (pulse with modulated) these require tuning sometimes as they age, like all of us. I have some manuals on sweo and west amp drives they date back to the late 80's so let me know if you would like any information , most drives use the same adjustments just located at different places on the boards and the set up proceedure should be very similar good luck

[Daniel Wilcox]

What you have is pwm drives (pulse with modulated) these require tuning sometimes as they age, like all of us. I have some manuals on sweo and west amp drives they date back to the late 80's so let me know if you would like any information , most drives use the same adjustments just located at different places on the boards and the set up proceedure should be very similar good luck

appreciate the offer! I think i have the manuals for my drives somewhere.

The weird thing was that it was only in one direction before, and i did try to re-tune it in the past. Sometimes it would move really rough, and sometimes it would move really smooth, which could even be seen on the scope in the software. Something was screwy somewhere, and it seems that the tach cleaning did the trick.
EMC2 tunes the servos via the software, which makes it nice and easy. I think i got it tuned pretty well at this point.. i've been running it all day at max velocity without an issue, and the following errors look great when i watch them on the scope. There seems to be no ocillation or anything.

[mike_Kilroy]

GREAT to hear you got it going!! Glad my little idea about checking the tach helped!

When you say old GE drives, sounds like the old GE PWM HIAC drives! Wow that's history! You might want to offer them to a local museum when you eventually replace them! Ha!

Anyway, the discussions of torque mode vs velocity mode are all good ideas but in the old machines like yours they seldom could handle the velocity loop in the CNC so you had to have velocity control drives. No matter what servo you use you still gotta have a position loop, a velocity loop, and a current loop, and if the position loop folks (CNC) can make the velocity loop internally too, then you are free to go torque loop only in the drive and get rid of the tach.

Only guess I'd make is that the renewed comm and tach brushes were not seated well after your cleaning until they wore in a tad? And that is why it was bad for a bit?

Good luck!

[Daniel Wilcox]

GREAT to hear you got it going!! Glad my little idea about checking the tach helped!

When you say old GE drives, sounds like the old GE PWM HIAC drives! Wow that's history! You might want to offer them to a local museum when you eventually replace them! Ha!

Anyway, the discussions of torque mode vs velocity mode are all good ideas but in the old machines like yours they seldom could handle the velocity loop in the CNC so you had to have velocity control drives. No matter what servo you use you still gotta have a position loop, a velocity loop, and a current loop, and if the position loop folks (CNC) can make the velocity loop internally too, then you are free to go torque loop only in the drive and get rid of the tach.

Only guess I'd make is that the renewed comm and tach brushes were not seated well after your cleaning until they wore in a tad? And that is why it was bad for a bit?

Good luck!

Yep! they're the old GE hiak amps. It had a big ol' GE controller on it before i converted it to EMC2. It had solid core wires all through the control pannel, so i dont think it worked for much of it's life. There were so many broken joints in there it was rediculous. I think the machine was circa 1978, serial number 416. Berfore i was born!
As long as it's working, i'm not planning on changing anything. 200ipm is plenty for this tiny little lathe, it would be scary to move much faster. My mill, on the other hand...

As for cleaning the tach, it seemed to stay about the same after i got it back in, the only difference was that it was bad in both directions. It never got much better or worse until i re-tuned it. The changes in the tuning were fairly substantual, but it has been running consistently good since i did it. I guess just cleaning the tach was enough to make that much of a difference, even though it didnt seem very dirty at all. Now my x and z axis tunes are pretty different, where they were exactly the same before. I wonder if the z axis will need cleaning soon. I wont touch it till it gives me trouble though.

[Al_The_Man]

One thing that often gets neglected are the main brushes, until one wears down to the spring, then it usually means a new motor, brushes are cheap insurance.
Also after several years of use, there is a gradual carbon build up which is not healthy unless taken care of.
If you have changed the controller then torque mode drives at a later date is still a possibility.
Al.

[Daniel Wilcox]

One thing that often gets neglected are the main brushes, until one wears down to the spring, then it usually means a new motor, brushes are cheap insurance.
Also after several years of use, there is a gradual carbon build up which is not healthy unless taken care of.
If you have changed the controller then torque mode drives at a later date is still a possibility.
Al.

i checked the brushes on the motor first... they were still like new. I dont think this thing got much use.


I ran the crap out of it since i fixed it, and she's still plugging along just fine!

[gridley51]

When you say old GE drives, sounds like the old GE PWM HIAC drives! Wow that's history! You might want to offer them to a local museum when you eventually replace them! Ha!

I have loads of Hiak drives running on machines and so do thousands of other people.They are a very reliable drive but in the event of a fault are also easily repaired.I would take a Hiak anyday over some of the more modern drives.
Off course maybe Hiacs are rubbish,I`ve never came across these.

[Daniel Wilcox]

I have loads of Hiak drives running on machines and so do thousands of other people.They are a very reliable drive but in the event of a fault are also easily repaired.I would take a Hiak anyday over some of the more modern drives.
Off course maybe Hiacs are rubbish,I`ve never came across these.

yea, they didnt seem overly complicated. I'm definitely not going to change them out for the sake of doing so.

[mike_Kilroy]

I have loads of Hiak drives running..................They are a very reliable drive.......
Off course maybe Hiacs are rubbish,I`ve never came across these.

Sorry for being a typical engineer who can't spell! I surely did not want to make you mad! I never said the Hiak or Hiac was rubbish, just that it belongs in a museum as a piece of Americana. I stand by that remark. As one who designs and applies servos, I am very sensitive to how well they hold up in the field. I have many even older than Hiac(k) servos running yet today by customers who likewise swear by them.