advice choosing servo amplifiers. Brushed DC motors, +-10V signal input

Some years ago I retrofitted a Bridgeport R2E3 (BOSS 8) 3 axis Mill with Linuxcnc. I used the popular 5i25+7i77 Mesa cards and kept the original Servo motors, encoders and drives because they were all working. Machine has been great and has already produced a lot of nice parts.

Yesterday one of the servo drives (Contraves NC400) failed. I swapped it with another axis to confirm that the drive was at fault (it was). Can't see anything visibly wrong with it, but I don't have an electronics background nor the knowledge to set about trying to repair it.

I'm considering replacing all three drives with something more modern. I won't be too sad to see them go because they sequel horribly during operation, and I wonder how much life the other two have left in them. Hoping the forum can make some suggestions about what my options are.

The motors are Brushed DC Servos (SEM 30M4-24). They have encoders and tachos built in. The encoders are fairly low resolution which is annoying because the DRO in linuxCNC is constantly flickering +- 0.003mm but they work so I'm inclined to leave them alone.
- constant stall torque 3Nm
- 96VDC
- 4000 rpm
- Pulse amps 75

The mesa 7i77 has been reading the encoders and controlling the drives with +-10V signal.

Q1. Does anyone have an old Contraves NC 414 they no longer want? (long shot but worth a try)

Q2. Can anyone recommend any cheapish (hopefully less than 200 Euro each) servo drives which would suit the SEM motors, and can receive a +-10V signal from the 7i77? I'm based in the UK so something available in Europe would be best.

Q3. Should I be considering replacing the 7i77 with a 7i76, then using some step + direction servo drivers, like the DG4S-16035 for example?

Many thanks in advance,
Tom

Thanks for replying.

Just had a look at the pico pwm servo driver (pico-systems.com/osc2.5/catalog/links/pwmservo.html).

How do you go about tuning a drive like this?

The description mentions 20A peak current. The manual for the original contraves NC400 quotes 45A peak, and the motor has "Pulse amps 75" stamped on it (I'm not sure yet what this actually means). Am I likely to lose some performance using the Pico drive? loss of acceleration for example?

Thanks!

Tom_R2E3 wrote:

Thanks for replying.

Just had a look at the pico pwm servo driver (pico-systems.com/osc2.5/catalog/links/pwmservo.html).

How do you go about tuning a drive like this?

The description mentions 20A peak current. The manual for the original contraves NC400 quotes 45A peak, and the motor has "Pulse amps 75" stamped on it (I'm not sure yet what this actually means). Am I likely to lose some performance using the Pico drive? loss of acceleration for example?

Thanks!

OK, the problem with the Pico servo amps is they are NOT a drop-in replacement for your velocity servo amps.
If you JUST want to replace the amps, take a look at Copely, AMC, Westamp and Servo Dynamics, to just name a few.
Yes, you would lose some acceleration with the Pico amps, as well as having to change out more of the system.
You ought to look for a servo amp with at least 40 A peak rating to get the same performance.
Your SEM motors have DC tachs in them, so you want to get real velocity servo amps that accept a tach feedback signal.

Jon

Hi Jon, thanks for the info and suggestions. I didn't realise it would be such an expensive exercise to replace the drives. For something with 35-40A rating i'm looking at £800+ for the three, probably more like £1000

There are some old NC400 drives listed on ebay with varying prices, maybe that's a better option for this machine. Don't want to spend too much at the moment, because I plan to upgrade to a better machine when I can.

Thanks, Tom

Or maybe there is a chance to repair the drive that I have.

There are two resistors on the top circuit board which look as though they have been hot. They are labeled 5W 2.5k Ohm but when I measure them I get 3.15 kOhm and 2.08 kOhm, could this be the problem?

I've attached a couple of photos. One of the legs on the transistor mounted behind them also looks suspicious.

Measuring resistances of installed components can be tricky, because readings from surrounding items will screw up your readings. If you had reading even remotely close on a resistor like that they are likely ok. Resistors usually die by going open. But you should probably re-flow those solder joints, and any others that look like they may have gotten hot.

Tom_R2E3 wrote:

Hi Jon, thanks for the info and suggestions. I didn't realise it would be such an expensive exercise to replace the drives. For something with 35-40A rating i'm looking at £800+ for the three, probably more like £1000

There are some old NC400 drives listed on ebay with varying prices, maybe that's a better option for this machine. Don't want to spend too much at the moment, because I plan to upgrade to a better machine when I can.

Thanks, Tom

I would NEVER use the NC400's due to the HORRID whine. I go to great lengths to protect my ears from damage.

There is an outfit that makes drop-in replacements for later model Bridgeport servo amps. Unfortunately, I am not going in to work much right now, and can't access the company name. They very well might have a replacement for your servo amps, too.

Otherwise, you can browse eBay and look for AMC and similar servo amps. They still have all their old manuals online, so you can look up the specs and wiring info for those models. Yes, high-power servo amps are expensive.

Jon

Tom_R2E3 wrote:

Or maybe there is a chance to repair the drive that I have.

There are two resistors on the top circuit board which look as though they have been hot. They are labeled 5W 2.5k Ohm but when I measure them I get 3.15 kOhm and 2.08 kOhm, could this be the problem?

I've attached a couple of photos. One of the legs on the transistor mounted behind them also looks suspicious.

You might resolder the leads on these resistors, sometimes after long periods of running hot, the solder "crystallizes" and you get bad contacts. But, those resistors don't look like they got hot enough to go bad. Most likely one of the power transistors failed. And, that may have taken other parts with it.

Jon