Ultra3000 ILimit strange behaviour

Does anyone have any experience with the Ultra3000 using the Ilimit, we have an application where we are replacing a 1398 with an Ultra and are having issues with the Ultra when the Ilimit ref is at 0 volts we get full current instead of 0 current. Can't seem to find any info regarding this issue so any help or insight would be appreciated.

I assume your Ilimit reference voltage is on pin 24 of CN1 and common on pin 22?

You assume correctly

Try viewing the Analog Limit Input in the Ultraware Monitor window to verify that it is seeing your reference voltage.

(Go to Monitor pane and click Setup, turn on "Current-Analog Limit Input" under Current Signals)

Been there everything appears normal as long as I am above 0 volts anything from about 50mV up is fine but below that it is where things begin to fall apart.

I would expect that there has to be some voltage applied at all times in order for the current limiting to work, simply because current increases proportionally from 0 to 10 volts, yet if you leave the input disconnected you'll get full current. But I don't see any mention of this in the manual. Can't you set your minimum reference voltage to be 50 mV or more?

That is an option I've considered and may do, at this point I'm just looking for info or insight because as you say the manual suggests I should have 0 current at 0V provided its not open. The 1398 behaved as per the manual however it seems the 3000 does not.

The issue is far from insurmountable however it would be best if I didn't have to start adding offsets that will obfuscate the controls for the troubleshooters

Understandable, sorry I can't be of further help. I've worked with the 3000's quite a bit but have never used the current limit.

I have worked with the predecessors to the 3000 as servo-only mode slaves being commanded from S-Class motion controllers. We always used torque mode in those older drives, even for positioning applications. The motion controller issued torque commands every 0.5ms to close the position loop.

This did not work well in velocity mode. It was only in torque mode with torque limits fixed first in both directions, save it to the "personality module" and then you could get the full range of performance and expected behavior as in the older 1391 servo drives.

I think you can configure a torque mode override, I have not used the 3000 as a servo amp slave, only in indexing applications, they are quite flexible, though, I recall default settings for reverse torque and velocity limits needed attention in the configuration. We ran the Ultra 100 and 200 as servo slaves to S-Class motion controllers.

I may be mistaken, but I think that if you're Ultra drive is confiured as a velocity controller, it will have full torque/current available at all times and at all speeds, but unloaded, the amp draw would indicate poor tuning or "hot" tuning for a demanding application.

You are correct in Analog Velocity Control it has full torque, but using the ILimit analog input provides current limitation which works except when you get below about 50mV then it gets full torque again which is not how it is supposed to behave and not how the Ultra 100 its to replace behaves.

Kick-It-Up said:

- - -
using the ILimit analog input provides current limitation which works except when you get below about 50mV then it gets full torque again

Click to expand...

Then I would assume noise can be ruled out since the old one works fine and you simply have a drive with a failed analog input. Is there another analog input which can be configured as ILimit?

I've got 4 different 3000's demonstrating the same behaviour so I'm ruling out bad hardware, and no the ILimit is a dedicated input. Also this has been demonstrated on the bench and in the field so noise should not be a factor.

At this point I'm quite confident this is an errata, so I'm gonna throw it back to Rockwell and see what they come back with.

I could understand if the error signal was a little "out of cal." at the zero, and perhaps you had to set a software limit to cancel a minor error, but switching to full torque in 4 different drives screams firmware bug.