Allen Bradley Ultra 3000 E22

[SpencerFayer]

I have an issue where the drive is not running, and I am getting this error. I have tried to slow the drive down through the interface on the machine. when I do slow it down, I still get the error. is there a way to plug the drive into a PC to get the settings off to slow down the drive or the accelleration?

The drive is not running at any point where I get the error, that is why I want to see if I can bypass the interface and use a PC to hook to it.

Thank you in advance.

 

[Ken Roach]

For those of you reading along at home, "E22" is the error code for "Motor Filter Error".

That's sort of a funny name for what this does; it's actually a "modeled overtemperature trip". The servomotors don't have a physical temperature sensor so instead the drive has a model of the cooling characteristics of each specific motor and measures the current and duty cycle of the motor.

The drive also uses this information to calculate a Motor Temperature parameter.

Unless this system is brand new and there's a configuration problem (the wrong motor might be specified), this error probably actually indicates an overheated motor, which suggests an actual mechanical change to the system that is causing too much current to be consumed to keep up the machine performance.

Ultraware is the software you want; you can connect to the serial port on the drive. If these happen to be connected via fiber optics to a ControlLogix, you could use RSLogix 5000 software to examine the motor configuration and monitor its temperature and speed.

I can't imagine a drive getting an E22 fault during a time when it is not running and has not been running for some period of time. Maybe when it's not moving; servo motors can overheat if they need to hold position against force for a long while.

Can you post more information about the context of this drive in the automation system and the history of the problem ?

 

[kolyur]

Ken, thanks for your usual thorough explanation. I too was getting this error recently on an Ultra 3000. The thing that made troubleshooting tricky was that it did not fault immediately when the overload situation occurred, as one might expect. Rather, it tripped out a short time later after the drive was disabled. Had me thinking it was some kind of feedback or interference issue.

In my case, it turned out that the relay controlling the motor brake had failed, so the motor was trying to run with the brake engaged.

 

[Amebeo]

Ken or Spencer,

Was there ever any resolution to this issue? I am having possibly a similar problem with a drive. I get an e22 on a Ultra 3000 while the light curtain is broken. What makes this seem impossible to me is that the motor leads are broken by a relay when the light curtain is broken. There can be no current draw from the motor when the light curtain is broken. I tried monitoring Motor Temp on UltraWare. At rest the motor temp was at 1% then I broke the light curtain. It took approximately 14 seconds for the drive to fault.

I believe we will be requesting tech support from AB today. If I learn anything I will be sure to post it if anyone is interested. Additionally, if you have any information that could be useful, it would be much appreciated.

Thank You
Amebeo

 

[dmargineau]

Have you tried replacing the drive's Output "isolation" relay?
The motor is physically disconnected, however, the LINE side of the "isolation" relay could be very well "draining" enough current to induce the fault you are experiencing.
This is quite an unusual Ultra 3000 application; at the moment the light curtain opens up the servo-drive's load side circuit from the motor (even when the system is at stand-still)there is a considerable amount of interrupted current which, over time, will "flash-over" the relay's isolation properties.
Just a question, how does the "live" Ultra 3000 "react" when is is disconnected from its "load"?

 

[Amebeo]

Hello dmargineau,

In response to the first question, yes we have replaced the relay. We thought perhaps this could be the issue as well, but this did not make any difference. Note the drive will fault without ever reenergizing the relay.

This drive is set up the same as another identical drive with identical motor with identical perameters. That drive does not fault at all. It runs exactly as expected.

In response to the second question, when the light curtain is broken the motor temperature trends up very quickly. If I keep the relay deenergized, the temp will continue to trend up. If I allow the relay to reenergize, the temp will trend back down slowly.

Is this what you are asking?

 

[dmargineau]

The reason for my question is the fact that your "safety" implementation is unusual for an Ultra 3000 Servo Drive application.
Do you have any other similar Servo-Drive applications in which you are opening-up the Load side (Output) circuit of a powered-up and "Enabled" Ultra 3000 drive? Are these applications functioning properly?
While installing Drive Output relays/contactors on VFD devices is a very common practice, the commands for the "isolation" device are to be given only when the drive is not running (except for Emergency situations of course); "removing the electrical load" when the drive is running will invariably and immediately fault it.
Back to the Ultra 3000, I believe your case is a similar issue; the Ultra sevo-drive is always "running" when powered-up and "Enabled" (even if it doesn't execute any mechanical "work") hence it needs an "electrical load"; removing the "load" via isolation relay could generate the issue you are experiencing.

 

[Amebeo]

dmargineau,

Thank you for the help. Further background on this piece of equipment. It was originally built by an outside source. We have since repurposed it for a new process. In this process we have redesigned, to some extent, the machine. A fair amount of this was done by an employee who is no longer with us. That being said, we do not have other drives set up like this one.

I believe what you are saying is that the source of our problem is not dropping the enable, but dropping the servo power while under holding load. Good call! AB agrees with you. We have gotten a response from technical support. I will give a synopsis below.

The basics of the response indicated that it was probable that we were not dropping our 24 VDC enable signal before dropping servo power. This means that the motor does still have a load on it when disconnected as dmargineau suggested. Additionally, with the drive still enabled, the servo can drift off it's encoder position and the drive will add power to try and increase command current to force the motor back into position. The drive does not know the motor is disconnected and continues to calculate the "Motor Temperature." This causes the E22 fault in our case.

Solution: We wired the enable voltage through the light curtain relays. This seems to have solved the issue.

Still unexplained, however, is how Motor 6 (identical to motor 4) does NOT have the same motor temp trends as motor 4. Instead Motor 6 runs exactly as expected with no symptoms like Motor 4. I would still like to know how this is possible, but for now at least this solution seems to be the best.

Again, thank you for your assistance

 

[dmargineau]

Well, this is a new one...RA TechSupport agreeing with my presumptions?!...
Did you ask them about the "identical" motors different drive feedback responses within the same given conditions?...Are the motors manufactured by Allen-Bradley by any chance?...
Some motor manufacturers have a lot "tighter" specifications and Quality Control than others; A-B is not one of the top tier ones; getting a lot better though...
That being said, it is almost impossible to get perfectly identical "two of a kind", especially when "talking" electrical equipment; expect "different reactions" and always try "tunning" the application depending of the installed harware "particularities" especially when precision is a must.

 

[OkiePC]

You always want to run you servo enable though the line or load side contactor with an early break contact if possible. Ideally, you turn off this enable signal long before the safety contactor opens anyway, but in cases where that isn't possible the early break contact should prevent this type of fault.

 

[Amebeo]

Dmargineau,
The two motors were mentioned I believe, but I don't recall for sure I did not have the pleasure of presenting the problem to AB. AB did not choose to comment on it if it were mentioned. I agree no two scenarios can be exactly the same. These drive were all tuned. The one in question was tuned multiple times hoping for different results. I'm not sure where the insanity delineation with that. In this process precision is not important.
P.S. I've found it's ok to be right from time to time. I just try not to make a habit of it lest humility finds me too in need of a reality check

OkiePC,

Part of the reason I mention the additional background was for the reason you suggested. Originally there was a time delay auxiliary block added to the motor control relay. I removed it not understanding its purpose. I believe the enable use to drop when the light curtain broke, then delayed the motor relay. Some of the original wiring may have been changed. Now we drop the enable and I will be adding the time delay controls back to the motor relays Monday. This is my first experience with servo controls, and a very good learning experience. Unfortunately most if the work was done when I got involved. At any rate, good observation. That is exactly what I was overlooking.

 

[OkiePC]

...Originally there was a time delay auxiliary block added to the motor control relay. I removed it not understanding its purpose. I believe the enable use to drop when the light curtain broke, then delayed the motor relay.

This is very common. Unless the servo is poorly sized or the controller has failed, nothing can bring the load to a stop faster than the servo amp. In that case, you would delay both the enable and the opening of the contactor with a safety relay off delay contact, and short the reference or give it a rapid stop command through software during this delay, drop out the enable via control output a very short time later (1/2 the safety relay delay is a good starting point) and then let the contactor open typically one to three seconds later, so no matter what goes wrong, it will coast when the contactor opens, hopefully after having already ramped to the fastest safe controlled stop possible.