Servo drive failures

rsdoran

Lifetime Supporting Member
Join Date
Apr 2002
Location
Birmingham, AL
Posts
7,371
I know this isnt a plc question but so many of you are familiar with this drive and/or systems like it and these kind of problems I had to ask here.

Ok the this drive is on a feeder that feeds flat metal into a stamping press, we are on drive 5. BTW it is an Indramat DCK 21.3 servo drive.

The company that built the press and feeder came in and installed a new drive, line reactor, capacitor module, motor and new cables. The new drive was not configured so they used the old programming module and DI card at first but we kept having faults. They determined from the previous problems that the 24vdc had noise and that was causing the failures, faults and loss of communication. The put a pull down resistor across the 24vdc power supply but still had faults. HERE is the part I think is wierd...as I said originally they DID NOT use the NEW programming module (actually its more of a fault indicator/memory module) and the DI (digital input card). They did put in the new Programming module but still had faults.

THe OEM tech decided that the dry contacts from the relays needed MOV's so drove downtown and got some. While he was gone we isolated some of the 120vac items that go thru the relay contacts and still had a fault. While he was gone the Indramat tech changed the DI module and entered the variables then the OEM tech returned and put the MOV's in...ACROSS THE Normally open contacts of the relays that passed the 120vac to the ESTOP, air solenoid and oiler solenoid...NOTE: Across the normally open contacts...NOT the relay coil.

The company has a Fluke 97 50MHZ scopemeter. I used but didnt see any noise etc at first but I was looking more for harmonics. We had another fault Friday morning so we investigated REAL HARD. The 24vdc power going to the drive is clean with very little ripple, maybe 50 mv and no noticeable harmonics. The output power from the drive has an AC signal of 6 volts with a timebase of 50ns....ie 20MHZ if my math is correct. We isolated every device we could that gets power from the drive or the 24vdc supply but the signal stays there. We removed the 480vac incoming power to the drive and enabled the drive and this signal is gone.

The drive has a carrier wave of 4khz, standard. WHY/HOW could I be getting a signal, 6vac@50ns, like this from the drives power output that is just suppose to be an internal jumper from the incoming supply?

The old drives that failed were sent to Indramat and the reports for the first 2 were basically that the programming module had to be reloaded with the firmware. We know the last one physically burned a DI card. All these are symptoms of overvoltage/spikes/surges on the 24vdc. The faults were unique because there was a combo of E1 E3, watchdog timer pertaining to high voltage discharge etc. At first the Indramat people said it couldnt have an E1 E3 fault.

Got any ideas what I am seeing...is it something that I need to ignore or fear?

All thoughts are welcome.
 
Ron

The only thing I can help, in some servo systems which I installed in the past I had problems like that.
first we separated the DC power supply of the servo from the rest of the system.
secound I used capacitors on the servo inputs to avoid noises.
I useally installed the servo in another panel.
It nasty problems which is very hard to find.
You can also try to feed the servo from another power source if it posiabel.

good luck
 
What do you mean the output power?

I work with Indramat a lot, and have worked with the DKC's. Do you mean the pass through of the 24VDC power or the actual output to the motor?

It is a pass through, but the drive also is connected to it.
 
Isolation

From your description I assume that there is currently no isolation transformer ahead of the servo drive. I suggest that you put one in, 480 VAC PRI / 480 VAC secondary. Ground the neutral on the secondary
side (needs to be WYE / Star).
 
fault

Wich fault appear on the digital module

F409... etc..??

when you connect your laptop on, wich is the fault description?

Have you read the help file??
 
To answer some of the question.
1. Output did mean the pass thru that is internally jumpered.
2. We have added an isolation transformer.
3. The main fault that I have seen is E1 E3, at same time, which are both watchdog timers. I have also seen the SDR overtemp fault..F292 I think it is.

Update:
I figured out I may have misstated something. The signal I am seeing is around 6vac (10vac if my probe was too close to the drive), very spiky, with a timebase of 50ns per division, there are 4 divisions per cycle. I believe this is a 4mhz signal...or so I thought.

I spent most of the afternoon taking readings and checking this system inside and out. With the drive enabled this "noise" will appear. I was seeing some wierd things...ie signal between ground and common ...which are bonded, very strange.

Anyway during all this testing/checking and a major headache I noticed I could get readings off the drive casing to the 24vdc common..which didnt make sense. I checked and the drive casing wasnt grounded...I grounded it, still had "noise" but had decreased in amplitude. I then ran another 24vdc power line to the output side (which is internally jumpered thru the drive and has connections for other uses internally) and the signal decreased more.

The strange part was I had nothing on the 24vdc power supply all thru this, after adding the ground and the second connection I now have...I THINK...around 400mv on ALL the 24vdc. I am hoping this is low enough that it wont be an issue.

I also noticed using the dual trace on the scope that the 24vdc did have a very small ripple and the AC coming out of the drive seemed to be a multiple of it...ie you could see the pulses/spikes happening to both at the same time.

I am not positive about what I have done so will monitor the system for at least a week or 2 to verify it stays low. I may add a DC snubber but will talk to the OEM about it first.

Any thoughts are highly appreciated.
 
Ron

Do you have the 400mV when the servo power in ON,Please check when the power is OFF.I would like to know if the servo generate the 400mV.
 
NO "noise" when drive not enabled.

The press encoder, BTV interface, transition box and EMD have not been changed BUT all but the EMD has been isolated from the drive and no "noise" found. An isolation transformer, line reactor, and capacitor bank have been added.

The drive seems to be creating the noise...but its at such a high frequency I dont understand how/why. We have even disconnected the incoming 480vac from the drive and did not get the "noise".

I will continue monitoring...wish my RS232 interface worked with Flukeview and the Fluke 97. Then I could log the signal and display it.

Appreciate the assistance.
 
What kind of cable wires are u using in the output? Maybe chnge the path of the output in the panel. What is close to the drive?

Try to keep the conntrols wires as far as possible from not close, paralel or twisted with power. Looks like the output wires are the dirty ones.

I dont really understand a MOV on contacts, maybe it make sense one end on contacts and the other on common but both ends there to dissipate the arc??? I dont understand.

Just a thougt, hope it helps
 
important

one thing that very important to do it's power on the 24 vdc before
input the 480v.

If you don't do that, you can burned the bladder of the drive. Be sure then the 24vdc is on before put 480v. Same thing when you disconnect the 24vdc, the 480v must be disconnect before.
 
I appreciate the fact that my IP is posted...thank you.
NOTE: its bellsouth.net

It should never be forgotten that I am a wanker...a fact I was recently made aware of. It seems its wrong to be a jack of all trades and master of none.

What else needs to be made public? My address?
1044 Birmingham Street, Tarrant AL 35217

Phone number? 205 841-5237 is home
Cell is 205 276-3357 but at this time I do not have a cell phone.

I am the soul owner of www.patchn.com

I am married and have 5 kids.

I graduated from Southern Technical College in Birmingham, AL with 4.0 gpa in Electronics.

I drive a 1980 Chevy 3/4 ton PU

Are there more details anyone cares to know? Or post?

I almost forgot...my birthday is Friday May 7, I will be 51 and my full name is Ronald Scott Doran. I have the nickname "Luke".

vipersig.png
 
Actually your IP and all that stuff is only visible to you. Each person sees their own info. I have no idea what it is.

http://www.danasoft.com/

You are correct that there is no such thing a privacy anymore. Anyone here should be able find my phone # and a map to my house within 60 seconds.
 
Last edited:
Rick Densing said:
Actually your IP and all that stuff is only visible to you. Each person sees their own info. I have no idea what it is.

NOW you do... :(

Ron, do yourself a favor and have Phil remove that stuff... :nodi:

beerchug

-Eric
 

Similar Topics

I am looking for some suggestions on good Ethernet/IP servo drives to communicate with an AB PLC. My customer has a non motion PLC but wants to...
Replies
8
Views
1,143
I am looking to upgrade some of our old Servo Drives to the newer kinetix 5700 style. currently we have 4 1394 axis that are all driven by 5kw...
Replies
1
Views
846
Hello there! I try to to drive a HF-KN13 servo motor with a FX5U PLC through MR-JE-10C. I tested the connection between the servo amplifier and...
Replies
1
Views
644
Dear all, we have a couple of SSD 637 servo drives that need to communicate trough easyrider. On the parker website you can download easyrider...
Replies
0
Views
500
Hi I selected a Mitsubishi Servo drive MR-J4-100TM-PNT for profinet communication. The supplier I am getting from this particular drive doesn't...
Replies
1
Views
1,380
Back
Top Bottom