Rockwell Custom Motor Files format, third party motor

WhiteBear

Member
Join Date
Aug 2013
Location
Earth
Posts
18
Hi,

Don't know if anyone had the same specific issue but still worth asking...

I'm trying to use a third PM rotary motor on a Kinetix 6000 (Sercos). It is very similard from a MPL-B3xx as it is fitted with a SRM50 or SKS36 with comparable characteristics. We are using Rslogix v16 to v20.

I managed to create a CMF file from one I managed to find and that imports successfully in the motion database.

Here's one attribute that is an big obstacle for integrating my drive : blobData.

Il seems it is an 512 Byte encoded value of the others attributes and it is the only data sent to the Kinetix drive (? - it seems to shrink to 256 Byte once stored in motion.db). By modifying any of the independant values, the import fail "corrupt file", proof it is encoded from the list of values provided. So far understanding the encoding method of the blobData has proven difficult. Rockwell won't provide any support and advised us to blindly swap any third party motor, which would cost way to much as well as bringing new mechanical issues, by the way...


Has anyone got any experience with creating or importing CMF files in the motion database ? I suppose infos on creating CMF files is very restricted.

I've also read that sick hiperface encoders need to be setup for use with third party motors by using a Kinetix 6000 and a specifig procedure, is it applicable in Sercos mode with Kinetix 6000 or only with Kinetix 6500 in CIP ?
 
(If somebody is interested...)

So far I managed to get the first 60 Bytes. They are the easy ones, ascii or hex attributes. Decimal values (eg : rated current, peak current) are all 11 decimal digits, so I tried in BCD, float and integer fixed but couldn't find any match...
 
I applaud your efforts. It would be a fantastic feature to have for niche motors or system control upgrade project.

On RA's report card the notes read "...is a good student, but doesn't play well with others." Remember you are dealing with a company that took an open standard and closed it. Your drives aren't SERCOS, they are SERCLOSE. RA promised us 3rd party drive support on SERCOS 10 years ago. Still waiting....
 
Rockwell do support third party motors, but for a few motor manufacturers only, not customers it seems. Quite a lot of motors in the motion database are third party motors included by rockwell. They made efforts with CIP motion axis with kinetix 6500 drives for example : third party motor setup from nameplate datasheet is possible... on paper.

Issues arise with intelligent encoders (Hiperface from Sick Stegmann or eidenhaim's EnDat ). Kinetix 6k are build to read motor settings from the encoder memory when a intelligent encoder is specified. If the structure is missing or corrupt, a encoder error E30 occurs.

But the same requirement exists with K6500 with nameplate datasheet configuration and the use of intelligent encoder ! The Rockwell Automation Publication 2094-RN020B-EN-P specifies in page 3 :

"Kinetix 6200 or Kinetix 6500 drives:
A one-time procedure must be executed via message instructions to program the blob file in the encoder (by using a Kinetix 6500 drive) so that it can be operated like any other Allen-Bradley motor. This is similar to the Stegmann encoder third-party motor requirement, except that a Kinetix 6500 drive is used instead of a Kinetix 6000 drive"

We want to setup a test motor in this configuration (EN2T+K6500+third party with intelligent encoder), does somebody know of the message instructions to use to perform this configuration ?
 
example CMF file

WhiteBear do you have an example CMF that you might be able to send me? I am doing some connection work and will let you know if I make any progress.
 
Hi guys,

I almost finished mapping datablob values (long and hard work to map and scale values by testing, logging, comparison), mesuring encoder angle and reprogramming the sincoder datafields accordingly. I'm running tests on my motor but so far motion is very erratic. Even marker and position test cannot be completed, a pole of the motor starts of finish by fleeing to the nearest winding. We switched motor, power phases, adjusted the encoder commutation angle and electrical values (R1 motor stator, XM mutual). But I still seem have issues with electrical tunning (?) since the motor pole usually move to the nearest winding when MSOed or when moved manually after MSO. I don't know if the flux is ok to lock the drive in position, or if a bad phasing or a bad encoder angle is making the motor going astray.

Is there a motor configuration specialist here ? I have the spécifications and all the Tools needed, but stuck with theses nasty problems.

I tuned the motor on a powerflex 700s with opt stegmann encoder card (needless to say my motor works very well on the powerflex 700s), but the values format read with the 1203-SSS on PF drives don't exactly match the one in a Kinetix 6000, tough I could check a few ones. I have doubts on XM motor stator in the kinetix, but it doesn't seem to make a important difference in my case... must have a bigger issue at hand.
 
I also have a few worries about back EMF (CEMF in Pflex). We never if it's V/KRPM or what, the value are rather hard to find.

For example, for AB MPL-B330P-M I couldn't find the V/KRPM value in datasheets. datablob value is 39,20. Ke VoltageConstant in motion database is 67,896. So it should be 39,20 V/1000 RPM.

I don't know if bad CEMF is the usual source of a motor run away.
 
Great progress was made, but there's still some issues related to commutation offset. I decoded and reversed the whole model, including CMF files and blob data. I worked more than 200 hours of the topic, over more than two years.

With an Elau Pac-drive SM (low inertia, PM sync motor), depending on some starting conditions, I can use it very well : stacking positionnal absolute commands, trend is good, no abnormal noise with good set of electromechanical parameters, jogging like a charm, precise positioning/lock tolerance is good, no error.

I had some trouble to get precise electromechanical values for the motor. Kinetix need precision, even more with PM sync low inertia motors. Just look how bad this type of motor runs at low RPM on an able powerflex drive...

I still have some issues with commutation offset on sick stegmann absolute hiperface encoders, hence my previous "starting conditions" from before. I'm working from the TI5000-EX documentation to understand how to read or define the absolute encoder offset and linked commutation offset.
 
An interresting thread.

We've had some custom motor files made by Rockwell for some Aerotech servo motors on k5000 drives and two of the four motor types trip occassionally on motor feedback errors. Rockwell are coming to investigate tomorrow.

Why did you choose to make the CMF yourself?

I dislike the fact that you cant just use a 3rd party motor intensely.

Nick
 
Rockwell was not willing to help us. They refuse to help you if they believe they have an equivalent motor available in their motion database. Same issue for the config tip on Kinetix 6500 with third party motor and hiperface or endat encoder.

So I had no other choice to do everything by myself.

I'm interested in your aerotech motor's CMF file or at least the datablob to check its values compared to datasheet specifications.
 
Last edited:
That's a bit poor isn't it? Even they didn't get to sell the motors, they at least sold the drives.

There is no Rockwell equivalent to motors we needed to use as they don't have any that work in a vacuum.

Nick
 
Yes that's a bit poor. In our case, the motor model had no torque and acceleration equivalent for its frame size in rockwell motion database.
 
Commutation offset issues are now history. I'm so nearing a complete reverse engineering.

Tests on a heavily loaded robot axis are next to come.
 
After many delays, low-load tests are fine, heavy load tests will be run next week on robot arms (10 Kg then more than 100 Kg loads on various reducer run by 3 to 6 kW motors)
 

Similar Topics

Hi!! I'm looking for Temperature rise calculation software from Rockwell, I just download "Product selection toolbox 2022" but this software is...
Replies
1
Views
97
Hi all. I'm building my first Grafcet using Logix, but I started from another project. From what we usually do with other PLC's, I was expecting...
Replies
1
Views
55
“The HMI files we cannot open—they were saved in V13—we do not have that—I cannot restore file –please have them save in V11 and send them back to...
Replies
4
Views
135
Working with AB/ Rockwell PLCs How do I get certain outputs to show up like the picture ? The one in the picture was made by a manufacturer, but...
Replies
4
Views
190
I want to program an Exor HMI using a Rockwell controller. However, when exporting the tag list from the controller and importing it into JMobile...
Replies
0
Views
92
Back
Top Bottom