Lexium 05 Drive Assistance

Greetings all, I am having a little bit of difficulty with an application I am working on and the company I work has placed me a bit under the gun. So I am hoping I can post what I have here and get some assistance.

I am working with a lexium 05 servo drive and a M238 Telemechanique PLC. They two are connected via CanBus currently talking at 250MB. I am using the function block style of programming. Now granted it has been very very long since I was in college and it has been equally long since I have used some of the tuning and such required in drives. Most of my knowlege and skills has been in PLC's. But anyway I need to make this servo start and stop on a dime, very precise, very fast, very accurate.

I have been tweaking and tuning for a couple of days now and I currently have the thing running at 4000RPM and its accuracy is about +/- 1/32. I need to get it dead on each and every time.

So has anyone worked with these products and can supply an example of how they tuned the drive and PLC so that it was very accurate?

Also what would the tuning have to be changed at all if the RPM were adjusted? I ask that because one I get the thing commissioned there may need to be times when the RPM is adjusted because of the quality of paper or the print or something. If you wish I can email my program as I have it thus far. Just PM me. And it is something with codesys but the name of the software is SoMachine. And Lexium use to be Berger Lahr. Hope this answers all the questions. Thanks and have a great day.

Hi mrtweaver,
I'd like to help but I need more information about your application and what you are trying to achieve.
Lets forget for the minute that you have an M238 and Lexium05.
I can tell you in detail how to tune the drive and I'm well trained on the SoMachine software but I need to understand exactly what it is that you are trying to achieve.
Chris

Chris, good day, thanks for your reply, what I am attempting to do is to replace an older set of controls on a guilotene form cutting machine. What the machine is designed to do is to take a roll or paper and cut the sheets to the given length. The main reason for redoing the controls is the older style controls that are currently on these machines are limited in what they can do, and more and more clients that we service are requesting cuts that can not be done on the existing machines. Cuts such as 3 cuts at 5 inches, then a strip cut at 3/8 inch. Or 2 cuts at 7 inches and a strip cut at 1/8 inch. Or the really unique one 1 cut at 4.5 inch, 1 cut at 6 inch, 2 cuts at 11 inch, and no strip cut. The existing machinery can only do 1 cut and 1 strip in any given cycle. So as you can see the cuts have to be very very accurate. As I said I currently have it tuned where I am +/- 1/16 but I need to get it right every time. I know I am close but just not close enough. I should also mention that in the testing I have done, I can cut 10 pieces, out of those 10, 6 are usually right on, the others are 2 that are over by 1/16 and 2 that are under by 1/16. It appears that when it goes over it self compensates on the next cut and goes under. But this is not always the case. I hope this clears up things.

chris_in_dav said:

Hi mrtweaver,
I'd like to help but I need more information about your application and what you are trying to achieve.
Lets forget for the minute that you have an M238 and Lexium05.
I can tell you in detail how to tune the drive and I'm well trained on the SoMachine software but I need to understand exactly what it is that you are trying to achieve.
Chris

Click to expand...

Hi,

OK.... That's not a massive help.
What is the motor driving,? Is there some kind of feedback of the paper position like an incremental encoder for example or an optical sensor detecting the leading edge and if so where is that connected to????? I need to understand the basic architecture of the mechanics and electronics.

The origional servo motor was an ABB it was a low inertia. It was direct drive. The new motor the Lexium is thru a 4:1 gear box to help with inertia. The item that is driven is a pin feed assy. The pin feed assy is all very lightweight aluminum and plastic, it has a distance of 6 inches for every revolution of the drive shaft. Of course the pin feeds drive the paper. The only thing that is being used for distance purposes is the encoder which is on the motor and is being fed back to the servo amp. The OEM did the same thing with the ABB they led the encoder from their motor back to their proprietary controls.

The following are on the system, M238 PLC, Lexium 05 motor and servo, 4:1 gearbox. The system goes as follows: a motion profile is made in the PLC this is the distance and velocity, it is sent to the Lexium via CanOpen at 250MB speed, the drive then performs the motion, when it has completed the motion it notifies the PLC, the PLC then triggers an output which causes the knife to fire, there is a sensor on the knife, when the knife has retracted to home the PLC then sends the next motion profile to the drive.

The cutter is off of a clutch and motor. Motor is constant spinning. Cluch also has sensor on it, the sensor is N/C. So the command to fire the knife monitors this sensor, when the sensor has a trailing edge it shuts off the output. When this same sensor has a rising edge it means the knife has returned home and now can move to next motion profile.

The current programmed of PPR of the encoder is 32768 which is the default value. In the lexium you can specify the denominator of the encoder so that if you want 500 pulses to equal 1 rev you can set that. But default is 32768 which is where I left it pending my trial and testing.

The over all purpose of the machines is very simple, cut forms accurately to given length. SO I think they would call it a length and cut machine. I did look inside a cabinet of a newer machine that does a similar function and they used Panasonic Servos, its to bad there is only one distributor here in USA that handles Panasonic otherwise I might just scrap the idea of using the Tele products and go to the Panasonic.

chris_in_dav said:

Hi,

OK.... That's not a massive help.
What is the motor driving,? Is there some kind of feedback of the paper position like an incremental encoder for example or an optical sensor detecting the leading edge and if so where is that connected to????? I need to understand the basic architecture of the mechanics and electronics.

Click to expand...

OK, all is a little clearer now It sounds like what you are trying to achieve is quite simple.

I guess you are using either MC_MoveRelative or MC_MoveAbsolute and then waiting for the done output of this function block to trigger the cutting mechanism? If you're not sure that the paper is stopping in the correct position then add a couple more checks before triggering the cutting. For example you could read back the actual position and compare it to where you expect it to be or you could check for the standstill output of MC_ReadStatus. It's possible to alter the parameters of the standstill monitoring in the Lexium05 drive.

If the code you produce is robust and the mechanics are good then producing a repeatable acurate cut should be easy regardless of drive tuning as the drive is stationary whilst cutting, correct? Drive tuning becomes an issue when you need to move from A to B quickly. I've attached a link to a PowerPoint presentation on manually tuning Lexium32 drives but it's just as relavent to the older Lexium05. Good luck.

http://cid-b4a7ca58f4db57dc.office.live.com/embedicon.aspx/Public/Tuning Lexium32.ppt