NI PCI 7342 and UMI 7764

[K_seeker]

Hello All,

I am using the PCI-7342 to control 2 servos for co-ordinated motion. I am using UMI7764 to breakout Motion IO to connect to my Servo Amplifiers. I am confident that the system will work fine because I have used a similar setup before. The issue I have now is that the UMI 7764 unit is farther than 2meters from the PC. I called NI and the tech said that NI doesn't have any cables longer than 2M to solve my problem. So she directed me to this forum. I was little disappointed that NI doesn't have any option or may be she is just not aware of it.

I have following questions here. First, is it OK to have the UMI 7764 mounted farther than 2 meters from the PCI card? (I need one extra meter here) Has any one used a setup where he or she needed to mount the Break Out Board farther than 2 meters and if so than was there any issues with the signal degradation that was observed? Is there any limitation on distance mentioned any where on NI documentation ?

Secondly if it's Ok technically to be farther than 2 m than who is the cable manufacturer? I was starting to search L-com.com. Is there any other source besides L-com.

Thanks

[Jochen]

As you were already told NI doesn't recommend using longer cables for motion control applications. There are several reasons for that. Here are the most significant issues:

• The quality of analog control signals decreases with increasing cable length
• The UMI converts differential encoder signals to ground referenced signals which also limits the maximum cable length.

I just mention these things as I want you to understand why we don't offer longer cables. A cable length of 2 meters is ok even for environments with significant electromagnetical noise. I have been working with customers who used 5 m cables and the application still worked reliably but in this case they used quite small motors in a low-immission environment. 3 meters should be ok in many cases but we can't guarantee for that.

Unfortunately I don't know any 3rd party vendors but maybe someone else in this forum does? Please make sure that you use high quality cables with proper shielding.

Best regards,

Jochen

Message Edited by Jochen on 10-09-2006 10:05 AM

[K_seeker]

Jochen,

Thanks for your response. I had not realized that UMI is converting the Encoder signal from Diffl to SE. That may cause some issue in our case.I am wondering now how the Robot controllers handle it. I beleive most of them uses +/- 10V to controls various axii of the robot.

Thanks.

[karlito20]

Hello K_seeker,

You had devellopped an application to control a servomotor with a servo amplifier. For that, you are  using the PCI-7342 to control the servomotor and the UMI7764 to breakout Motion IO to connect to your Servo Amplifiers...

In my company, I have to devellop the same application on LabVIEW and I bought the PCI-7342 and the UMI7764... 

I would like to know what is the logiciel you used to devellop this application.,

If it's LabVIEW, i would like to know if you can send me the VI because I don't know LabvIew...

Thanks a lot...

Murat

[K_seeker]

Murat,

For most part you should be able to look at the help files and figure out how to control the motion. Basic motion control vis are same for stepper or servo. So there is no magic to it. the only thing that we have to figure out is how to connect the servo amplifiers with the UMI7764. If all that connectins are valid than next step in the process is to tune the motors from the MAX (you can skip this step if you are using steppers) . If you have not done tuning before than you may be a bit frustrated initially. At times its a bit of trial and error approach to get the perfect tuning parameters. You will find lot of resources on internet to give you a better understanding of this process. Once you have aconplished this you should be good to go. To start with you can lok at example files. That will give you a good feel of various motion functions available within LV.

I will go through my code and see if any of my code will help you but meantime give me more detail about you system.

GoodLuck

[karlito20]

I have a testing machine (which permits to stretch or to compress samples) that I have to analyze with an aim to re-do the processing part (command part).

I send to you a diagram of the machine and a diagram with the connection to the UMI 7764.

I have to do moving the horizontal part, obtained thanks to the rotate of the two long screws.

This screw turns thanks to the speed that supplies the motor (gearwheel + driving belt).

To control the motor, I have to apply to the terminals AOUT and AGND, the consign voltage (є [-10V; +10V]), image of the motor speed rotate.

+10V → 3000 rpm

    0V → 0 rpm

-10V  → -3000 rpm

Cordially

Murat

[Buechsenschuetz]

We are currently connecting the PCI7344 with 5m (custom made) cables to our system which converts differential to single-ended signals (similar to the UMI). We do not use the UMI box (except for certain tests) but we never had problems with 5m cables.

Anyhow, this always depends on pulse frequency, environment noise etc etc. In other words, it is hardly predictable whether an application will work with cables longer than the ones recommended by NI. In any case, you should do the differential to single ended conversion as close as possible to the 734x card - the differential signals were designed to avoid noise problems and you should use this benefit as far as possible.

[karlito20]

Hello K_seeker,

I sent you more detail of my system (look at the last message I sent)... Are there any similarity with your system???

Cordially.

Murat.

[karlito20]

K_seeker,

Can you help me??? I really need your help!

Thanks!!

[K_seeker]

Murat,

Sorry about the delay. The connection you have mode between these devices seem fine. (I just want to let you know that I have not used the brand of servo drive and amplifier you are using) Have you been able to move the motor from MAX. You should be able to move even prior to tuning. Start with low P values . Gradually introduce I and D. I can't tell you what these values could be. You will have to figure out your self. If you are able to move your motor from MAX I believe you are set to move on to Labview programming. Prior to that do not waste your time trying to start run it from labview.

Let me know how it goes with this step than we can take it to the next level. Good Luck.