Random DeviceNet Nodes dropping off Network

[MorphuisOGrady]

Guys,

Thanks very much for all your suggestions. I am going back to site tomorrow to investigate further. Unfortunately our Rockwell Rep was unable to provide me with a DNet monitoring tool today so tomorrow all i have is my good old multimeter. Not the man for the job but the customer is screaming so It will have to do.

What i propose to do tomorrow on the troubled networks are as follows:

1. Check (for the 100th time) for loose connnections.

2. Verify DeviceNet voltage is within spec at both ends and middle of trunk line.

3. Ensure termination resistors are ok.

4. Change interscan delay on DNB's from 10mSec to 100mSec

5. Up baud-rates on Networks to 500k. (Cables are well within spec to allow for this)

Any other suggestions would be greatly appreciated. I accept without the correct tools its going to be tough if even possible.

Thanks

 

[MorphuisOGrady]

Guys,

Forgot to mention in last post.... This loss of communications for 41 seconds is really bothering me. Can anyone explain why this may be?

Because these starters are withdrawable cassettes (buckets) its impossible to access the E3 Plus while pumps are running. I suspect that the E3 devices could be powering off. But they cant all be rouge E3's.

They are 24VDC E3's. There is dual 20A power supplies supplying the network but is it possible the Random E3 Plus devices themselves are powering down? If so why does it lose communications for 41 secs every time? I did a bench test on one E3 on DeviceNet today and on loss of power it takes max of 5 secs to establish communications.

When I get a comms failure (Error 78) on DNB on my networks it takes 41 seconds to establish communications again.

 

[MartB]

Guys,

Thanks very much for all your suggestions. I am going back to site tomorrow to investigate further. Unfortunately our Rockwell Rep was unable to provide me with a DNet monitoring tool today so tomorrow all i have is my good old multimeter. Not the man for the job but the customer is screaming so It will have to do.

What i propose to do tomorrow on the troubled networks are as follows:

1. Check (for the 100th time) for loose connnections.

2. Verify DeviceNet voltage is within spec at both ends and middle of trunk line.

3. Ensure termination resistors are ok.

4. Change interscan delay on DNB's from 10mSec to 100mSec

5. Up baud-rates on Networks to 500k. (Cables are well within spec to allow for this)

Any other suggestions would be greatly appreciated. I accept without the correct tools its going to be tough if even possible.

Thanks

Also check all grounding and shielding.

 

[MorphuisOGrady]

Problem solved by replacing DNet scanner to a new Rev C.

Thanks Guys