Time synchronisation of IEDs

[dfdt]

We use time synchronising of the relays via DNP3 protocol as most of our IEDs are connected via DNP3 over ethernet LAN.I am told that this method of time synchronising is not very accurate(+/- 10 ms?) and bit non deterministic depending on network traffic.I would like to hear your views on pros and cons of time synchronising of IEDs via DNP3 protocol versus via IRIG B signal.
Regards,

 

[DTR2011]

I suppose it depends on what accuracy your are looking for. For protection, I have only seen IRIG. I recently did a project where the metering was via DNP via ethernet. There were issues with the DNP meters only refereshing once per 24hr and the two meters would loose about two seconds between themselves. We had to fiddle around in order to get DNP to sync every second.

In my opinion IRIG is a better and simpler approach. You connect to the source and it works. I understand having to run coax along with the cat5 is more wiring, but there is not much to screw up in the coax(rg-58) connections.

 

[mgtrp]

"but there is not much to screw up in the coax(rg-58) connections."

...or so you'd think. I've seen multiple installers struggle with crimping the terminals onto the coax cable!

 

[dfdt]

What is the best accuracy and resolution achievable via DNP3 protocol? I am looking to achieve an accuracy of +/- 1mS

 

[davidbeach]

IRIG-B, no other option seems to make sense as I see it.

 

[dpc]

We've used DNP for time synch in situations where the device had no IRIG capability. I think the resolution was around 10 ms or so.

For 1 ms accuracy, I think you will have to go with IRIG. If the device has an IRIG input, it would be an easy decision for me. The cost of the GPS time clocks gets lower every day it seems.


David Castor

http://www.cvoes.com

 

[rcw retired EE]

"The cost of the GPS time clocks gets lower every day it seems."

With the cost of labor escalating, we have determined it is cheaper to add a GPS clock at each of the elelctrical rooms in the plant than to pull, terminate, and test the IRIG Coax all over the plant.

DNP does not work well for synchronizing event recorder clocks in protective relays. It is better than nothing but gives errors when comparing event reports. During a plant trip the communications links can get busy and add additional delay errors to event records. What's the hold saying:

"A man with two watches never knows what time it is."

 

[oldfieldguy]

DNP works well when "close" is several milliseconds.

When you need tighter synchronicity, IRIG is the way to go.

And seriously, the cost of IRIG receivers is dropping fast, so they're not a major installation cost any more, more like a rounding error in the cost of a decent substation.

old field guy

 

[slavag]

Yep, today price of GPS is nothing.

 

[bacon4life]

We usually use IRIG-B, however, all our new IEDs in the substation yard are using fiber back to the control room, so we are left with using DNP to time sync them.

 

[electricpete]

Just curious... what is an IED ?

=====================================
(2B)+(2B)' ?

 

[slavag]

Intelligent Electronic Devices = protective terminals.

 

[electricpete]

Thanks slavag.

Good to know. That way I won't duck for cover if someone tells me there's an IED when touring a substation.

=====================================
(2B)+(2B)' ?