Fault recorder for a power plant 1

[SA07]

Hi
We need to install a fault recorder on 2 power power plants ( 3.8 MW & 1 MW) which are connected to the utility network. We already have a fault recorder Ben 6000 from Qualitrol installed on 2 other power plants ( 45 MW ) These 4 power plants are located next to each other.
We have a quote from Qualitrol which is quite expensive 66 000 euro.
We want to record the utility voltage, current, alternator voltage, current, protection relay trip, turbine trip. So that we can analyse any disturbance, trip and know the sequence of event.
Usually we have disturbance from the utility network (22kV) due to lightning.
Are there less costly fault recorder from other suppliers?

 

[davidbeach]

Decent relays make dedicated fault recorders unnecessary at this point. Up to 8kHz (160 samples per cycle at 50Hz) and up to exceedingly long at much lower sampling rates. Using the relay also means that you're looking at what the relay saw, not what the fault recorder saw, they may not be the same thing.

 

[sushilksk]

Cost is variable on many factors - like no. of analogue / digital inputs, sampling rate, storage capacity, competition among vendors. You may try getting quote from other vendors like Ametek etc.

Regarding disturbance records from relays, it gets the job done but due to limited no. of analogue / digital channels, DR from many relays to be analyzed in case of disturbance.

 

[dpc]

I have not seen a new fault recorder installed in the past 20 years so I'm with David Beach on this one. There could be special situations, but even so, it would probably be more cost-effective just to purchase digital relays strictly for use a fault recorder.

As for analyzing data from multiple relays, this is probably a strength, not a weakness - provided you have synchronized all of the relay clocks vis IRIG. If all of the relay clocks are left free-running, then deciphering event reports from multiple relays for a complicated event is ...... not fun.

 

[davidbeach]

Good time to every relay, trigger on disturbance, not just trip (we do that with the ‘B’ relays) and gather the records. Unless you’re already at 100% latest and greatest relays I don’t see how money spent on DFRs instead of relays is a wise use of resources.

 

[thermionic1]

There is a DFR at one of our larger thermal plants - circa 2006. It is known for false alarming and locking up. Countless times a relay or SCADA tech has to be sent out. The DFR has to be power cycled to reset the alarm.

All the relays in the control house are synced via IRIG and from what I understand system protection taps directly into the digital relays for any events that require analysis.

I have a project next year to remove that DFR....

 

[dkjfnvfd]

What's the difference between a fault recorder and using a numeric protection delay? (Genuine question - not a challenge).

A good numeric relay should capture voltage and current waveforms pre- and post-fault for analysis and have plenty of other useful features.

 

[sushilksk]

1.Higher sampling speed
2.Higher storage capacity
3.More analogue and digital channels, which can be wired to all elements of a plant. For example in a plant there may be disturbance in one element, numeric relays of other elements may trigger or not trigger recording during this event. But in case of fault recorder, signals of all the elements will be recorded in a single file. Which may be useful for analysis

 

[dkjfnvfd]

That makes a lot of sense. Thanks!

 

[davidbeach]

Relays can be cross-triggered. The sampling and storage issues may or may not be important. Something every now and then and the DFR would be very hard to justify. Regular events and the DFR might be more attractive; but fixing the problems might offer a better return on investment than simply being able to document them.

I'd still put the money into better relays until such time as you've reached top of the line on the relays.

 

[sushilksk]

DFR have their own issues like maintenance, spares etc. Most of the disturbances can be analyzed from the DR of individual relays, DFR data is seldom required. DR files of DFR are larger in size (~10 MB) compared to individual relay file (~100kB), working on them is a little sluggish.

IMO, DFR is a special equipment and have special applications, not required in normal substations.

 

[SA07]

We have the following relays as spare in our store:


Relay protection SEL 501 203X151XXB
Relay protection differential SEL311L 1HDA3213XX
Relay protection differential transfo SEL387E 013X1H2X3X
Relay Siemens 7SJ6022-5EB91-1FB0+L0A
Relay Siemens 7SJ6222-5EB92-1HF0+L0A
Relay Siemens 7SJ6021-5EB91-1FA0+L0A
Relay Siemens 7RW6000-4EA00-1DA0+L0A
Relay ABB Ref615 Com003 2RAA005844A0006B SN 1YF191800550 131758 AIM006 2PCA02 13
Relay ABB Reg 630E-C generator protec. relay SBGNABACABAZA BNBXC 1VHR91113426

Can someone advise which one we can use as a fault recorder/disturbance analyser for our power plants to connect & monitor the following signals?

Signals for Power Plant 1

Digital signal

1 Tie breaker open/close
2 Tie breaker relay ABB REF 615
3 Generator breaker open/close
4 Generator relay ABB Reg 630
5 Turbine Trip from Woodward
6 Boiler trip

Analogue signal

1 Tie line voltage VR
2 Tie line voltage VY
3 Tie line voltage VB
4 Tie line current IR
5 Tie line current IY
6 Tie line current IB
7 Tie line frequency
8 Alternator voltage VR
9 Alternator voltage VY
10 Alternator voltage VB
11 Alternator current IR
12 Alternator current IY
13 Alternator current IB

Signals for Power Plant 2

Digital signal

1 Tie breaker open/close
2 Tie breaker relay ABB REF 615
3 Generator breaker open/close
4 Generator relay Alstom Micom P40
5 Turbine Trip from Woodward
6 Boiler trip

Analogue signal

1 Tie line voltage VR
2 Tie line voltage VY
3 Tie line voltage VB
4 Tie line current IR
5 Tie line current IY
6 Tie line current IB
7 Tie line frequency
8 Alternator voltage VR
9 Alternator voltage VY
10 Alternator voltage VB
11 Alternator current IR
12 Alternator current IY
13 Alternator current IB

 

[sushilksk]

You dont have to use spare relays, existing relay's available DR function can be used.

Say for Alternator, REG630 relay is used, Alternator voltage and currents will be available in its DR. Similarly, say SEL311L is used for Tie line protection, Tie line Voltage and current will be available in its DR.

 

[SA07]

Sometimes when there is a plant trip, we need to determine what has tripped first; the alternator, the network, the turbine or the boiler. We need to determine the sequence of event. Not all these signals are connected to one relay now. May be we should see if it is possible to connect these signals to existing relays. Thks for the advice

 

[davidbeach]

All relays need precision time. Then it's much less important that a single device captures everything.

 

[sushilksk]

Relay time sync using NTP have accuracy upto 1ms. Still if you want all information in one one device, in your case with only 24 analogue inputs, one REL670 or REC670 with 2 TRM cards (with 6 currents and 6 voltages each) can serve the purpose. You may contact ABB for this, it will cost much lower than rate quoted by qualitrol.

 

[davidbeach]

Time distribution via IRIG-B can provide microsecond or better resolution. NTP is better than nothing but I wouldn’t consider it high accuracy time.