Interlocks 1

[mjeshtri]

Please can somebody answer in my question.
In old substations 400/220kV we had by the switch gears (disconector, earthing switch,circuit breaker) wire interlockings. We have now in implimentation 400/110kV new substation, but the contractor (I dont want to say the company name) give as the drawings and there are only logical interlocks, but no wire (electrical) interlocks. I ask for help: From your exprience It's enough only logical interlocks or there must be and electrical interlocks.

With respect
G.Kabashi .

 

[dede61]

Probably your contractor proposed a bay controler that contains the programmable logic to achieve your interlocks by programming. It is generaly accepted now. Of course nothing prevents you of imposing hardwired interlocks. (or do both if you like max. security and complexity)

regards,

Danny

 

[ZapSib]

I would request from contractor to supply hardwired interlock as well.
zapsib.

 

[nightfox1925]

I would highly recommend to add hardwired interlocks to obtain redundancy which leads to better reliability of the system especially in EHV systems.

 

[AllanUK]

In my experience, different clients aproach this issue differently with regard to their willingness to rely on programmable logic of intelligent relays.

It's quite common for ESD and emergency stop push-button trips to be hardwired into the control scheme (so that they are not reliant on any programmable logic), with other inter-locking being done via programmable inputs into the intelligent relay.

Regards,
Allan

 

[waross]

Those of us who saw the advent of programable controllers can remember horror stories of programing glitches and or computer related hardware failures. I remember a multi billion dollar project when all the hard wired stops were over-ridden by the computor. When you pressed an emergency stop button, the contactor would drop out. When you released the stop button the computor would re-energise the equipment. after 5 or 10 seconds the computor would respond to the stop signal and then stop the drive. It was unnerving.
We have been told that the PLCs are so good now that there will never be a problem. Yes, and we were told the same thing BEFORE the problems arose. The problem is that the trust is gone.

For mission critical safety issues, "You can have my hardwired safety interlocks when you pry them from my cold dead fingers!"
respectfully

 

[Ashereng]

waross, that sounds like a really good topic to start a new thread in

Control Systems engineering Forum
(1947 members)
forum830

That is a debate that is still currently going on, passionately, in my industry.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[Siemens1]

Definetely your contractor is using a bay controller. The bay controllers are proven technology. I have been using bay controllers for the last 8 years from different manufacturers.
Redundant hardwire, as is recomended by some of the respondants, will add cost and the reliability of the system will suffer.
What I recomend to you is that you request the contractor that for the status of each switching device they use one open and one close contact to verify the status. This arrangement will prevent misoperations.

Regards

Hans

 

[waross]

Hi Siemens1

The bay controllers are proven technology. I have been using bay controllers for the last 8 years from different manufacturers.

To re-phrase this;

The nuclear power plants are proven technology. We have been using nuclear power plants for more than 8 years from different manufacturers.

ergo; 3 mile island didn't happen.
For the younger members of the forum, the nuclear power industry told the public for years that a serious accident could not happen in a nuclear power plant. The chances of a major accident were non-existent. There were backups and redundant systems for any and all eventualities. You may have noticed that in the U.S public confidence in nuclear power is very low. The phrase "Remember 3 mile island" by itself is still almost enough by itself to stop a proposed new nuclear plant.
After the accident the excuses and apologies started.
I agree that adding hardwired safeties will add cost and the reliability of the system will suffer. But not that much in either case.
Those of us who have been in the field for a few more than 8 years have all seen instances of new, improved technology failing. Sometimes we have participated in discussions as to the need for extra safeties and been over-ruled, and then the proven technology failed.
I give safety more priority than reliability.
I guess we agree to disagree.
respectfully

 

[Ashereng]

waross,

Correct me if I am mistaken, but 3-mile Island accident was caused more than less, by human error?

The accident happened during a "test" of the cooling system or something like that? The safeguard system was either bypassed or reduced because of the test?

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[itsmoked]

Ashereng; That is exactly what happened at Chernobyl.

Keith Cress
Flamin Systems, Inc.- <

http://www.flaminsystems.com>

 

[Ashereng]

itsmoked,

Thanks for the confirmation.

So, if the problem was human error, than a hardwired system requiring operator involvement would be less reliable and less safe than a strictly "safety system only" one?

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[itsmoked]

I agree with you.. No! It may even make it more dangerous because now you have some local guy who can manually command something not having thought out all the ramifications that a designer might have done thoroughly before. I think the general concern is that some transistor or logic device blows and then things start happening out of proper sequence.

Keith Cress
Flamin Systems, Inc.- <

http://www.flaminsystems.com>

 

[waross]

Hi Ashereng
As itsmoked said, Chernoble in Russia was a test gone bad. My understanding is that they were testing the plant's ability to cope with a sudden loss of load and found out that it couldn't cope.
My understanding of 3 Mile Island is that it was an instrumentation failure that led to human errors based on faulty information. The point is, it was said to be fool proof, but the fools messed up anyway. The pre-event hype was that an accident was impossible. PERIOD.. After the event, the justification of human error was raised. When you tell me something is impossible under any circumstances and then it happens, don't expect your explanations to be well received.

On a multi-billion-dollar pulp and paper mill project, the failure rate of the interface cards was over 50% on initial commissioning. Over half of the interface cards were faulty out of the box. These are the cards that we had to depend on to tell the computer to please stop a motor and then we had to depend on the same cards to relay the stop signal from the computer to the motor controller.
The computer group went one further. The computer had the ultimate control. A stop button would interrupt the current to a motor starter, as we would expect it to. It would also send a signal to the computer to stop the motor. When you released the stop button, the computer would immediately re-assert control and re-start the motor. Eventually the computer would respond to the stop signal and re-stop the motor. The response time was 3 or 4 seconds. That's a long time if you’re trying to stop a motor in a panic situation. It's not very good for a large motor either.
We also had to rely on the I/O cards to get the signal to the computer and back. The cards with a greater than 50% failure rate out of the box.
This was a control system designed and built by one of the leading industrial conglomerates in Western Europe. I think I have justification to distrust both their quality and their programing practices.

I am not suggesting going back to total electromechanical controls. I recognize the greater reliability and the greater versatility of computer-based controls.
I'm actually in favor of the new controls, but I still want to see one or two hardwired safeties as back up in the event of the failed I/O circuit that will never fail.
respectfully

 

[davidbeach]

The PLC is a control device. It is not a protection device, nor is it a safety device. The folks who make their living from PLCs might not agree, but don't let them make the important decisions. Anything that has to work should the PLC fail should not be dependent on the PLC.

 

[nightfox1925]

waross, I totally agree with you. I support the present trend of system integration and eletronic trends that advance technology is offering. However, as far electrical protection is concerned, it is worth to consider providing hardwired controls (which has been proven to be reliable in the past)as a redundant backup safety feature in fault isolation. Another thing to consider is the level of technical capability of the user to respond with total microprocessor based controls. I have seen an on-line generator fried up just because the PLC based load shedding scheme did not respond well. Later inverstigations found out that the logic itself was not properly done. It is always worth the cost providing hardwired controls in the protection system as backup for the logic operated controls.

 

[Ashereng]

waross, my apology - I mis-read. Chernobyl was the site of accident during a test.

To everyone:

I have started a new thread dealing with the issue of the ultimate decision maker being the "safety system" or "operator" at the following location.

Safety System vs. Hardwire / Human Intervention
thread830-151978

It would be greatly appreciated if you would contribute to that post.

Thanks.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?