Generator Engine Control

You can use a PLC for auto cranking but have an output turning on a starter relay in the same panel, which in turn provides power to your starter solenoid mounted near your starter motor.
 
I would think that is a big 'no no'. Our 'Emergency' back up systems use a synchronous motor relay for timing 'start' on gens.

I would put an HMI at the Nurses call station. When power goes out, give them a choice of 'Cold', 'Cold/Can't see', and 'Bobby pooped his pants in the frozen toilet', perhaps?

Fire that 'Emergency Timer' accordingly, in logic of course.

Johnny
 
I can see a synch timer in the transfer switch, but not really in the cranking panel/circuit.

Basic CAT cranking circuit is 7 relays, 4 lights, a timer, and two switches.

In 1972, four of us were working at CAT's leading supplier, and each proposed a solid state cranking panel for consideration. mostly radio shack parts. four completely different approaches. They thought solid state was far off in the future.

A lot of panels still being shipped with the stock 7 relays and a timer.

regards.....casey
 
Generator auto start

Casey,

From a few incidents at Boeing where generator would not start (batteries were not charged) I am a little skeptical over all this auto stuff.

We have a Cat 275 Kw (??) at work 5 (??) years old and it has all solid state control (I think). there is not one manual switch on or a pressure guage for oil even. All is displayed on a screen.

PLC control would be great.
BUT
IF it were my unit I would have a manual set of controls manual transfer and if I really had my way a synch scope to parallel. That way when I wanted that unit on line I could get it there. Still have to keep batteries charged of course.

Dan Bentler
 
Ded batteries can be a bummer. wful hard to kick start a V-16 CAT or Cummins. When I started in 72, industry standard AT THAT TIME was to have the battery charger fed through relay contacts in the cranking panel, which allowed the charger to be hooked up at all times EXCEPT during crank.

Later, I saw some chargers had a terminal that hooked up to the starter circuit, for a crank disconnect relay.

Some of the more moder stuph I've seen is almost an insult. Specially some of the solid state panels that do not have components rated for typical operating voltages or current. But you can save a few dollars.

Oh well.

regards.....casey



.
 
I went to build an auto start and transfer for my home. Issues that I came up with included:

How to detect when started: possable cent.speed switch on crankshaft?
How to give up on starting when it wont start...bla bla bla.. it turned out a whole lot more complicated than I initally thought.

Wife got tired of my "project"; we were losing power often so wifie ended it all and bought a pull start generator at Costco.
 
Generator and Load Control systems are one of my "specialties". So far I have done three hospitals, two small cruise ships, an airport and both main buildings for the NZ parliament.

Yes it is a learning curve.
Wife got tired of my "project"; we were losing power often so wifie ended it all

Mike if you think wife is tough, try turning off the power to the entire executive wing of your government!! Sheesh. Usually I only worked live on it late Saturday nights/Sunday mornings, but any outage over 30 seconds still needs an "explanation".

No it is not rocket science, but much depends on the configuration of the system. Simple transfer systems are quite straighforward enough , but the Parliament Exec Building project for example has a split main board each fed with its own mains (utlity) and generator, and a bus coupler that combines the two sides whenever any one main is disconnected. The generator can only carry about 45% of the building's maximum load so there are about 40 controlled loads distributed about the building on two Devicenet segments to manage the total load as per the capacity of the connected supply.

Sounds simple enough but the basic switching map for this system has 246 different possible Main Breaker (MCB) switching combinations alone. That is before you start taking into account all the possible generator and MCB fault conditions!! This nightmare was eventually tamed with a nine state Karnaugh map and at 800 hours of work I stopped counting how much time I put into it. I would love to post the file, but it is definitely not my property to do so.

What Casey is asking for is quite doable in a small PLC as long as the genset is going to remain isolated from the utlity supply, ie no synchronising and PF/VAR regulation. As soon as you want to do that then it all rapidly gets more interesting. If you have a more detailed spec for what you want to achieve feel free to email me direct if you wish.
 
Last edited:
I have just had a look at the list of PLC/design jobs that I have done over the last 10 years. There are no less than 132 generator jobs done in that time. They are all PLC controlled. There have been with Cummins, CAT, Perkins, MTU, MAN, Ruston, Detroit, English Electric, Allen, Dorman, Paxman sets and a few others.

Have never had a problem with a proper design. Most systems have been multiple engine with synchronising and load sharing (use specialised modules for this and then a PLC bypass system can be designed in), SCTT (soft transfer), Co-generation, base load power stations, 415V generation, 415V with step up to 11kV, 11kV generation, laod control, load shedding, capacity control, auto start up after a shut down (base load power stations), building load control (up to 11 networked PLC's) - you name it.

In all this time the only real PLC problems I have had is an occassional faulty power supply.

SCADA systems are nearly always used over the top of complex systems. I always put an "automatic return" button on the control cubicle in case Mr Gates operating system does what it does best. That way they can put the building, or whatever, back to normal without the computer.

Have used remote I/O etc (usually Device Net). The largest distributed I/O system I did for one of these jobs was 92 Device Net I/O blocks into 1 PLC plus a total of 23 Allen Bradley Powermonitor 3000 devices, also on Device Net into the same PLC. There were 9 Device Net scanners on the backplane. Works like a charm. The AB PM3000 units are addressed by explicit messaging. Data is taken into the PLC and then extracted by Citect SCADA on an Omron network. Updates to the SCADA from each of these devices is better than twice a second. Awesome.

Have used AB, Mitsubishi, Hitachi, Siemens, IDEC (Izumi), Schneider, GE 90-30, and a couple of others over the years.

The systems are very reliable and run on battery (for obvious reasons). Batteries are not a problem at all provided proper maintenance checks are carried out. We recommend that every 3 months a discharge test is carried out on all battery systems. The data is then entered into an Excel spreadsheet for each battery bank. If there is a noticeable drop in performance from a battery set this can be shown in Excel as a chart and the client can then decide whether to replace the batteries or not. I have not experienced a catostrophic battery failure at ANY site where this has been the practice. Have seen plenty of catostrophic battery failures where the client has decided to "save a buck"!!!!

During these years I have also formed an opinion about the most reliable PLC DC power supplies, due to monitoring any failures. I will not go into that here but suffice it to say I now have a favourite brand for lots of reasons, not the least of which is reliability.


I cannot remember the last time I did a generator job without a PLC, including several fairly large base load power stations. Nearly all output cards are transistor with diode protected relays to drive load. A lot easier to replace an external relay than have to pull a relay out of a PCB in the middle of the night. Yes, Murphy was an optimist - it always happens in the middle of the night.
 
Last edited:
I had not seen any PLC's running engines in the field, granted I don't get around them like I used to. I have used a large variety of solid state controls and modules that were rammed down my throat.

I did a large prison job several years ago, with lots of remote monitoring and load shedding where the "heart" of the system was a 9030, with a backup 9030 for other critical operations, and the PC with cimplicity was for remote monitoring and control, not the "brains" of the system. But the engine was still controlled by 7 relays (well, actually 25-30) and a timer.

I always thought PLC's were the way to go on an engine control, but it is not the best environment, even when the controls are remote mounted, a lot of spikes and noise, even with filtering and conditioning.

There are a lot of small bricks out there for the simple systems, most likely would use AB, GE, Omron, or Entertron. Like the idea of entertron, some enterprising service tech isn't going to mess with my timers.

Larger stuph, probably 9030 with load control capabilities.

more later, time to give up the phone line.

regards.....casey
 
Casey

I have been using Omron in these things for years. Write my own load control, engine duty order selection to SCADA screens (Citect), etc etc.

Got a call the other day for a PLC tech to diagnose problems in a gas set (Wauhesha). The PLC was an old Omrom C40K. The set gets dragged all over the place on a skid in an open cut mine. PLC has been in the set for about 8 years. PLC was OK, of course. Emergency stop buttons were falling apart, relay bases blown up form being under water, relays were "well worn" to say the least.

Replaced these parts and away she went. Do not underestimate what a PLC can take. It is quite remarkable where I have seen them working. But of course, when the set breaks down it is always the PLC, as above.

Larger jobs, larger Omron PLCs.
 
"But of course, when the set breaks down it is always the PLC, as above."


Of course...Dont you know that the "mystery box" is ALWAYS the problem....

Most of the younger techs have come around, but still dealing with some of the older guys...
 
Whenever I am called into a plant room to diagnose a fault I have made it a point not to carry in the laptop. You really get some strange looks.

The first thing I ask for is a set of up to date drawings. The second is "what is happening?" The third thing is drag out the trusty old multimeter (new now - I just picked up a new Wavetek) and chaeck out power availability and level. One of the first things I was taught as an apprentice is check that all 3 phases are there. Amazing how many times you find a blown fuse.

The laptop is the last thing, when I have my head around the plant and what the fault is.

When commissioning a new job that I have designed, I find I walk in with the tool bag and laptop, download the program and then start checking out the I/O to make sure the wiring is correct. Have been held up for weeks by mechanical people and then it is a case of "just dump in the program and it will work". Like a dollar for every incorrectly connected wire or faulty mechanical device I have found. Then the worst in the world because something has to be fixed. Even had a mechanical guy ask one day if I could fix his faulty low oil pressure switch by "tricking" the software.
 
Bob, experience is a wonderful thing no?

Actually my wife says that "experience is something you get just AFTER you need it".

But I could not agree more with the mode of troubleshooting you outline in the post above. Almost always the PLC program is NOT the cause of the fault, and time spent doing basic fault finding always pays off.

Yesterday I got an urgent call to a pump station...crisis mode stuff...the local guy was very terse about how the !@#$ PLC had "died" again. Literally within 20 seconds of arriving I found that the 24v control supply was only at 9.6v due to a faulty PSU.

After a while you learn to just look and listen for a while and the clues just start talking to you.
 

Similar Topics

Hello, I have been working on a project that uses linear actuators to control a diesel engine generator. I have all the control part done with...
Replies
9
Views
22,121
Hello Experts, I am trying to create a SINE Wave generator and i cannot seem to make it smoother. I have done this before but i forgot the exact...
Replies
6
Views
239
Hi dear forum, I need to create a test sine wave that so that I could check settling time in bidirectional inverters. Function that needs to run...
Replies
12
Views
1,126
Hi, I have an older Kohler Generator/ATS controller that has a Schneider Electric (Modicon) Momentum PLC and has a Advantech HMI. I assume this...
Replies
3
Views
677
Anybody have any experience using a tone generator on analog signal wires? Trying to locate two analog level sensors in an old plc5 panel that...
Replies
7
Views
964
Back
Top Bottom