Self-excited generator

[ScottyUK]

A question for the small generator gurus.

We have a little 290kVA set driven by a steam turbine. Machine was built by Electric Construction in England many years ago, but they have long since passed in to history, and I have no data at all, not even a nameplate.

The machine had a fault before I arrived at site and the exciter was rebuilt. The rotating rectifier diodes were also replaced. I've done all the cold and live commissioning tests I can reasonably do at present - IR's, winding resistances, open circuit curve, voltage balance, exciter diode drops, etc, etc, and all looks reasonably normal. Rated voltage no-load excitation is about 26V and 0.85A.

It stubbornly refuses to self-excite from the residual magnetism in the rotor. The AVR appears to be good - it's an "Anciac 80" by the same manufacturer as the machine - but it isn't giving enough output to build output voltage. If I drive the field externally the AVR tries to regulate once there's a bit more voltage to power it. I am deeply suspicious that the rebuilt exciter parameters are different to the original - fewer turns, different conductor size - but am unable to prove it conclusively. I am getting about 6.4V AC 50Hz between lines on the main output and this is fed back into the AVR. I get about 450mV DC of the correct polarity out of the AVR, but I am not at all convinced that there is enough output voltage from the exciter armature to forward bias the diodes of the shaft-mounted rectifier, which is why it does not build up voltage correctly.

Any thoughts would be appreciated (other than scrap it - that's already on my list ). I wish I was working on a big turbo machine, I know how they work!


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[waross]

I would be looking at the AVR, Scotty. 6.4 Volts sounds like you have enough residual. Only 450 mVolts out of the AVR sounds low. Exciter faults often "smoke" the AVR.
Reversed field leads will also give this symptom.
What is the resistance of the exciter field? I have run into quite a few AVRs that require at least 16 Ohms in the field.
Most AVRs have a lower limit of field resistance that they will work into.
You may try feeding the field with about 3 or 4 Volts DC and see if it starts to come up. I used to flash fields with a worn out 1.5 Volt flashlight (lantern?) battery. I would put it in series with the AVR output. It was usually enough to hick the machine off and re-establish the residual. I have also seen AVR circuits that fed the residual voltage through a bridge rectifier and directly to the exciter field. As the voltage was building up, the diode circuit was disabled so that the AVR could assume control.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

My first step would be to just reverse the leads. Then the size "D" drycell in series.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ScottyUK]

Hi Bill,

Agree the 450mV out of AVR is low. I've swapped the AVR with our spare and both have similar behaviour so they're either both good or both dead. When I power the field from an external DC power supply the output voltage comes up nicely. 4.8V on the exciter field gives me about 70V on the main terminals, and it's pretty linear all the way up to rated voltage.

Once the machine raises terminal voltage the AVR output definitely seems to be behaving normally and can be observed trying to regulate as the output voltage reaches setpoint. Quite what it is doing during build-up I'm not sure - I had expected it to rectify whatever residual output was available and feed it straight through, then to transfer to the regulated output once it became established. A circuit diagram would be wonderful!

I've tried swapping the leads but with basically no effect. If anything the reversed leads acted to reduce the available voltage a fraction but it wasn't convincing.

The most infuriating thing is that there's a 110V DC auxiliary supply from the critical services battery that is used for control within the panel. I am _so_ tempted to set up a field flash from the auxiliary supply using a timer, resistor and a couple of diodes, but equally I shouldn't have to do that because it used to work.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[waross]

I'm sorry Scotty. I have seen AVRs with a straight through flashing circuit but I have never had a circuit diagram. I jury rigged one at a fishing resort out in the jungle of the Moskito Coast. I used a relay to disconnect the diodes when the field started to build.
Another option may be to power the AVR from a separate source. Some AVRs have common power and sense terminals and this is not possible. You may lose Black Start capability.
Another option is a new AVR. Depends on the budget and it may not go well for your reputation if there is another fault.
Your field resistance may be too low if the exciter field was rewound to a different spec.
Was this originally brushless or has it been converted from a brush type exciter?


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ScottyUK]

Hi Bill,

I'm certainly not against a new AVR although I'm not sure which specific types are 'good' or 'bad'. I had a quick look at Stamford's SX460 and AS440 and on paper they seem reasonably equivalent to what I've got but I have no practical experience with them. There are so many aftermarket AVRs that I'm sick of looking at them, so if anyone can say "Use Brand X, they're great!" or "Never in a million years do you want one from Brand Y!" then that would be useful.

The field resistance is 28.5 ohms so I think that is fine, I'm more concerned that the re-wound exciter effectively has a lower gain than the original and with marginal conditions for establishing excitation that might be enough to have tipped it over the edge.

Does anyone out there have the OEM data for an Electric Construction BRF 450 generator dating from the mid-1970s?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[waross]

Hi Scotty, I 'm on my way out to take William to school. Don't have time to check the stats just now but I have had good luck with Stamford AVRs. SX 440, & SX 460 ring a bell. Basler also has good AVRs for small sets.
It may be 10 hrs or so until I can do some research on specific AVR models for you.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[edison123]

Scotty, I would check/replace the wiring from the AVR to the exciter field. Loooks like bad connection that self rectifies.

Muthu

http://www.edison.co.in

 

[waross]

Good point Muthu. I haven't seen that for years. Had a bad connection in a motor armature circuit so bad that the motor would not run in reverse. Somewhere between 5 HP and 15 HP.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

Hi Scotty;
How are you making out with the generator? I found some old instruction sets for FG Wilson generator sets. They were using all Stamford AVRs. The model varied according to the set size and a few other factors.
The SX 440 was always dependable. Price and what is in stock are factors I most often use to choose when more than one AVR will do the job. In Central America, mostly it's;
"This is the only one we can get. Will it work? Yes it probably will work."
I have always found AVRs on small diesel sets to be pretty forgiving. Sort of like hammers. One size doesn't fit all but one size fits almost all.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ScottyUK]

"How am I making out with the generator?" If that machine was a girl we would not be having a second date!!

Joking aside, haven't done much this past day or two other than look for AVR suppliers. I can get the antique repaired by this company and they've also offered a functional replacement for about £640 GBP (their type REG 7). The replacement seems to have a bit of excess capability compared to the original which should mean it runs cooler and lasts longer. I'm just a little wary if it is a one-man-band that I'm storing up problems for the future in terms of spares.

I'm waiting for Cummins to get back to me with a couple of prices on the Stamford AVRs. Price-wise I might do better going to the aftermarket, but from reading a few comments on the 'net there seem to be a fair number of counterfeit Cummins AVRs out there. Amazing.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[glendale]

The Electric Construction Company of Wolverhampton was taken over by Brush Electrical Machines (Loughborough) in the mid 70's, I was a commissioning engineer with Brush at the time and remember the drawings etc all transferred to them, might be worth a call to their service department.

 

[Hoxton]

I seem to think that the designs (generator designs - perhaps not AVR's) were purchased by someone who carried on making them for a while - i cannot find them on the web, so thay may have gone as well.

Try brush, as Glendale says.

ANCIAC (AVR) stands for "A New Concept In Automatic Control", by the way!

 

[rmw]

There you go Scotty, back dealing with the Service Dep't at Brush again. Isn't that wonderful news??????

They are such angels as regards responding promptly about their own kit, they should be just WONDERFUL when dealing with a bunch of obsolete stuff.

Best of luck.

rmw

 

[ScottyUK]

Hahahahaha...

Glendale / Hoxton,

Thanks for the lead. I'm gonna guess that Brush will probably want to sell me a MAVR or a PRISMIC regulator. Both are a little bit overkill for my application. It's certainly worth a call though, if only to confirm rmw's expectation.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[ScottyUK]

A quick update: we replaced the AVR with an aftermarket type from Roper Electronics which is a nice little unit, old technology but maintainable and field-repairable. The machine still wouldn't self-excite, so I've given up on using residual magnetism and fitted a field flashing supply derived from the 110V station battery via an isolating converter. Comes up beautifully every time.

For anyone who wants the manual for the ANCIAC 80, I've attached it here.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[waross]

Thanks for the update, Scotty.
Yours
Bill

Bill
--------------------
"Why not the best?"
Jimmy Carter