On a small diesel set (under 1 MW) there is about a 2:1 ratio between no load excitation voltage and full load excitation. You may be able to calculate the value if you have a lot of information about the set which may not be readily available.
But the Automatic Voltage Regulator usually takes care of that.
If you are trouble shooting, don't be surprised if the no-load to full-load ratio is close to 2:1 or more.
Bill
--------------------
"Why not the best?"
Jimmy Carter
I'd agree with Bill on that but qualify it a bit more, in my experience, on PM excited machines up to about 5MW the 2:1 has always been pretty close. On SE machines up to about 4MW it always seemed to be a bit higher, like about 2.5:1.
This is based on units built by CAT, KATO, EM, Ideal, AVK, Newage, Marathon and Baldor.
This is also based on loads with low THD, high load harmonics can have a big effect in some cases.
Thanks (lps) to waross and catserveng for the quick answers. I am doing some troubleshooting of sorts and was just looking for confirmation.
In my case, I am looking at three cummins/marathon 400kw slip ring generators. The no load excitation current readings are, in fact, right at 50% of the full load nameplate value. I am considering the current readng instead of voltage because the tests were performed cold (right after start-up) and the nameplate voltage values are for the hot resistance of the rotor.
Is the excitation proportional (linear) to load from no load to full load?? If 50% excitation is no load and 100% excitation is full load, will 75% excitation equal half load? And, for a specific number, will 67% excitation equal 34% load?? ( there is a reason that I picked this specific number)
My thinking is that the voltage is proportional to excitation but, I would like some confirmation (or some correction if I am wrong).
Honestly I haven't looked at part load exciation data that much, usually I look at the no load info when setting up a machine and verify the full load point, but can't answer to the proportional nature of the excitation.
My experience is similar to yours Mike. But from first principles I have assumed that a generator has a regulation voltage drop similar to a transformer but much greater due to the greater impedance of a generator. I would expect the excitation to be fairly linear with the load as long as the field does not approach saturation. I have not seen anything to suggest that the relationship between load and excitation voltage is not linear, but as I have mentioned, I leave it up to the AVR to worry about the fine details.
Bill
--------------------
"Why not the best?"
Jimmy Carter
The excitation current will vary based on both MW and MVAr loading. You could be generating 0 MW, but outputting maximum MVAr (see a generator capability curve), and be right at maximum excitation current. Or you could also be at rated load and power factor (lagging) and also be at maximum excitation current. Or you could be at rated load and power factor (leading) and be quite a bit less than maximum excitation current.
There are some diagrams and equations in the following link which might help, especially around slide 8 on field current limit:
"...rated load and power factor (lagging) and also be at maximum excitation current."
Actually that would be quite an unusual condition - rotor heating is normally a limiting factor only at low lagging power factors. Under the conditions you noted it is typically the stator heating limit which dominates.
----------------------------------
If we learn from our mistakes I'm getting a great education!
It is fairly common for the generator rated power factor to be at the intersection of stator and rotor thermal limits. This is the point on the capability curve that I was meaning, maybe I'm not expressing myself well?
The locus of rated excitation current shown on most capability charts is quite often inside the true maximum excitation limit imposed by saturation and rotor thermal limits - using that line moves the intersection with the stator limit considerably.
Somewhere I've got a marvellous old document from the CEGB which is one of the most detailed works I've seen on capability diagrams. I'll see if I can find it if you're interested.
----------------------------------
If we learn from our mistakes I'm getting a great education!
The full-load excitation current is mainly dependent on the power factor and the generator synchronous reactance. A large hydro generator with 0.9 p.f. and a synchronous reactance of 1.0 p.u. with a no-load excitation current of 100% has a full-load excitation current of about 160 to 170%. I assume that small generator sets have synchronous reactances in the region of about 1.5 p.u. and higher. The air gap ampere-turns portion of the no-load characteristic then is smaller in comparison with the armature reaction ampere turns. A 2:1 excitation current ratio therefore is quite realistic.
How high is the synchronous reactance of the low voltage set in question? And why don't you measure the excitation currents over various load conditions? Should be easy.
