Doosan G90+wj dWJD-2a-T3 (C24) laid down on me...

[Tony Wells]

Thanks for the explanation, Birken. I'll run a few more tests Monday and see what I can turn up. I won't be working on it over the weekend, other than mentally. If another test should cross your mind, please don't hesitate to note it here.

So basically if the exciter is working (and it is) to generate some voltage on the final output, the power for the AVR is coming from the Aux windings, and it seems to be or there would not be 26 volts on the exciter terminals (d/c), then something may be related to the sensing circuit. If that was having problems it would kill the exciter voltage....but it doesn't, unless the primary field coil is connected. Seems very odd to me. I'm no stranger to electricity or electronics, but I haven't gained a full grasp of this method of regulation/generation yet. With everything connected, the main output is a measly 16-18 volts (not truly zero).....so maybe there is a false flag sent to the AVR by the sense circuit causing to fall off. IIRC, there are toroids that each of the three legs go through just above the main breaker. Seems they are labeled "load sense". Can hardly see how they could fail. But I'll check them.

 

[Birken Vogt]

Those toroids are just for some meter or controller to display how much current the thing is making and won't be anything to do with this.

The 26 volts may be what it is making open circuit, but connected to the actual load of the exciter coil, it might fall to nothing.

Operation is really quite simple. When it starts, there is some residual magnetism that makes a weak voltage on the output coils. The 12 main coils, plus the 2 aux coils. The AVR takes the output of the aux coils, rectifies them, and sends them out as DC onto the field coil. When sensed voltage approaches the set point (480) then the AVR will cut back on the DC going to the field and hold it regulated at 480.

The sensing wires might be tapped to sense the full 480, or they could be center tapped to sense 240, or maybe 277, who knows.

 

[Tony Wells]

That's pretty much as I thought. Same as most generators. Since the aux windings are powering the AVR, and the 26 VDC sags (or disappears completely, yet to be measured) when connected to the primary field, that seems like a possible trouble spot. The 26 VDC should vary depending on load, but try to maintain a set voltage that keeps the main output stable. But they should supply close to spec just running with no load just to hold the voltage setpoint. That being the case, I would think that voltage would vary with the voltage adjust pot on the front panel. I think it's a 3W 470Ω.

Since this is a brand new AVR, I assumed it came in with the trimmer pots adjusted, but I got no paperwork with it. If that assumption is correct, then either it is bad, which since it behaves exactly as the original doesn't seem too likely, or I have to suspect something external to it. It did arrive with the 50Hz jumper in place, so I had to move it to 60Hz.

Makes sense on the toroids. Forgot about driving the panel meter for current.

I do have the full schematic for the power side of the machine, so I'll have a look to see where those sensing wires connect. I'll check that circuit if they are not providing a signal. However, if the sensing wires were dead, for example, how would it be making the 26 VDC? Or shorted/open?

I don't like the idea, but I have a known good G70 with the same AVR I could swap in if I really think the AVR is bad. This will probably come down to a single wire broken or burned that I just haven't found yet. I sometime fall for those easy assumptions. Bad habit.

You've been a great help Birken, and I appreciate it. I've not had to work on too many generators, so this is a little unfamiliar territory. For now at least. But it's getting clearer.

 

[Birken Vogt]

The pots adjustment will only be fine tuning. It will still work no matter how they are adjusted.

It is the 26VDC open ckt because the residual voltage is getting rectified and supplied in its attempt to get the thing started. But is the 26VDC collapsing when connected to the actual field?

Normal sequence is, 1. residual makes weak AC voltage. 2. Weak AC is rectified making weak DC. 3. Weak DC makes more AC. 4. More AC gets rectified into more DC and it takes off and makes full voltage.

If the sequence breaks down, driving 12 volts onto the field momentarily will cause it to gain some AC and then take off. I think this is not happening for you.

I would hook everything up, apply 12 volts to the field, correct polarity, along with the VR still connected, then measure AC voltages on the sensing and on both aux windings. If any are missing, there is a problem. The wildcard is one of the aux windings here is apparently a current derived one and not voltage derived. So it might not show much with no load. That is apparently a L-S unique design.

Actually, what I personally would do is attach a simple AVR that I carry on the truck for things like this. Sensing is also power on the simple ones. So feed AC in the one end and DC comes out the other end, case closed. That can prove that the generator at least can work. But the battery method kind of proves that too.

 

[Tony Wells]

snip<It is the 26VDC open ckt because the residual voltage is getting rectified and supplied in its attempt to get the thing started. But is the 26VDC collapsing when connected to the actual field?>snip

It would seem so, since it will not produce mains voltage with the primary field connected. I have only checked it while running the 12 VDC to the primary field. Connected to the battery, it is present. I will reconnect the primary field to the AVR, AND apply 12 volts (or 24 possibly, since it is closer to the OCV on the AVR terminals) and confirm. If it were holding, the battery would be unnecessary to power the primary field.

 

[Birken Vogt]

Just beware, if the regulator tries to "take off" and power the field, due to the battery, it will also be trying to charge the battery which of course it does not have enough power to do. So the battery test is only for a short blip, or done through a diode so the AVR cannot charge the battery.

 

[Tony Wells]

On this last mention of connecting 12 VDC to the primary field, while connected to the AVR, you mean only a momentary contact of the battery, if I'm understanding you correctly. Or through a diode. I have a few 1N5404 diodes on hand. Along with a few others.

 

[Birken Vogt]

Yes that's right. Momentary contact to see if it will light off, if leaving it hooked up and there is a chance it might light off, then use a diode on the battery.

 

[Tony Wells]

Gotcha....thanks!