AC vs DC amps

[ChrisOlson]

......and I found out, sampling one leg, that I get more DC amps (out of all three rectifiers) than I get sampling that one leg on AC.  It's a little hard to get exact numbers because the meter keeps varying slightly, but this is an approximation.

Why is this?  I always assumed this was a one-to-one ratio.

Using the conventional method of measuring power output from a three phase generator, I'm seeing a "clamp" voltage of ~10.8 VAC between any two legs, and using my example of ~3.5 amps, this generator is putting out 37.8 watts.  On any diesel-driven three-phase generator this is what it would be - RMS volts x amps = watts.

However, looking at the actual DC output going to the battery I'm seeing a "clamp" voltage of ~14 VDC and ~10 amps = ~140 watts.

Who ever can explain this to me wins the prize.  I don't know what the prize is, but they win it.

--

Chris

[Flux]

This is a messy business, much depends on the conduction mode of the rectifiers.

Taking the star case, the rectified dc peak is 1.414 times the ac rms. The dc mean which is a more reasonable thing to consider is 1.4 times the ac rms. You have to subtract diode drops on low voltage systems and that adds more confusion.

So yes the dc will be higher than the ac Take 1.4 times your 10.8v ac giving 15.12 dc and loose a couple of diode drops and you have your 14v battery volts.

Current is not as simple as rectifier conduction mode comes into this but with an air gap alternator you wouldn't expect diode overlap so you have the dc current carried at any instant via 2 diodes. The dc current flows through 2 phases at any instant with the 3rd idle.

You therefore have a current equal to the dc value flowing for 2/3 of the time. The rms value of this  comes out to something like 0.77 of the dc value.

So you end up with both the ac rms voltage and the ac rms current being less than the dc components.

If you try this with iron cored alternators with leakage reactance you will get overlap in the diode conduction and the rms current is different. I am fairly sure that if you take the IRP connection you will again find diode overlap when supplying high current and this may be why under extreme overload conditions it shows less stator heating ( or is claimed to do so).

You must not confuse this rectified load with the normal power in a 3 phase system where Power is root3 x rms line volts  x rms line current x power factor.

Flux

[ChrisOlson]

You therefore have a current equal to the dc value flowing for 2/3 of the time. The rms value of this  comes out to something like 0.77 of the dc value.

I'm more of a nuts and bolts guy, but I sort of understand what you're explaining.

Watching the two meters, and the wind is very powerful and gusty here today so it's hard to get a definite "snapshot" of it, it appears that what I'm getting on the DC side is roughly 3x that of what the AC ammeter says on the leg it's sampling.  I say "roughly" because it's not exactly 3x.

If I wait for a lull in the wind so it stays more steady, and I can use 10 amps DC as a reference, it looks like it takes about 3.5 AC amps to give me that 10 amps DC.

I wonder if it makes a difference as to how the rectifiers are hooked up?  I have dual 30 amp block diodes on each leg.  I run a leg to one AC terminal, then jumper that leg to the other AC terminal on the same block, then jumper that to the other block, which is jumpered too.  Then I have all the DC terminals on all blocks hooked in parallel.

I had to do this on this 12 volt unit because I kept burning those block diodes out if I only used a single block diode per stator leg.

--

Chris

[Flux]

You seem to be getting a fair bit more dc for a given ac current than normal.

As I said I would expect over 7 A on the ac side for 10A dc.

I have no idea what your ammeter measures, unless it is a true rms digital or a moving iron or dynamometer analogue unit it will not measure the rms value and the waveform will be virtually square wave. A normal cheap multimeter calibrated to measure rms on sine wave will be wildly in error.

From your description you seem to have a normal 3 phase bridge rectifier made up of blocks of single phase units. You should be measuring the ac in the incoming lead from the alternator not in the link between the ac connections on the bridge. If you are in the link then you would see half the current.

Flux

[ChrisOlson]

Yes, I just discovered I had a problem with my setup here.  I had a shunt-type DC ammeter that was measuring both turbine output and inverter load to run my laptop and lights in the shop office here.

I waited for a lull and pulled the DC side wires off the rectifier heat sink and quick hooked up a regular inline DC ammeter between the rectifiers and the battery bank.  Now it shows more accurate numbers (I think).

When the inline meter says 10 amps the AC ammeter (digital and jumping around a lot) says somewhere between 7.5-8 amps.  Which agrees with that .77 number you quoted.

Thanks for explaining this to me, flux.  I can design and build a complicated mechanical device, but when it comes to electronics stuff I'm a little lost sometimes.

--

Chris