DC shunt/meter amperage problem

[brutus1955]

figured i would try you guys on this one

i am working on a prototype alterntor, 500hz
57.6vdc at 75-100amps

i have a yokogawa analog dc amp meter (50mv) with its
500amp shunt, so i can read up to 500amps with decent resolution.

it has worked well with other alternators, but not this one
it reads about half of what it should at best if at all
usually it just sits there reading zero

if i move the shunt from between the battery bank and alternator to between the batteries and the load it works fine

so i put my oscope on the shunt and i have a bunch of AC noise on the DC line, evidently the diodes in this thing
are less than clean in operation.

so the question, how do i clean this up sufficiently so that my shunt/meter will read when inserted between the alternator and battery?

it has been suggested to place a capacitor of 10uf
across the shunt along with a .01uuf to filter out the noise

does this sound appropriate? can i have too much capacitance
and get a false high reading?

how does one get around this problem without doing something
different in the alternator?

looking for ideas here

thanks
bob g

 

[Skogsgurra]

Use a DC clamp. There are good ones at a low cost. Or buy a LEM DC transducer. Even lower cost.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[ScottyUK]

A capacitor across the alternator output terminals? A few uF should shunt the noise although I'm not sure where it is coming from if this is a rectifier type rather than a commutator type. A polypropylene film type would be ideal - maybe a power factor correction cap from a fluorescent luminaire? Possibly better to work out where the noise originates: have you looked for common ground connections? Your scope is (hopefully) earthed but is the alternator also earthed? That will often cause problems.



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

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

 

[Comcokid]

The problem most likely does not originate with the shunt. You might carefully check the specifications of the Yokogawa instrument you are using. You may be exceeding it's upper frequency limit. Realize that the upper limit is probably specified (measured) with a clean sine wave. The alternator will produce a non-sine output which means lots of harmonics. Furthermore, the Yokogawa may internally be sampling at some interval to make it's measurement - it's going to miss a lot of the voltage and/or current in it measurement depending upon sampling interval/speed and frequency of the signal. (I mention voltage as I don't know if this instrument you mention is actually a power meter). A capacitor may help, but if a lot of your energy is in DC + 'ripple' then only a large value capacitor or a battery will be the only thing that will filter or average the ac energy components to a lower level so the meter will read more correctly.

I presently do some switching power design, and in my case frequently get fooled in rushing to make measurements using Fluke hand-held meters looking at voltage or current - most of the ripple is beyond the specified frequency response of the meter. Sometimes you just have to use a scope to make a DC measurement that's not purely DC.

 

[brutus1955]

Thanks for the help, i finally figured out how to measure
and account for everything, even if it isn't highly accurate
it is 75 amps +/- less than an amp

for now that will have to be good enough, but

i guess i will be working on finding a way to filter the AC components from the DC enough so that my amp meters work.

upon further research it would appear my controller/regulator which is pwm in design is to blame for part of the problem, a significant part i don't know

and the 500hz ripple is another component, which is significant because of this thing being 3phase,,and

the diodes because of their rugged nature are not the quietest switches in the world, but not much i can do about that, at least that i am willing to do.

so time you figure a major ripple of 500hz and three phase, with approx 25khz pwm superimposed on that, and the noisey diodes, harmonics, and the kid down the street with his boom box on wheels, etc

its basically on a par with the early spark gap transmitters of last century!

the batteries do a good job of cleaning up the mess, but
how much power that is going to the batteries to charge them in the future will be a mystery if i don't get this sorted out somehow.

in testing the efficiency has proven very good, so i gotta press forward with this one and see if there is a way to
clean it up enough to get a standard ampgauge to work.

bob g