Fieldlines.com || Make your electricity from scratch!

Front Page - [Homebrewed Electricity-- (wind) (solar) (hydro) (steam) (controls) (storage) (mechanical)] - Classifieds - Site News
Everything - Newbies - [Remote Living-- (housing) (heat) (light) (water)] - Reviews - Diaries - Our Products

DC current measurement problem
By bob g, Section Controls Posted on Fri Jul 10, 2009 at 11:44:08 PM MST
I am having trouble with a dc meter with shunt...

i have used it in testing before without issue, it is a 50mv movement
and a 500amp shunt
i have measured and verified its accuracy to within the spec's i need
and it has never given me any trouble, till now
i am working with yet another alternator, it is running at 500hz
57.6vdc charging into a set of grp 31 batteries, which are a bit long in the tooth
but will take a charge just fine.
i have a pair of inverters (48volt) driving a pair of 1500watt heaters
and all is just fine on that end, 120volts ac, good sinewave, ~3000 watt load
i also know the inverter efficiencies are right at 87%
so 3000 watt load divided by inverter efficiency gives me the output from alternator
to the batteries, basically i know it is all coming from the alternator because it is maintaining the 57.6volts instead of the at rest volt reading from the batteries of just over 50 volts.
therefore 3000/.87 = 3448 watts
and 3448 watts / 57.6 = 59.86 amps
the problem is the amp gage shows approx 20amps which cannot be right because
20 x 57.6 = 1152 watts, not even a third of what i am using.
so i put a scope on the shunt to check the waveform and it has a heavy AC component
i added a capacitor of unknown mfd to the alternator outlet and that helped a bit
the waveform improved and the amp reading went up a bit to about 40amps or approx
2300watts,, still not enough!
now i understand AC and DC but am a bit foggy on DC with an AC component overlaid on it. seems like this is what is giving me the erroneous error readings?
so what do i do about it? add more microfarads to the alternator DC output to clean out this component or do i need to also put in an inductor to filter this stuff out?
it doesn't appear that this "noise" on the buss is harmful to anything, but it keeps me from getting the needed accuracy in testing i am after.
any thoughts or direction from the guru's would be appreciated
if i can use a capacitor, how do i size the thing? can i put too much capacitance
on the output?
thanks
bob g

DC current measurement problem | 13 comments (13 topical)

Re: DC current measurement problem (3.00 / 0) (#1)
by commanda on Fri Jul 10, 2009 at 10:08:44 PM MST

I would put some capacitors directly across the shunt. Something in the range of 10uF, paralleled with 100nF. Leads as short as practicable.

Amanda

Re: DC current measurement problem (3.00 / 0) (#2)
by bob g on Fri Jul 10, 2009 at 10:20:35 PM MST

by across the shunt, do you mean across where the leads to the meter are attached?

i tried on cap of 1uf and it seemed to help, but didn't know how much to add

can i add too much and get reading that are falsely higher than they should be?

thanks
bob g

[ Parent ]

Re: DC current measurement problem (3.00 / 0) (#3)
by commanda on Fri Jul 10, 2009 at 11:48:14 PM MST

Large electrolytic capacitors have inductance due to the way they are made, that's why I said to put a small (100nF) in parallel as well.

where the leads to the meter are attached?
Yes.

1uF probably a bit small. 10-100 educated guess.

Amanda

[ Parent ]

Re: DC current measurement problem (3.00 / 0) (#4)
by bob g on Sat Jul 11, 2009 at 12:23:13 AM MST

thank you very much!

i will give it a go tomorrow afternoon and report back

i have several low voltage cap's to choose from, but only a couple with
voltages high enough to cover the alternator output.

thanks again

bob g

[ Parent ]

Re: DC current measurement problem (3.00 / 0) (#5)
by Flux on Sat Jul 11, 2009 at 12:39:50 AM MST

I assume the dreaded meter is digital. A moving coil analogue meter measures dc mean with no problems. If it is analogue and reads low you have the wrong lead resistance between the meter and shunt.

If it is one of those digital things then I really have no idea what it measures, it probably measures the mean of the peak ripple or something and if the ripple is near its internal ramp integrating frequency it may do very strange things with aliasing.

If it is digital it will have a very high internal impedance and I would add a filter of perhaps 10k and 10uF to filter the waveform from the shunt to mean value. It will be a bit slow responding but it should at least see mean value. When you get sense you can reduce the capacitor value if it is too slow.

If at all possible get a decent analogue meter and calibrate the digital monstrosity against it, they usually have a small adjustment pot hidden somewhere. It seems strange to me that very expensive digital panel meters are often worse than a very cheap multimeter on anything but perfect supplies.

If for some reason you have a rms reading digital meter it will be no use for battery charging if there is any ripple or rms component present, batteries see mean current.

Flux

[ Parent ]

Re: DC current measurement problem (3.00 / 0) (#6)
by bob g on Sat Jul 11, 2009 at 03:35:54 AM MST

Flux:

the meter head is an analog unit, d'arsenval type movement and is made by
yokogawa

it works very well on my other alternator mod's without issue, and it will work just fine if i place it between the battery bank and load.

it just won't work between the alternator and battery bank.

i am also concerned as to the suitability of the output for charging batteries as well, although it seems to be good enough in quality to actually charge the 4 group 31 truck batteries i have connected in series to make up a 48volt test bank.
the ups/inverters (a pair of apc 2200va/1600watt) don't mind the power at all coming from the system, apparently the batteries smooth things up enough as evidenced by zero ripple nice clean power to the ups's illustrated on the oscope.

perhaps i should be looking at a large bank of caps to smooth the alternator output
before this alternator is put into service charging batteries i care about?

my thinking is it might well take 400,000uf or more to make much of a difference,
but then again when i look at what is shown on the scope it looks to be a rather complex signal made up of various ac components, do maybe a large and small uf across the alternator output much like commanda suggests across the shunt?

bob g

[ Parent ]

Re: DC current measurement problem (3.00 / 0) (#7)
by boB on Sat Jul 11, 2009 at 10:51:46 AM MST

Adding TOO big of a capacitor directly across the shunt is not the correct place cuz then all the AC components will just be bypassed instead of being measured. You will need to add a resistor, like Flux said, from one (or both) of the shunt terminals to the meter leads and then a capacitor across the meter leads. The capacitor needs some impedance to work from to actually filter the voltage from the shunt.

Even just a few Ohms should help immensely.

boB

[ Parent ]

Re: DC current measurement problem (3.00 / 0) (#8)
by Flux on Sat Jul 11, 2009 at 01:41:38 PM MST

If it is a moving coil meter it will read mean. There is no need for filters capacitors or anything else. Don't add any series resistance to try to filter things. It will be calibrated with a specific resistance of connecting lead between the shunt and meter movement, if you don't know what this is, it will be low and perhaps .027 ohms. Keep the lead resistance as low as possible if you don't have the exact figure.

What it measures will be what actually charges the battery. You may have a rms current considerably higher than your analogue meter is showing but for battery charging the moving coil meter will give the correct answer. If some of this ripply current is going direct to the inverter rather than the battery it may confuse things but for the battery itself just trust the analogue meter.

Flux

[ Parent ]

Re: DC current measurement problem (3.00 / 0) (#9)
by bob g on Sat Jul 11, 2009 at 04:44:14 PM MST

this is the problem, i cannot trust the meter!

it shows no more than 20amps, when i know i am passing in excess of 70 amps

if i add some capacitance across the meter i can get it to read 40amps, but no more.

from doing some more research it appears the problem might be the pwm controller/regulator imposing high frequency/switching onto the DC output.

i am thinking of placeing a filter on the field to clean up what is being fed
to the rotor, in hopes that i might get a cleaner DC output, we shall see

also thought about mounting the shunt closer to the battery instead of close to the alterntor and also thinking of shielding the field wire as well.

more to come

thanks everyone for your input
it is appreciated

bob g

[ Parent ]

Re: DC current measurement problem (3.00 / 0) (#10)
by Flux on Sun Jul 12, 2009 at 12:30:25 AM MST

If you can get it up to 40A with capacitance across the meter then I suspect 40A is your true current. I have never had to add any capacitance or filters to a moving coil meter but if it makes a difference then there may be some major ripple on the current waveform, way beyond what is expected in a normal circuit.

Because of the very low lead resistance needed you really can't add any significant series resistance between the shunt and meter. You could add inductance as long as the choke has a low dc resistance. Try something like the secondary of a low voltage transformer in series with the same value capacitor as before and if it still only reached 40A then you have 40A mean available for battery charging.

You may of course have more than 40A flowing if the rms component is higher than the mean, it may or may not help to feed the load but it will not contribute to the battery charging.

If you can find a large low resistance choke to put in the dc line you could reduce the ripple on the actual current and then the meter itself would measure the current you have without trying to filter. It will take very little inductance to remove fast switching edges from pwm controllers but if it is a high frequency alternator it may take a bit more to remove the rectifier ripple but that is not normally sufficient to affect a moving coil meter.

With these strange conditions you have to decide which components of the waveform are meaningful to you and choose suitable metering to measure it. The moving coil meter measures mean. If you want the true rms then the only analogue devices to use are moving iron or dynamometer but most may fail with frequencies as high as you seem to have.

Measurements under strange conditions can be very difficult and just taking the answer from the first meter available may be miles off the mark.

As far as the actual battery is concerned it works on mean current and if you have a low mean and enough ripple to raise the rms way above this the battery will only be charged on the mean bit even though the rms may heat the cables way beyond what is doing the charging. With inverters having large input filter capacitors I am again inclined to think that the ripple and rms component is not going to do much more than heat the connecting wires. I suspect that if you ignorer any rms component you have no more than 40A available from your comments.

Anyway best of luck, just keep experimenting to decide which component of the actual current is doing what you want and then choose suitable meter types to measure the bit you are interested in of better still find a choke capable of carrying the charging current and get rid of the largest part of the ripple.

Flux

[ Parent ]

Re: DC current measurement problem (3.00 / 0) (#11)
by Madscientist267 on Sun Jul 12, 2009 at 12:40:14 PM MST

Throw a scope across the shunt and post a pic. :)

Steve

[ Parent ]

Re: DC current measurement problem (3.00 / 0) (#12)
by boB on Sun Jul 12, 2009 at 12:52:29 PM MST

Remember when we had to watch the "Ohms per Volt" input impedance spec on a VTVM ?

boB

Re: DC current measurement problem (3.00 / 0) (#13)
by TomW on Sun Jul 12, 2009 at 05:15:29 PM MST

boB;

Careful, there. bud. Half these guys likely never seen a voltmeter / heater combination. You are giving away your age. And, yes, I remember that little quirk.

Tom

The Truth is the Truth, even if no one believes it; and a lie is a lie even if everyone believes it

[ Parent ]

DC current measurement problem | 13 comments (13 topical)

Menu

� create account
� How to use the board
� FAQs
� search the board
� Google search the board

Login

Make a new account

Username:

Password:

Total Views

��114 Scoop users have viewed this posting.

Related Links

� Also by bob g

You must be a registered user to post here. It's easy and free, and the link is on the upper right side of your page.
All trademarks and copyrights on this page are owned by their respective companies. Postings are owned by the poster, but may be deleted or moved at the ADMIN's sole discretion. The Rest � 2009 Forcefield.
You can Email the board ADMIN here. PLEASE include the username you signed up with!