How To Convert DC panel meter to read AC AMPs?

Discussion in 'Electronic Basics' started by [email protected], Feb 2, 2007.

1. Guest

I have beautiful DC panel meters that currently reads 0-5Volts DC by
using an in-series resistor. I want to use this meter to instead read
0-5 AC AMPS. This is to help me judge excessive power consumption on
my train layout.

I have a .15 Ohm shunt resistor in parallel with the meter, but as you
might expect, the meter needle vibrates due to the AC. So, I put a
diode in series with the meter to feed it DC. This works. To help
keep the meter as sensitive as possible, I put a small electrolytic
capacitor in parallel with the meter. This gets me closer to measuring
Peak voltage drop across the shunt instead of RMS.

Is this the best way to do this? I tried a full-bridge rectifier but
found that I had even more voltage drop across the bridge. Are there
low-voltage drop diodes I can use that will work better here?

thanks,

2. JamieGuest

Look at a precision rectifier.

3. Ross HerbertGuest

Firstly, you will need to use a bridge rectifier to convert the AC to
DC. Then you will need to use a shunt in series with the AC line as a
means of detecting a suitable voltage range to drive the meter.

Read Rod Elliot's article on meters and shunts.
http://sound.westhost.com/articles/meters.htm

4. ahwiiGuest

Thanks for the references but building precision rectifiers seem a bit
overkill. Are there any pre-built precision rectifiers that come on a
single chip?

I really didn't want to make this such a big project that I would have
to breadboard the supporting circuitry. I was hoping a simple diode
would do the trick. That is why I tried a half phase, single diode
test using 1N914/1N4148 silicon switching diodes. I made up a bridge
rectifier (made of the same diodes) but this dropped the sensitive of
the meter quite a bit. Using a standard 50V silicon bridge rectifier
delivered almost no voltage to the meter! I need .9mA for full meter
deflection.

Building two op amp circuits (one for each supply that I am measuring)
and feeding the Op Amps their 5v source is more than I wanted to take
on. The power supply is delivering 17V AC at about 5AMPs to my digital
controllers.

thanks,

5. kellGuest

You're only getting about 1 volt peak across the sense resistor at
maximum current, so you don't have much to work with.
Schottky diodes or germanium diodes have lower voltage drops than
standard silicon. Using schottky or germanium bridge with a double
diode drop might still be low enough to give you a reading, but I
don't see why you want to use a bridge. Wouldn't the extra drop
affect the accuracy of your meter in the lower range? You're getting
very little voltage from that sense resistor.
Until there's enough current through your sense resistor to exceed the
drop of your diode or diodes, the meter will read zero. The more
diodes you use, the more inaccurate. If it was my railroad and I
wanted the meter accurate I'd probably use a single shottky or
germanium diode.
You could also look into using a current transformer instead of a
sense resistor. Theoretically at least, you could get a linear, full-
scale reading on your 5 volt meter. The devil is in the details and I
haven't used current transformers, so I'll have to defer to persons
with more expertise on how to execute that.

6. defaultGuest

If you can sacrifice some more voltage in the shunt here's a simple
idea:

Put a pair or silicon back to back diodes in series with the shunt
resistor and connect the meter to the shunt resistor and diode pair.
The diodes have to be rated high enough to handle 1/2 the AC current
that you will use - and there should be a fuse or some current
limiting (or just use a pair of 20 amp diodes)

That will drop Point .6 volts AC and compensate for the diode drop in
the meter's rectifier diode. You will have six tenths of a volt less
going out to the train layout - if that's tolerable to you.

Likewise you could use two back to back pairs and compensate for a FWB
rectifiers and waste 1.2 volts in the compensator diodes.

****

Another different and slightly more complicated approach might be to
bias your meter rectifier on all the time with a small current
returned to the opposite AC supply rail.

connected to the other end of the shunt (same place the minus of the
meter is connected). Your basic set up.

Add another diode and resistor connected in series from the junction
of the meter and cathode of the meter rectifier and the opposite AC
lead. The cathode of the compensation diode goes to the AC lead.
Resistor value? That depends on how much current your meter needs but
a watt or two of power dissipation should be plenty. Both diodes
(meter rectifier and compensator diodes have to be able to handle the
compensation current - so a 1N4148 might not be the best meter
rectifier in that app)

I think that idea would work, but would welcome any criticism or
critique, if some guru would like to lambaste me. The first idea
would work the second idea doesn't lower the voltage out to the track
so may be more appealing.

7. Homer J SimpsonGuest

Basically you want a circuit to let you read a very low AC voltage with a DC
meter. This is difficult to do. The original method was to use a current
transformer, however today it is more usual to use a 'precision rectifier'
circuit. See http://sound.westhost.com/appnotes/an001.htm for an example.

You would, of course, be better to purchase a 5 A meter.

8. defaultGuest

Way back, during the dawn of time (1930) they used copper oxide
rectifiers for meters. Very low forward drop and very low (6 V) PIV.
They were used for meters, battery chargers, and radio detection.
Reasonably linear for meter applications since most of the meters were
only good for a 5% accuracy.

I scoured the web looking for a supplier for them and didn't turn up
any - but they still make moving coil AC meters that use them.

I have a few moving vane meters I got surplus. They are repair stock
for an old X-ray machine and were made by G.E.. Iron vane meters can
be very accurate and they don't care if the power is AC or DC. Mine
have something like 15 turns of wire on the coil and were intended as
current meters. I suspect all the bad press iron vane meters received
is due to the cheap battery checker and charger meters that were
produced with 20% accuracy.

Nice thing about old iron vane meters is you can take them apart and
rewind the coils for different uses and change the scale markings.

9. Ross HerbertGuest

should have directed this response to Jamie who was the one suggesting
precision rectifiers.

10. Ross HerbertGuest

copper and selenium oxide rectifiers.

11. ahwiiGuest

I just bought a few 1N5817 Schottky diodes on eBay (DigiKey way too
expensive and my local electronics shop had none for the taking - at
least that I found) and will give these a try instead of the silicon
diodes I have been trying. I guess the small transformer might be the
next step to try. Yeah, I need to keep the voltage drop across the
shunt resistor pretty minimal as the DCC controller wants to have a
pretty steady voltage supply and needs to be able to pull anywhere
from 0 to 5.5AMPS.

thanks,