# 120V accurate current source - how?

Discussion in 'Electronic Basics' started by Shawn, Sep 6, 2005.

1. ### ShawnGuest

I'm seeking opinions on the best way to generate an accurate current
at 120V AC. I need to calibrate a power meter design, and it calls
for the ability to accurately generate specific currents between 1A
and 5A. I've considered the use of power resistors, but 5A would
require a 600W resistor, which, while not impossible to find, is a bit
drastic. I also considered six 100W bulbs, but since it's a power
meter I need a degree of accuracy greater than the tolerance of a
light bulb.

Any suggestions on other possibilities I might explore, or old
equipment that I might be able to pick up from eBay?

Thanks all!

2. ### Dan HollandsGuest

Power meters generally have 2 inputs, one to measure voltage and one to
measure current.
Calibrating them is usually done with a variable 120V voltage into the high
impedance voltage input and a variable current input at a low AC voltage
into the low impedance - close to zero - current input. It is not necessary
to have the current input come from a 120 volt source. Commericial current
sources suitable for calibration are available and are expensive. Ideally
the phase angle between the voltage and current sources is also adjustable
to test the effect of inductive or capacitive loads.

Dan
--
Dan Hollands
1120 S Creek Dr
Webster NY 14580

www.QuickScoreRace.com

3. ### ChrisGuest

Hi, Shawn. If you need 5A at 120V, you're going to have to dissipate
600 watts. No way around it.

Current calibrators are drastically expensive -- that shouldn't be
considered unless you have a really hefty budget for this job.

Light bulbs are a no-go here. At rated voltage, their wattage can vary
by more than 20%. Also, their resistance is non-linear -- as the
applied voltage decreases, the temperature of the filament decreases
and the resistance decreases.

You might as well use power resistors. This is what they're made to
do. And they needn't be expensive, either. Nebraska Sales and Surplus
has 120 ohm, 120 watt adjustable power resistors available in quantity
for \$9.00 USD each.

http://www.surplussales.com/Resistors/WireWound/WW111-250.html

String 5 of these in parallel, give each one it's own series junkbox
switch, and voilla! You have a 1-2-3-4-5 amp 120VAC load. You won't
find an AC current calibrator at ebay for \$45 bucks. I would guess
shipping alone would probably be more. ;-)

If you want more accuracy (line voltage typically varies a bit, and
power wirewound resistors are +/- 10% unless otherwise specified), you
can still use these resistors. Just borrow a big Variac, and get a
resistance meter that can measure accurately. Adjust each resistor to
115 ohms +/- 1%, and then apply a measured 115VAC through the Variac.
This will give you what you need, with good accuracy.

Before you start, though, find out if you can do calibration with only

Good luck
Chris

4. ### john jardineGuest

I needed to do something similar without spending any money. Used the
workshop Variac (every workshop should have a Variac) and 2kW electric
fanheater (every workshop should ... ).
Monitored the current with a DVM.
regards
john

5. ### mikeGuest

I hate to rain on your parade, but that's what I do best.

1:What's your definition of "accurate"???? Just how accurate?

2:What's your definition of "power"??? If you expect to measure RMS
power but calibrate it using some kind of (non-true RMS) voltage
measurement and a resistor...see #1.

3: If you expect to get a pure sine wave out of the wall plug...see #1.

4: Do you expect your measurement bo be accurate with reactive
loads...over what PF range? see #1.

5: How about non-sinusoidal currents? Range of peak to average ratio?
See #1.

The theory is trivial. Acutally (accurately) measuring power over
a range of non-ideal conditions is not so easy.

mike

--
Return address is VALID but some sites block emails
..
Wanted, PCMCIA SCSI Card for HP m820 CDRW.
FS 500MHz Tek DSOscilloscope TDS540 Make Offer
Wanted 12" LCD for Compaq Armada 7770MT.
Bunch of stuff For Sale and Wanted at the link below.
MAKE THE OBVIOUS CHANGES TO THE LINK
ht<removethis>tp://www.geocities.com/SiliconValley/Monitor/4710/

6. ### ehsjrGuest

Incorporate a 10:1 shunt in your power meter design.
Calibrate the circuit at at 12 volts and 1/2 watts.
For 120 volt measurements, use the shunt.

Ed

7. ### Bob MastaGuest

Ecellent idea. If the OP doesn't have a DVM for the rated
current, he can always use a known small resistance
shunt and measure the drop to determine current.

Also, note that even without the variac there are
devious ways to handle high power in lower-power
resistors. First off, you can apply the power just
for the second it takes to get a reading, then give
them a long cool-down. Better yet, immerse the
resistors in a paint can full of oil to dissipate the
heat. This trick was used back in the '60s by
hams needing dummy loads for transmitters.
There are of course limits to just how far you can
go with this... don't use a 1/4 watt resistor and
expect it to work. But if you can get something
in the 10s of watts, I think you will be fine.

Best regards,

Bob Masta