Most Accurate Way to Measure Power

ip

ng

the

er.

ies.

There's another trick for efficiency measurements; use the same DMM, front = therminals for the input, rear terminals for the output, or just swap the l= eads around. That way any error or slight drift of the DMM has less effect = on the measurement if the input/output figures are not to far from each oth= er.

Cheers

Klaus

He posted this:

How about using an incandescent lamp and a photocell? Get a reference point for the unknown DC+AC signal then substitute pure DC + measure current. Or measure the temperature?

How about the light output. If you used a relay to chop between the unknown and a known pure DC level at a rate allowing the illumination to stabilize between the two sources. Then you just trim the DC until there is no change. Use a photo transistor or diode to measure the light output and see the balance point. It should make a fairly large change with a very small DC change (power). Use a very small filament to get a fast time constant.

Just trying to bring an electro-optical solution to the table :)

tm

Most (digital) scopes have limited accuracy due to low number of bits. Most Tek stuff is 8 bits with averaging to get ~9 bits, there are some LeCroy that have 10 and now 12 bit ADCs. If you can get a scope with a 12-bit ADC, that should help. Multiplying an 8-bit sample times an

8-bit sample, it is going to be real hard to get much better than 1% accuracy. Hmm, that 2.5 V P-P ripple could sure complicate the measurement.

Jon

Sounds really accurate >:-} ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

.

=20

=20

We usually think of an ammeter as having zero resistance. And, a voltmeter as having infinite (very high) resistance. But, those aren't always neglig= ible, and your 'scope and metering setups might be the source of the discrepancy. At 100 Hz, I'd think only the resistive impedances need be considered.

hip

ow

S

ing

, the

a

Getting Mike Engelhardt to look after it seems to have been a another good move.

-- Bill Sloman, Nijmegen

chip

nning

ct, the

=A0 =A0 =A0 =A0 =A0...Jim Thompson

o u

uter.

nt

ncies.

e n

=A0 =A0 =A0 ...Jim Thompson

=A0 =A0 =A0| =A0 =A0mens =A0 =A0 |

=A0 =A0 | =A0 =A0 et =A0 =A0 =A0|

=A0 =A0|

=A0 =A0 =A0 =A0 |

=A0|

=A0 |

d.

d.

nt.

I didn't do any serious soldering before I had to build my own electronics for my Ph.D. project. Childhood electronics obviously helps, but it's not strictly necessary. We had a fair number of impractical graduate students - like the one who wanted to dismantle a vibrating reed electrometer, so that he could measure the voltage drop over the 1G resistor from the electrometer with a 20k moving coil multimeter - but some of us had some manual skills.

-- Bill Sloman, Nijmegen

Unfortunately you have form the product of voltage and current - to get power - before you average.

e

An AD734 could do a much better job, but Jim is unlikely use one of Barry Gilbert's integrated circuits, and he's probably not up to getting the necessary analog signals to the AD734 accurately enough to take advantage of its accuracy.

And you'd have to low pass filter the product of current and voltage. Much too much analog design for an integrated circuit designer.

-- Bill Sloman, Nijmegen

A Fluke 8920A will read DC+RMS of the AC component. It uses a bolometer for the RMS conversion and will display ac or AC+DC to 20 MHz on most ranges.

All for it! In general you need the resistance to be adjustable, though.

That's a bit like the half-shade method that was used from the 1930s for polarimetry. In general it's easier to do electrically nowadays, unless the original signal is optical.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

Gossen Metrawatt has one (the Metrahit Energy), and there's also a Fluke that does it, iirc.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

Thanks! I'll so advise the client. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

Why can't he just use another of the same meters to measure output power? Something like these, perhaps:

[snip]

Good question. I had the same thought. Can those instruments work on DC as well as AC I/O? ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

How about a Valhalla 2101?

It'll do both. There's one listed on used line for $500.

Thanks for the lead! ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

Valhalla's good stuff I hear...

Or, maybe even better yet... (certainly more expensive though)...

These are typically what we have used in the inverter / charger industry.

boB