Re: Problems with "true" RMS Readings

[Terry Fritz <twftesla-at-xxxxxxxxxx>]

Hi Harvey,

	Both my digital scope (60MHz) and my fancy HP 34401a (200kHz at 6 digits
(I paid more than you did ;-)) do true RMS measurement to a bazaar level.
It takes a while to get used to those readings compared to the old 'peak
level times 0.707' meters.  All these meters use one of the popular true
RMS chips in them which are all very good.  It does not sound like your
meter is bad as long as it is measure something within an order of magnitude.

	""True RMS meters read the equivalent DC voltage that would heat a
resistor up as much as the odd waveform you are measuring.""

	The old meters will just read the peak value and multiply it by 0.707
assuming that it is a sine wave, but they could be very far off while the
true RMS device will be very accurate.  Unfortunately, I must tell you that
you should probably take all your old meters back and keep the new one ;-)
True RMS meters are a wonder that modern electronics has brought us normal
folks.  My scope and HP meter cost about a $1000 each, but 10 years ago
they would have cost 10X that.  For most things, the true RMS value is what
we want.  The old meters are just all they could do with a diode and a
resistor...  Since many things use sine waves that was usually enough.
Doing the true real-time Root Mean Squared calculation requires some pretty
heavy duty A/D converter and fast math power.

	I would be typing all night explaining all the ins-and-outs of RMS
voltages and currents but rest assured that the meter is doing it's job
well and you will find it a valuable tool.  Oscilloscopes that have RMS
readouts really make things obvious since you can see the messy signal and
directly associated it with the RMS value.

	If you are really worried I could test the meter for you but I think it is
fine.  Perhaps others could suggest a good book or web reference about RMS
measurements that would make all this more clear?

Cheers,

	Terry


At 05:50 PM 11/1/2000 -0800, you wrote:
>I recently purchased Radio Shack's Auto Ranging LCD
>Digital Meter. This meter additionally has the quality
>of reading the "true rms". All normal digital meters
>have this rms reading also, and I have used Radio
>Shack meters for years with no problems. But this one
>is giving me a real headache Perhaps it is defective
>but for now I can only ask for a replacement, but I
>have a feeling that it will do the same thing. First
>let me refer to the manuals explanation;
>
>WHAT DOES TRUE RMS MEAN
>RMS stands for Root-Mean-Square. RMS is a standard
>method for determining the effective value of a
>varying or alternating voltage. The RMS value is the
>DC voltage that would produce the same amount of heat
>as the voltage you are measuring, if you were to apply
>both voltages across identical resistors.
>
>The true RMS value is helpful when you need to measure
>an AC voltage that is not a pure sine wave. Most
>meters use average-responding AC converters. These
>meters are calibrated with pure sine waves. If you
>measure a sine wave that does not have harmonic
>distortion, the meter gives an accurate reading. But
>if the sine wave is not pure- there are distortions
>within the waveform, or you measure triangle waves or
>square waves- the average responding meters display an
>incorrect value.
>
>For example a 2 volt peak to peak, 50% duty cycle
>square wave measures correctly as 1.000 VAC on this
>meter. An average responding meter displays 1.111 VAC-
>an error of more than 11 percent!
>
>The maximum frequency your meter can accurately
>measure is 5000 Hz for AC voltage and 1000 hz for AC
>current.
>
>Now these are all good things to know since I am
>taking measurements from a three phase AC converted
>car alternator at what the meter says is 188 hz. 
>Since these amperage figures were derived from
>resonant circuits giving a lot of figures that didnt
>jive I decided to compare the old meter "rms "
>readings from the old 60 hz  resonant amperage
>conductions of the high voltage BRS using 56/59 Henry
>coils. I found great discrepancies not easily
>explained by chop/chop analogies.
>So I went back to measure the amperage input at short
>in the high voltage wall driven BRS. This is been
>measured numerous times by the old Radio Shack digital
>meters at between .25 ma and .3 ma. However the new
>"supposed true rms" reading from the new meter gives
>1.15 ma! If Radio Shack is worried about a 11% error,
>I am more surely worried about this >300% error and
>will use more of those!!!!
>
>In short the meter seems very useless for small
>amperage measurements. As I have complained to the
>store where I purchased this the following;
>
>{I did the following tests by measuring the amperage
>input at 4 times the former values by inputing 4 times
>more the voltage using a 440 transformer input to the
>shorted high voltage 
>BRS}
>
>Taking a known circuit that consumes 1.12 ma by the
>former meters -at- 440 volts the new meter only measures
>.86 ma when the ampearge selection is at the 400 ma
>range. Selecting the range down to the proper setting
>of 4 ma then shows a reading of only .142 ma, a value
>only 12.6 % the value the former Radio Shack "Normal
>and not true" rms reading gives. In short the readings
>the two ranges themselves give do not even resemble
>each other!!! Getting more intrigued I decided to
>measure the maximum wall amperage the high voltage BRS
>gives at 10,000+ volts open condition using a 440
>transformer, the normal meter reads .65 Amps but the
>new quote "true" rms reads .528 A, so this is all BS,
>those values cannot possibly be true. These are some
>problems along the way and I certainly think Radio
>Shack has some serious engineering problems with their
>product, and unless someone could explain what is
>wrong here I am sending them this letter.
>Sincerely Unsatisfied; HDN
>
>
>
>
>
>=====
>Binary Resonant Systemhttp://www.insidetheweb-dot-com/mbs.cgi/mb124201
>
>__________________________________________________
>Do You Yahoo!?
>>From homework help to love advice, Yahoo! Experts has your answer.
>http://experts.yahoo-dot-com/
>