Re: funky voltage readings [solved]

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Terry Fritz" <teslalist-at-qwest-dot-net>

Hi Ken,

I had never used the HP34401 with the HV probe before...  120VAC reads 
61.536mVAC on the HP and 115.37mVAC on a cheap China meter.

R301 and R302 going into U301 (buffer) are 500K each and that is where the 
1M comes from.  The impedance would be "high" without those resistors.  But 
they need to have that low impedance to get the frequency response up into 
the 100kHz region.  They actually use programmable variable caps to 
compensate the meter for higher frequencies and apparently mix in a 50kHz 
calibration signal too.  Page 100 of the 34401A service guide tells about 
it.  somebody spent a lot of time (too much) thinking about this circuitry!

I guess if you wanted a "normal" meter, you would not use the HP 34401 
;-))  Low-level high-frequency AC readings on the HP are also weird 
too.  You can probably just go to 1 digit of precision in that case 
:o)  Otherwise it is a super meter but it does have some odd moments.

The probes sheet at:

http://we.home.agilent-dot-com/cgi-bin/bvpub/agilent/Product/cp_Product.jsp?LANGUAGE_CODE=eng&ID=1000003703%3aepsg%3apro&COUNTRY_CODE=ZZ&CT=0

States: "The Agilent 34136A is only compatible with the Agilent 
34401A.".  I bet they blew it and that was supposed to say "NOT" compatible!!

BTW - www.bkprecision-dot-com  Has their version the Model#PR 28A for 
$59.  About $100 clams less than HP, Tek...  The price of that probe varies 
wildly depending on who is selling it...  Off course, you can just peel of 
the label and replace it with any logo...  I think the fabulous RCA WG-290 
is the father for TV repair use.  At least back to the 1960's...  I am sure 
they are all made in China somewhere now...

There are some very ancient version of this probe:

http://cgi.ebay-dot-com/ws/eBayISAPI.dll?ViewItem&item=2536516507

If you get the RCA version it is 50kV ;-)

http://cgi.ebay-dot-com/ws/eBayISAPI.dll?ViewItem&item=2537428890

And they are only $2.00

http://cgi.ebay-dot-com/ws/eBayISAPI.dll?ViewItem&item=2538531681

Do I hear $1.00  And check out that old box ;-))

http://cgi.ebay-dot-com/ws/eBayISAPI.dll?ViewItem&item=2538974480

Thanks for figuring all this out. ;-))


BTW - I am sorry to report that the Wavetek 27XT has been discontinued 
(very hard to find now) and the "replacement" 37XR does not have the wide 
ranges (10A, 10MHz, 40Mohm) anymore...  Someone in China is making a 
similar version...

http://www.altadox-dot-com/specifications/vc9808_specs.htm

This place will do bulk orders and place your own label of choice on them.

http://www.altadox-dot-com/private_labeling.htm

I think "Tesla" is open :o)  Altadox make a lot of "oddly familiar" looking 
things, but no 40kV probes....


Cheers,

         Terry


At 02:30 PM 6/14/2003 -0500, you wrote:
>Sent: Saturday, June 14, 2003 12:18 PM
>Subject: Re: funky voltage readings
>
>
> > Original poster: "Terry Fritz" <teslalist-at-qwest-dot-net>
> >
> > Hi Ken,
> >
> > I have done that test many many times with similar probes (mine is the B&K
> > version) without problem.  It sounds like you have done all the tricks
>like
> > check it against the 120 VAC line and tried two different NSTs.  I can't
> > think of anything that would cause this.  But, It really "should" work
>just
> > fine...
> >
> > Those "same" probes are also sold by Fluke, B&K, Tektronix...  They are
> > well proven and free from funkyness for tests like this.
>
>I found a manual for the Fluke version of this probe:
>
>http://energybeam-dot-com/resources/Fluke80K40.pdf
>
>It seems Fluke may be the one who makes this probe for everybody else. The
>math on that sheet is right on with my readings that were about 50% of what
>they were (1Mohm meter impedance vs. 10M as in DC, which any DMM can do.)
>
>The schematic shows the probe is just two resistors, and unless you have
>10Mohm meter impedance, you're going to get off readings unless you apply
>some corrective math.
>
>I tried the probe again with only a scope connected, and the voltage reading
>was around 7V for about 7000 on the tranformer, probably higher because the
>scope probe is more than 10Mohm. Amusingly, a hamfest special Fluke 73
>series II meter (inexpensive and older) had direct readings of 7 to 7.3
>volts. When I connected the 34401 benchtop DMM, voltages dropped to about
>3.4 again.
>
>Lessons learned:
>
>- An expensive digital meter can be worthless for AC readings. I'm glad I
>leaned then Agilent 34401 meter has a 1Mohm input impedance in the AC
>ranges, which is poor if not disturbing.
>
>- a $20 hamfest DMM can have superior AC input impedance. Play around, check
>specs.
>
>- Be careful when using big stick high voltage probes, you can can easily
>get trash readings that may be way low. It seems the Agilent site says the
>probe I'm using is only for use with the 34401 meter, and can be used for
>AC, but there is no info about how direct readings in the AC range are
>wrong.
>
>- instead of CE compliance and other nonsense stickers on the probe, there
>should be a warning aside from "use the ground lead."
>
>Maybe somebody has a HP 400 series AC meter, or Leader millivolt meter
>that's for sale. Ebay is a great way to waste a week looking at the HP
>meters before some fool pays too much at the last 15 seconds.
>
>KEN