[Test-Equipment] Re: HP 403B pegging meters

[Jeff Furman]

As an electrical engineer, my background suggests the turnon behavior is 
similar to an underdamped transient response of a second (or higher) order 
system. My 403b shows a few cycles of high amplitude followed by 
additional cycles of substantially reduced amplitude. I convinced 
myself that this oscillation is exponentially damped, as shown in every 
controls textbook.
The turnon capacitor charging currents take long enough that they 
initially confound the actual 
small signal response. The long time constants needed for 5 Hz response 
make for long settling time. I haven't done this, but I suspect a spice 
analysis will show the same behavior.
I replaced all the mercury cells with the nearest equivalent strings of aa 
alkalines; I decided rebiasing things wasn't strictly necessary. A purist 
might want to maintain the transistor operating points via bias circuit 
changes, or use outboard low dropout regulators to duplicate the mercury 
cells' voltages (and voltage stability over discharge lifetime.)

Jeff Furman AD6MX

>
> Message: 1
> Date: Thu, 13 Mar 2008 09:46:47 -0400
> From: "Grant, Ken" <Ken.Grant at uhn.on.ca>
> Subject: [Test-Equipment] HP 403B pegging meters
> To: <test-equipment at mailman.qth.net>
> Message-ID:
> 	<2B1B8F53103478458FAAF8DBB14B345B0418B11B at UHNVMAIL002.uhn.ca>
> Content-Type: text/plain;	charset="iso-8859-1"
>
> When I first got mine, and not knowing better, I thought that the pegging
> meter phenomenon was a sign that the instrument was broken.
>
> So, I disconnected the original meter ( 100 uA FSD ) and, in its place,
> hooked up an external 0-500 uA meter, allowing for different meter internal
> resistances.
>
> The external meter would go over to about 400 uA on power-up, settle
> towards zero after a few seconds, then do it again before finally zeroing.
> This seems to be normal.
>
> Here's my best guess as to what's happening.  There's a big 1200 microfarad
> de-coupling cap in the DC negative feedback loop. This is probably necessary
> in order to get the frequency response down to 5 Hz. Until it charges up (
> thorugh an 8.2K resistor ) and the loop can correct itself, theres an
> initial rush of current through the meter movement. I still can't account
> for the second 'peg', though.
>
> It would be interesting to know if anybody has devised a way to prevent the
> pegging. Perhaps using some component or trick that wasn't available back
> when the instrument was designed?
>
> Ken
>