[Test-Equipment] Problems debugging LCR bridge

[Richard Knoppow]

----- Original Message ----- 
From: "Barry" <n4buq at knology.net>
To: <marconi_test_instruments at yahoogroups.com>; 
<test-equipment at mailman.qth.net>
Sent: Monday, April 23, 2012 7:03 AM
Subject: [Test-Equipment] Problems debugging LCR bridge


> I'm still working on my Marconi TF1313 LCR bridge.  The 
> problem it's exhibiting is in R mode, I can't get a deep 
> null reading like I can in C mode.
>
> I have traced this to a 60 cycle signal that surfaces in R 
> mode even when there is no signal coming from the output 
> of the bridge.
>
> This model uses an EF86 for the first amplifier followed 
> by two more stages of amplification that use both halves 
> of a 12AX7.
>
> I have disconnected everything except the AGC line from 
> the grid of the EF86 and have disconnected all other 
> connections from the bridge (one coupled to the plate of 
> the EF86 and one to the input of the 2nd amplifier stage) 
> to the amplifiers.
>
> When in C (or L) mode, the output of the 3rd amplifier 
> shows a very small 1kc (or 10kc depending on the bridge 
> frequency setting switch) at the output of the final 
> amplifier.  This small signal, once detected, is too small 
> to cause any reading on the meter.
>
> When in R mode, though, there is a significant 60 cycle 
> signal that, once detected, is sufficient (about 4v 
> peak-to-peak before detection) to move the meter well off 
> the NULL point.  When I'm measuring resistances, this 
> makes for weak null indications and that's what I'm trying 
> to fix.
>
> With the only thing connected to the grid of the first amp 
> (the EF86) is the AGC line (that is sourced from the 
> detector via a 10M resistor), I can't figure out why the 
> 60 cycle signal is introduced in R mode and not in C or L 
> mode.
>
> I assume that with nothing else connected to the input, 
> then (in a perfect world) the output should be zero.  With 
> nothing but the AGC line connected, even in an imperfect 
> world, I don't understand why I see the 60 cycle signal in 
> one mode and not the others.  If it were heater noise, 
> then I assume I would see this in either mode.
>
> Any advice as to what to look for?  The heater of the EF86 
> has a hum balancing pot and I can somewhat minimize the 60 
> cycle noise with that, but I can't get it to disappear.
>
> Sorry this ramble a bit but I've done several things to 
> try to isolate the source of this stray signal and wanted 
> to try to paint a clear picture of the current state of 
> the amplifier/detector.
>
> Thanks!
>
> Barry - N4BUQ

     I'm not specifically familiar with this bridge but have 
a couple of general suggestions:
Is there a half-wave rectifier in it somewhere, for instance 
where does the DC for resistance measurements come from?
60hz hum is often from heater to cathode leaks in tubes. 
Substituting tubes is sometimes the only way to find this.
     If the AVC system has a fixed bias source see what that 
looks like.
     Sometimes gassy tubes can cause hum, again the best way 
to find them is substitution.

A good scope will help with this.


--
Richard Knoppow
Los Angeles
WB6KBL
dickburk at ix.netcom.com