David,
Yep, I know slowly....Looking for a time where I can spend about
4 hours on it. Otherwise I just set everything up and then I have to
take care of my toddler. Almost in her terrible two's but, she's acting
older than her age.
Anyway, what is the best way to lower those third harmonics (60Hz source
and 180Hz offsets? Are these a result of AGC and/or Demodulators on the
A4 board? Is it filterable? IN the HP tech notes they recommend putting
200 ohm resistors in front of L1 and L2 which feed each demodulator.
That would create an RC filter network but not sure if that would be in the
180 Hz harmonic region.
In the mean time, did you see Richard's posts to me about removing them from the circuit...?
Post # 406, 440
Hope every one enjoyed their Holiday, Winterbreak, Christmas, Hanukah, etc.
Last edited:
For low leakage Tanalaum is the first choice. There are specific low leakage eletrolytic caps targeted at Tantalum replacements.
Thanks Demian,
I have to look to find one. It looked like a electrolytic from Sprague,
from what I recall. Tant's of that value run 230.00 to $550.00 at mouser.
In the mean time I'm stuck with what I put in there. Almost wondering
if I could put two electrolytics back to back with a film bypass.
Not sure how that affects leakage.
Thinking what I have now is a Panasonic FC with a small wima film cap
should prevent leakage.
I have to look to find one. It looked like a electrolytic from Sprague,
from what I recall. Tant's of that value run 230.00 to $550.00 at mouser.
In the mean time I'm stuck with what I put in there. Almost wondering
if I could put two electrolytics back to back with a film bypass.
Not sure how that affects leakage.
Thinking what I have now is a Panasonic FC with a small wima film cap
should prevent leakage.
Paralleling does not reduce leakage -- the leakage will be determined by the big electrolytic. Measuring the leakage is important, and every cap will be different, so for low leakage applications, they are measured/selected.
However,
I could use a small value cap .1uf on the
pin or in series with the feed to the
quad opamp. This should block
the DC from the causing distortion
on the opamp.
OR
Use that small caps then wire the other
side to ground ground. w/ wo a diode
to block other signals from entering
the opamp.
There is another cap that has even
lower specs. This is A2C23 for
the meter. It originally was a 10uf
cap, but now is a 15uf cap with a
maximum leakage spec of 0.05uA
after 2 minutes with 14VDC applied to it.
"If the dc leakage through C23 exceeds .05uA
a dc voltage offset occurs at pin 4 of U5 causing
and erroneous meter reading.
I'm trying, this is tricky stuff and standard parts just
don't seem to work very well. Heck I even cleaned off most of the bench
looking for that 650uf cap. A lot of tants I've seen don't
even meet that spec. The Panasonic FC I used has a spec
of 3uA leakage. : (
Now I'm wondering if I should have used switch mode supply
capacitors for the mains filters in the power supply.
...and the beat goes on.
I've found some schematics for measuring leakage current
and in general it seems that it is a black art.
Cheers,
I could use a small value cap .1uf on the
pin or in series with the feed to the
quad opamp. This should block
the DC from the causing distortion
on the opamp.
OR
Use that small caps then wire the other
side to ground ground. w/ wo a diode
to block other signals from entering
the opamp.
There is another cap that has even
lower specs. This is A2C23 for
the meter. It originally was a 10uf
cap, but now is a 15uf cap with a
maximum leakage spec of 0.05uA
after 2 minutes with 14VDC applied to it.
"If the dc leakage through C23 exceeds .05uA
a dc voltage offset occurs at pin 4 of U5 causing
and erroneous meter reading.
I'm trying, this is tricky stuff and standard parts just
don't seem to work very well. Heck I even cleaned off most of the bench
looking for that 650uf cap. A lot of tants I've seen don't
even meet that spec. The Panasonic FC I used has a spec
of 3uA leakage. : (
Now I'm wondering if I should have used switch mode supply
capacitors for the mains filters in the power supply.
...and the beat goes on.
I've found some schematics for measuring leakage current
and in general it seems that it is a black art.
Cheers,
Do not modify circuits whose function you don't really understand. David has been repeating this point for a while and he is correct. HP knew a great deal about making this thing work, so just follow their lead and do your best to do what they recommend and get everything working to spec before you start messing around. We all want to see you succeed, but there is really a lot you don't grok yet, and it's causing you all kinds of problems. Go back to the beginning, start over, and follow the rules, at least for now.
Click on this following link, and read it carefully until you think you understand it. Then carefully read the section on leakage. Conrad is a great guy and knows what he is talking about.
Checking Caps
Checking Caps
After being away from the 339A for awhile... I came back to it and it appears that the analyzer's minimum reading may not be noise limited but is seeing the notch frequency residual. How can the notch be made deeper... 10+dB would be good.
THx-RNMarsh
THx-RNMarsh
Last edited:
The notch depth can be increase by increasing the gain of A4U1.
Keep in mind that removing more fundamental may result in the harmonics having greater magnitude than the fundamental. The phase detector may lock onto a harmonic rather than the fundamental. If this happens then the opposite of what you're trying to achieve will occur. Fundamental suppression will be lost.
I needed a break also.
So I got bogged down in trying to make an octopus rig/curve
tracer. Still trying to get it to work properly. That said,
Richard, do you think that might be what is going on in
my case?
It's a great new year, wish everyone the best.
cheers,
So I got bogged down in trying to make an octopus rig/curve
tracer. Still trying to get it to work properly. That said,
Richard, do you think that might be what is going on in
my case?
It's a great new year, wish everyone the best.
cheers,
@SYNC -- My 2 cents, no -- that's not what's happening in your unit -- you have gone back to the original gain for A4U1, right?. Your oscillator is not working correctly.
As I recall, you changed its gain resistor to 200 Ohms (5X increase). But then there was something about scaling being off... Could the gain for proper scale be similarly reduced at the output past the filters (U4) ?
THx-RNMarsh
I don't remember the scaling being off from that change. The tracking was lost for the said reason. You could try a lesser gain increase. X2, 3X.
@y'all, no not yet.
I've made no additional changes nor mods, nor undoing since
my last post a couple of weeks ago. Been in a holding pattern.
I think I'll put a pot in there with a smaller resistor in series so that
I can change the gain of the A4U1 for now. Nothing fancy while
I fiddle with the circuit.
Some of the issues I think I'm having are feeding the
oscillator back on the A1 board.
@ Richard, what electrolytic cap are you using there
at the oscillator and JFET on the A1 board? I think it is A1C30?
Brand, Series, Value? Is that a Nichicon polarized electrolytic?
Goal: pinpoint/fix the oscillator problem.
Then I can get off the waterboard. : )
I've made no additional changes nor mods, nor undoing since
my last post a couple of weeks ago. Been in a holding pattern.
I think I'll put a pot in there with a smaller resistor in series so that
I can change the gain of the A4U1 for now. Nothing fancy while
I fiddle with the circuit.
Some of the issues I think I'm having are feeding the
oscillator back on the A1 board.
@ Richard, what electrolytic cap are you using there
at the oscillator and JFET on the A1 board? I think it is A1C30?
Brand, Series, Value? Is that a Nichicon polarized electrolytic?
Goal: pinpoint/fix the oscillator problem.
Then I can get off the waterboard. : )
Tried just a X2 gain change and no affect on thd+n result ----- so I guess it only leaves noise. So, still need to find way(s) to lower noise.... might be circuit R's limited and not old opamp's noise. but need to test that also.
THx-RNMarsh
All the caps that have no DC on them are replaced with low Z/ESR Bipolar types. That one is 100Mfd/100Vdc 105C. But I dont see the brand name on it and dont remember, either.
THx-RNMarsh