You have a cracked trace or pad. There is no way for there to that much resistance in a low Z trace. Power down and wait for the caps to bleed. Measure the resistance between TP1 and pin 6 of A1U1. The trim pot R31 is usually set about mid way. It is easy to crack a pad or crack a through hole when removing a part on a double sided board.
Also, if R30 is at its lowest resistance, then that lowers the amp gain, which will reduce the output, not raise it. I agree with David -- I think you have board damage of some kind.
Hmmmm,
I measured 105.8 ohms.
This is the value of the resistor across the trace that I cut.
On the output A1u1, pin six.
I measured 105.8 ohms.
This is the value of the resistor across the trace that I cut.
On the output A1u1, pin six.
Why did you cut the trace and insert a resistor?
Take it out. Then recalibrate.
The 1468 has a peak around 7meg IIRC. At any rate you have to return FB at high frequency to null the input capacitance.
If you insert a resistor this won't happen unless there is a FB cap about 7pF. But don't do this.
Just remove the resistor and repair the trace. There is enough FB through the bridged T for stability.
The op amp is probably oscillating at high frequency.
Last edited:
David,
I think that was one of your earlier recommendations.
I think RichEEM, recommended 47 ohms that he used.
If it is oscillating at HF...it's above what my scope can
see, or my ability to see it on the scope. The sine waves
at various frequencies look like they are supposed to.
The sine wave does look clean and not fuzzy.
Unless it is so slight that I can't see it.
I performed the Oscillator performance Tests
from the manual, 4-15, 4-16, 4-17 &4-18.
From Table 4-3 of the manual:
Out freq out level
10Hz 2.930 - 3.070 in spec
20Hz 2.965 - 3.035 in spec
100Hz 2.965 - 3.035 in spec
10kHz 2.965 - 3.035 LOW 1.98v
20kHz 2.965 - 3.035 LOW 1.98v
110KHz 2.930 - 3.070 LOW 1.98v
The higher frequencies are a problem...but that might be
not able to properly tune with the adjustable cap.
Root cause may be HF Oscillation.
I'll remove resistor and repair the trace.
I can also plug the LME chip in the U12A,B,C,D..and
see if the voltages change. It's in a socket so not hard to
do.
@RichEEM, I"ve already changed the resistors around the Jfet to pots.
Those are R51 and R50, which are 2K each.
I think that was one of your earlier recommendations.
I think RichEEM, recommended 47 ohms that he used.
If it is oscillating at HF...it's above what my scope can
see, or my ability to see it on the scope. The sine waves
at various frequencies look like they are supposed to.
The sine wave does look clean and not fuzzy.
Unless it is so slight that I can't see it.
I performed the Oscillator performance Tests
from the manual, 4-15, 4-16, 4-17 &4-18.
From Table 4-3 of the manual:
Out freq out level
10Hz 2.930 - 3.070 in spec
20Hz 2.965 - 3.035 in spec
100Hz 2.965 - 3.035 in spec
10kHz 2.965 - 3.035 LOW 1.98v
20kHz 2.965 - 3.035 LOW 1.98v
110KHz 2.930 - 3.070 LOW 1.98v
The higher frequencies are a problem...but that might be
not able to properly tune with the adjustable cap.
Root cause may be HF Oscillation.
I'll remove resistor and repair the trace.
I can also plug the LME chip in the U12A,B,C,D..and
see if the voltages change. It's in a socket so not hard to
do.
@RichEEM, I"ve already changed the resistors around the Jfet to pots.
Those are R51 and R50, which are 2K each.
For those following along:
To recap here is some of what I've done and undone so far:
In pic above Left opamp is U3, U1, U2 Righ quad op amp.
The are OPA134, LT1468, LME49740 respectively.
All the open holes above the left and center op amps
is compensation that's been removed.
Center op amp:
Upper right white box PolyCarb, PolyStyrene stacked cap (C10)
replaced with small multi layer C0G NP0 type.
C47 (two stacked silver mica's replaced with)
one 250pf mica.
R50,R51 replaced with yellow multi turn pots.
R30 replaced with blue multi turn pot.
Left OpAmp:
C49 larger PP film replaced with C0G NP0 type.
C40, C41 replaced with Bipolar 100uf type,
same as C30 (by U1).
All the open holes above the center op amp
are compensation that's been removed.
To recap here is some of what I've done and undone so far:
In pic above Left opamp is U3, U1, U2 Righ quad op amp.
The are OPA134, LT1468, LME49740 respectively.
All the open holes above the left and center op amps
is compensation that's been removed.
Center op amp:
Upper right white box PolyCarb, PolyStyrene stacked cap (C10)
replaced with small multi layer C0G NP0 type.
C47 (two stacked silver mica's replaced with)
one 250pf mica.
R50,R51 replaced with yellow multi turn pots.
R30 replaced with blue multi turn pot.
Left OpAmp:
C49 larger PP film replaced with C0G NP0 type.
C40, C41 replaced with Bipolar 100uf type,
same as C30 (by U1).
All the open holes above the center op amp
are compensation that's been removed.
I think I tried that but withdrew it. It's not necessary.
Leave it out until the problem is solved.
Need to start from a clean slate.
Last edited:
For those following along:
To recap here is some of what I've done and undone so far:
In pic above Left opamp is U3, U1, U2 Righ quad op amp.
The are OPA134, LT1468, LME49740 respectively.
All the open holes above the left and center op amps
is compensation that's been removed.
Center op amp:
Upper right white box PolyCarb, PolyStyrene stacked cap (C10)
replaced with small multi layer C0G NP0 type.
C47 (two stacked silver mica's replaced with)
one 250pf mica.
R50,R51 replaced with yellow multi turn pots.
R30 replaced with blue multi turn pot.
Left OpAmp:
C49 larger PP film replaced with C0G NP0 type.
C40, C41 replaced with Bipolar 100uf type,
same as C30 (by U1).
All the open holes above the center op amp
are compensation that's been removed.
Have you swapped your Jfet with the spare?
Anyway how are things now?
Don't have a spare yet.
Going to order tomorrow, should be
here tuesday.
I"ll have to look and see is there another
JFET on the A4q3 (removable board) Should
work 2n4392. 😉 That should answer the JFET question.
Will sub it in and see.
But maybe not the voltage question.
Last edited:
Well, the problem persists and is made worse by
the 2n4392 that I swapped in. In the R30 pot
There is very little variation with the pot, that is
only about three of twenty turns is required
to go from 9.8 volts and 6.4 volts. at A1TP1.
At 9.8 volts output, the magnitude of harmonics
is about 30 dB down from the fundamental
instead of 60 to 80 dB down.
the 2n4392 that I swapped in. In the R30 pot
There is very little variation with the pot, that is
only about three of twenty turns is required
to go from 9.8 volts and 6.4 volts. at A1TP1.
At 9.8 volts output, the magnitude of harmonics
is about 30 dB down from the fundamental
instead of 60 to 80 dB down.
It was working but had gremlins. Sometime it would be intermittent.
To verify your other question/statement, yes, the voltage at A1TP1 changes
with range as follows:
10Hz 9.98V
100Hz 9.93V
1kHz 9.86V
10kHz 8.49V
So it is pretty close. Probs as all frequencies:
png
To verify your other question/statement, yes, the voltage at A1TP1 changes
with range as follows:
10Hz 9.98V
100Hz 9.93V
1kHz 9.86V
10kHz 8.49V
So it is pretty close. Probs as all frequencies:
Last edited:
@Sync --
As you vary R30, is the DC at the gate of the FET changing? We need to know if the AGC is doing anything. Target is about -0.5V at 1kHz. The DC range will vary with frequency setting -- mine hovers around between -0.3 to -0.6 on the lowest ranges. The A1U1 output should be close to 20Vp-p or about 6.3VRMS give or take a few hundred millivolts. Your 9V readings are what? RMS? P-p? Can't make any sense of this.
As you vary R30, is the DC at the gate of the FET changing? We need to know if the AGC is doing anything. Target is about -0.5V at 1kHz. The DC range will vary with frequency setting -- mine hovers around between -0.3 to -0.6 on the lowest ranges. The A1U1 output should be close to 20Vp-p or about 6.3VRMS give or take a few hundred millivolts. Your 9V readings are what? RMS? P-p? Can't make any sense of this.
Last edited:
Going back through everything at least on
A1U1 board, I pulled the pots and redid them.
Hopefully I didn't cook the JFET as when
I was resoldering R30, instead I sat on the JFET.
Fun, turning it from side to side.
I removed the pot at A150 and replaced
it with the 2K resistor. I added a 5.6K resistor
in parallel with the other pot at R51 so it adjusts
from about 1k to 2.5K.
Mechanically it seems to be working better.
I need to check pin six on A1u1.
RichEEM, I'll let you know...I think JFET is
changing DC with the turn of R30. Not sure
how much though...will get back with you on that.
I'm doing the best I can..and sometimes I don't know
what should be measured, with what parameters either.
This is all research for me, as Einstein would say,
if we knew what we were doing we wouldn't call
it research.
And
If you think you are uncertain and impatient, frustrated
please know if you multiply it by ten or one hundred
then you get to where I am. I understand many of
you have done this kind of stuff all your lives.
I started learning ELE in my fourties and I'm fast
approaching the sixties....So if I'm kinda slow at
picking up things please understand I don't mean
to be. I"m just trying to keep my mind active and
learn new things.
I'm trying to learn what y'all have forgotten.
A1U1 board, I pulled the pots and redid them.
Hopefully I didn't cook the JFET as when
I was resoldering R30, instead I sat on the JFET.
Fun, turning it from side to side.
I removed the pot at A150 and replaced
it with the 2K resistor. I added a 5.6K resistor
in parallel with the other pot at R51 so it adjusts
from about 1k to 2.5K.
Mechanically it seems to be working better.
I need to check pin six on A1u1.
RichEEM, I'll let you know...I think JFET is
changing DC with the turn of R30. Not sure
how much though...will get back with you on that.
I'm doing the best I can..and sometimes I don't know
what should be measured, with what parameters either.
This is all research for me, as Einstein would say,
if we knew what we were doing we wouldn't call
it research.
And
If you think you are uncertain and impatient, frustrated
please know if you multiply it by ten or one hundred
then you get to where I am. I understand many of
you have done this kind of stuff all your lives.
I started learning ELE in my fourties and I'm fast
approaching the sixties....So if I'm kinda slow at
picking up things please understand I don't mean
to be. I"m just trying to keep my mind active and
learn new things.
I'm trying to learn what y'all have forgotten.
Last edited:
HP 339A Update
I've had a few problems that I had to work out.
I had a spike right at the top of the sine waves.
Then some other stuff trying to fix the board
with an unknown trace of through hole problem.
I have a distorted power supply.
Took a look at the -15 volt rail via A1TP -15,
and find some fuzz and AC. It is unstable
at 10 Hz. On the QA400 Plot, still not so great.
I've had a few problems that I had to work out.
I had a spike right at the top of the sine waves.
Then some other stuff trying to fix the board
with an unknown trace of through hole problem.
I have a distorted power supply.
Took a look at the -15 volt rail via A1TP -15,
and find some fuzz and AC. It is unstable
at 10 Hz. On the QA400 Plot, still not so great.
Sometimes a spike at the top (or bottom) peak indicates a burst of oscillation as the wave begins to run into the power supply, which is called clipping. As the amp runs into the rail, the gain-controlling negative feedback disappears, letting the amp go a little (or a lot) nuts. When working correctly, the output will be sufficiently far under each rail voltage -- by about 5V -- that there is no clipping, even when changing ranges, etc. Remember the output is 10 peak plus and minus, and the rails are at 15V plus and minus, so there is headroom for the amp to swing both directions. If the amp output is offset away from zero VDC, then all bets are off, because the amp can clip at one peak of the other even though the peak signals are OK.
This may not be what is happening in this case, but it is something to watch for. The spike can also be caused by the gain-control FET not working or running out of its working range. If you have a good scope, set it for DC and use it to monitor the gate of the FET so you can see the DC shift as well as AC ripple, if any, while you adjust R30. I don't think you have any AGC working, but without being able to see the levels at the FET gate and drain, it's hard to know.
This may not be what is happening in this case, but it is something to watch for. The spike can also be caused by the gain-control FET not working or running out of its working range. If you have a good scope, set it for DC and use it to monitor the gate of the FET so you can see the DC shift as well as AC ripple, if any, while you adjust R30. I don't think you have any AGC working, but without being able to see the levels at the FET gate and drain, it's hard to know.
