Whew, just finished reading the entire thread. I scored a 339a at a local Ham Fest for $70! It's super clean on the inside, but had a completely missing OSC output jack. It turned out to be punched through the front panel, and was dangling between the panel and the frame. Not wanting anything to short out, I took it apart before I turned it on. I broke two switches, just by looking at them hard. With the help of DIYA I pulled the front panel and made 2 repairs (the MON output jack was also damaged). However, I missed the tip to make sure and keep the switch rods aligned.
I'm not 100% sure that my switches are mis-aligned. The unit appeared to work (mostly) and I was able to test loopback and an external amplifier; but there were some oddities. At some values, the MON monitor out looked clean on my OScope, but at others it was a golden mess of strange signals. It also appeared that the directional lights for Frequency were trying to get me to go the wrong way (or past the end of the lowest settings). I have only had time to play with it for a couple of hours, so operator error is a big possibility.
I'm not looking for help (yet), I plan to do the entire self-check in the manual before I do anything else. I am enjoying the discussion here, and am trying to learn as much as possible!
Thanks!
p.s. Sync, I used to live in Lewisville, TX (not that far from you, it seems). Now I'm in GA.
I'm not 100% sure that my switches are mis-aligned. The unit appeared to work (mostly) and I was able to test loopback and an external amplifier; but there were some oddities. At some values, the MON monitor out looked clean on my OScope, but at others it was a golden mess of strange signals. It also appeared that the directional lights for Frequency were trying to get me to go the wrong way (or past the end of the lowest settings). I have only had time to play with it for a couple of hours, so operator error is a big possibility.
I'm not looking for help (yet), I plan to do the entire self-check in the manual before I do anything else. I am enjoying the discussion here, and am trying to learn as much as possible!
Thanks!
p.s. Sync, I used to live in Lewisville, TX (not that far from you, it seems). Now I'm in GA.
You might also enjoy this thread.
http://www.diyaudio.com/forums/equi...stortion-audio-range-oscillator-new-post.html
If you think this was a lot to get through wait till you try the above thread.
The 339a mods began at the low-distortion-audio-range-oscillator thread.
Cheers,
Hi Accelr8,
Yep, it is now one big mass of humanity, and will only get worse.
If you get lost on the front panel, with every knob turned
full CCW the set screws should be vertical. If they are off
mis-aligned then someone's been in there and didn't align
them back properly.
Hope that helps.
Yep, it is now one big mass of humanity, and will only get worse.
If you get lost on the front panel, with every knob turned
full CCW the set screws should be vertical. If they are off
mis-aligned then someone's been in there and didn't align
them back properly.
Hope that helps.
waiting for replacement opamps. : (
I still have major oscillations of the meter
at minus 80dB is the worst. Then it goes
to small fast movements on VU then back
to slower larger oscillations on the NORM
selection.
It stops with any of the filters engaged.
I still have major oscillations of the meter
at minus 80dB is the worst. Then it goes
to small fast movements on VU then back
to slower larger oscillations on the NORM
selection.
It stops with any of the filters engaged.
Mine does this too when set to the most sensitive setting. Cleans up it switched to level. But in analyzer mode it will oscillate at a low level. The oscillation stops if the analyzer is set to higher input level or higher distortion range.
I'm not sure if what you have there is really a problem. As far as I know they all do this.
The 339A I have has the extended range option.
I can't remember if the oscillation cleaned up replacing the analyzer op amps. Hp really did leave the 339A on the edge. Probably to achieve it's wide bandwidth.
Mine is not the extended range version, and it doesn't do the oscillation thing on the most sensitive range.
So, I'm still with the LT1468 with distortion/harmonic content
which appears to be on the verge of clipping.
I don't know what to do to adjust it. I've tried different caps
at C47 1000 pf, 750pf, to 250pf, it doesn't make a difference.
I run out of adjustment on R30....
On the bright side I got my parts HP339a working. I think it
works better than the one I'm upgrading...go figure. On #2,
I replaced the PS Electrolytics. I pulled the little 1uf tants
and they all tested find. I then replaced the Meter's power supply
power regulator and its controller. Replaced the Tants and did a
pre check. Who ever had it before me didn't have the power input
switch positioned correctly, so I positioned it correctly.
I plugged it in and powered it up. It works, well.
which appears to be on the verge of clipping.
I don't know what to do to adjust it. I've tried different caps
at C47 1000 pf, 750pf, to 250pf, it doesn't make a difference.
I run out of adjustment on R30....
On the bright side I got my parts HP339a working. I think it
works better than the one I'm upgrading...go figure. On #2,
I replaced the PS Electrolytics. I pulled the little 1uf tants
and they all tested find. I then replaced the Meter's power supply
power regulator and its controller. Replaced the Tants and did a
pre check. Who ever had it before me didn't have the power input
switch positioned correctly, so I positioned it correctly.
I plugged it in and powered it up. It works, well.
Lets slow down here and take a breath.
Do you have a scope?
If so then connect it to TP1 of the the A1 board. Confirm you have a 10VPP sine signal.
Adjust R30 until you do. If you can't then something is wrong with the leveling circuit.
Did you replace the Jfet?. If no then remove it, test it, see that it's okay. When I fried mine it was shorted at one of the junctions. Gate to source or gate to drain. Can't remember which. Did you remove that zobal, R29 C46. Remove C44, C45, C48, CR7, CR8,. Keep C32.
Did you replace U2A,B,C,D. If so put the old part back in. It doesn't make any difference anyway.
The LT1468 works. No fuss or adjustment required.
Start with this.
I thought I replied. This is strange on the both my HP339a, I measure a resistance of about 22 - 28 ohm to ground. More than a dead short?
You need to look at the signal amplitude a the FET drain with a scope. You now know the possible resistance range of the circuit and if you measure the feedback resistance at a given pot setting you can see what the signal level should be. If there is a big difference from expected in signal level at the drain, then that's a help in troubleshooting. Sticking an ohmmeter across the circuit could result in a variety of readings that won't help you figure out what's going on. At the FET drain, you should see a small amplitude sine wave at the frequency of the oscillator, probably not more than a couple hundred millivolts p-p; I haven't looked at mine in ages, and am not ready to open it up until I get on with serious modding, so it'll be a while.
If you run out of range on R30, which end is it at? If it's at max R, then you know that the R at the ground leg (the FET and the resistors) is too high. If it's at the low end, then you know the ground leg R is too low. David has pointed you to the AGC circuit, and he's right about the quad amp, the original is fine there if you've still got it. Don't mess with caps until you get the level stuff sorted, because that mostly only affects the highest range.
If you run out of range on R30, which end is it at? If it's at max R, then you know that the R at the ground leg (the FET and the resistors) is too high. If it's at the low end, then you know the ground leg R is too low. David has pointed you to the AGC circuit, and he's right about the quad amp, the original is fine there if you've still got it. Don't mess with caps until you get the level stuff sorted, because that mostly only affects the highest range.
When powered down a Jfet is full on. It s a depletion device. So of course you will get a low reading. You can take it out and bias if off with a nine volt battery. Put a 1k resistor on the gate for safety. -9V to the gate from the source. You can tie the gate to the source and load the drain with a 1k series resistor. Put 9V between the resistor and source. The jfet will function as a current source. Measure the voltage drop across the resistor and calculate the current. Compare that to the data sheet. If the resistor burns the jfet is f'd. I don't think a 9V battery is enough to do that but you get the point.
Yes.
Ok will do.
Checked with the other one, they almost test the same,
at least in circuit.
Zobal no, not completely. Will do it and yes
I've removed those others component, will reinstall C32, or keep
it there.
Yes, put the original back in.
I'll let you know the result.
Two other questions,
When you replaced C30 with the 100uf bipolar, did you also
add a second bi polar in series with R52?
Were each of those bypassed with a .1uf cap?
Adjusting R30, I can only get it up to 9.39 Volts.
This is with Zobal removed.
When I started adjust R30, it was at 6.4 volts.
Perhaps lowering the pin 6 resistor, to 33 or 40 ohms
in stead of 100 ohms?
Got some level issues I would say.
This is with Zobal removed.
When I started adjust R30, it was at 6.4 volts.
Perhaps lowering the pin 6 resistor, to 33 or 40 ohms
in stead of 100 ohms?
Got some level issues I would say.
Continuing further, I checked on A1 TP3,
Standard: 45mV
Measure: 16mV
A1 TP4 is also low
Standard: 75mV 0Vdc
Measure: 21mV +9.3 Vdc
Looking to find adjustments for them.
To Do: Remove Jfet and test.
Standard: 45mV
Measure: 16mV
A1 TP4 is also low
Standard: 75mV 0Vdc
Measure: 21mV +9.3 Vdc
Looking to find adjustments for them.
To Do: Remove Jfet and test.
The level is not controlled by the pot. It's controlled by the leveling circuit which has a fixed DC reference. The DC reference ultimately controls the level, The pot sets the operating point of the Jfet. The trim forces the gate voltage to change for a given frequency. The only reason you see a big level change is because the Jfet is force to a limit. The reference voltage at the input to the controller may be out of spec. 9.36Vpp is very close. Compare one analyzer to the other. Remember what's in the manual is nominal.
Changes where made over the years so the manual is a generalization. A guide.
Pin 6 of what? Can you include part reference as well?
For now don't mess with any resistor ratios. This just makes it harder to locate the problem.
Changing an op amp will have no effect on operating parameters unless it is oscillating.
The 1468 works fine in the A1U1 position provided the compensation and clamping is removed.
The 45mVpp at TP3 ride on the peak of the sine. The 9Vp is assumed to down from the 10Vpp input and only with selector switches as indicated. The 1.5uF I think id for the x100 range. It will be different for x1000. Same for the 75mVPP at TP4.
The +15Vdc, the resistor ratio of R34 and R35 of A!U2A establish the level reference. If the rail voltage is off so is your level. When the oscillator is level there will be a 0Vdc average at TP4 with the 75mVpp riding on it. When this condition is met the integrator output stops changing relative to the dc average input. There will still be some ripple voltage at the output of the integrator. The integrator's job is to store the total dc error. The error is established by the difference of the dc reference and the peak value of the sine input to the controller. The integrator behaves as an error accumulator and drives the error at the proportion amplifier output to zero. Once this state has been reached the loop is in lock and the oscillator is leveled.
And do not mess with the resistors in the FET circuit -- changing any of them will change the dynamic control range of the FET and mess up things all the way round. If the DC levels in the AGC circuit are where they should be, try adding some resistasnce to the feedback leg with the pot in it, say 470 ohgms... But I think David is right and there are problems in the AGC or power supplies.
David, RichEEM,
That is pin 6 of A1u1.
Supply voltages are good entering A1. Id rather them be right on the money
but I'll take +15.18 & -15.02.
R30 is running out of room on the low side. It gets to .7 ohms and stops,
voltage can't climb enough.
There is about 1/2 volt difference between A1U1 pin6 & A1TP1.
9.83V vice 9.37V. But, there are losses before that. I'll spin the Freq
multiplier and see what it dishes up.
Thanks for the explanation David.
That is pin 6 of A1u1.
Supply voltages are good entering A1. Id rather them be right on the money
but I'll take +15.18 & -15.02.
R30 is running out of room on the low side. It gets to .7 ohms and stops,
voltage can't climb enough.
There is about 1/2 volt difference between A1U1 pin6 & A1TP1.
9.83V vice 9.37V. But, there are losses before that. I'll spin the Freq
multiplier and see what it dishes up.
Thanks for the explanation David.