Hmmm. The osc is pretty simple in design. Just switched C's to change freq range. The output is rectified and active filtering to drive the jFET control. If opamps have been replaced and jFET is good.... not much left to go wrong..... except really bizzar things....
One BIG source of trouble is if you remove the individual switch wafers to better clean them.... there is a very very high risk that you won't replace one of the switch wafers in the proper direction. Then the osc will never work. I have seen this happen before.
Similar disaster can result if coupling shaft set screws were loosened and allowed the switch wafer shaft to rotate and then retightened in wrong position....
THx-RNMarsh
One BIG source of trouble is if you remove the individual switch wafers to better clean them.... there is a very very high risk that you won't replace one of the switch wafers in the proper direction. Then the osc will never work. I have seen this happen before.
Similar disaster can result if coupling shaft set screws were loosened and allowed the switch wafer shaft to rotate and then retightened in wrong position....
THx-RNMarsh
Last edited:
Switchology 101
Actually, I was thinking the same thing, but didn't want to butt in because, Well you know too many cooks and you end up with Roo in the stew! Now, something in the brownies...is a whole 'nuther story.
What is someone before the OP messed with it and never new what he did to screw the pooch.
And to hear Richard's discussion about someone gifting him a switch and the length of time and concentration it took to figure out putting it back together again.
So it works in one position only, then when the switch is turned it doesn't work any more after that until the initial position again.
Switchology, We had to develop this in various Mil-lessons,
groupings called switchology, and if we didn't have the actual information we had to SWAG it. It either got caught during review (if incorrect) or hoped we'd get feedback if we specified it incorrectly after publication. Contract also specified on going support for the lessons too.
Please keep us posted on what you find.
Cheers,
Actually, I was thinking the same thing, but didn't want to butt in because, Well you know too many cooks and you end up with Roo in the stew! Now, something in the brownies...is a whole 'nuther story.
What is someone before the OP messed with it and never new what he did to screw the pooch.
And to hear Richard's discussion about someone gifting him a switch and the length of time and concentration it took to figure out putting it back together again.
So it works in one position only, then when the switch is turned it doesn't work any more after that until the initial position again.
Switchology, We had to develop this in various Mil-lessons,
groupings called switchology, and if we didn't have the actual information we had to SWAG it. It either got caught during review (if incorrect) or hoped we'd get feedback if we specified it incorrectly after publication. Contract also specified on going support for the lessons too.
Please keep us posted on what you find.
Cheers,
The shafts have two flats one with a ridge down the centre line so they cannot be passed through the wafers wrongly, so if all the ridges are in line I don't think it can be wrong. Anyway, I loosened the coupling on the digit shaft and moved the null side one up one down and the null was worse both ways.
Using an external oscillator I get a lesser null when set the frequency controls in the two ways possible to set for say 1kHz ie 1-0-00 and 0-10-00 better null is the first way should be the same I think.
I am still puzzled as to why the trim pots on A4 do nothing at all even when using an external oscillator which negates null circuit errors.
Using an external oscillator I get a lesser null when set the frequency controls in the two ways possible to set for say 1kHz ie 1-0-00 and 0-10-00 better null is the first way should be the same I think.
I am still puzzled as to why the trim pots on A4 do nothing at all even when using an external oscillator which negates null circuit errors.
Hi Microx,
I can try to elaborate on the A4 pots if the previous bit was lacking.
Please help me review the current state of affairs and correct me if I'm wrong.
On A4 board, U3 opamp is likely failed, U5 health uncertain. But nice clean sine waves on TP 3,4,and 5.
On board A3, clean sine waves at TP4, TP3, and U3 pin 6. But nulling is poor, so TP3 and U3 pin 6 are down only a few dB re TP4.
TP1 is a clean sine wave, but it's phase shift seemed incorrect.
------------------------------------------------------------
I'd like to reassure myself about the TP1 phase-shift re TP4.
So back to dual channel scope display, 1KHz signal, with both sine waves centered so there's equal amplitude above and below scope center line. (TP1 amplitude should be the same as TP4 if the phase shift amplifier is working properly.) Scope should confirm period is 1 ms. If phase shift is 90 degrees, the time between zero crossings of the two signals should be 250us. What is the time difference in actual fact? If other than 90 degrees, would you measure resistance between TP4 and U1 pin3?
I'm still optimistic about eventual success.
I can try to elaborate on the A4 pots if the previous bit was lacking.
Please help me review the current state of affairs and correct me if I'm wrong.
On A4 board, U3 opamp is likely failed, U5 health uncertain. But nice clean sine waves on TP 3,4,and 5.
On board A3, clean sine waves at TP4, TP3, and U3 pin 6. But nulling is poor, so TP3 and U3 pin 6 are down only a few dB re TP4.
TP1 is a clean sine wave, but it's phase shift seemed incorrect.
------------------------------------------------------------
I'd like to reassure myself about the TP1 phase-shift re TP4.
So back to dual channel scope display, 1KHz signal, with both sine waves centered so there's equal amplitude above and below scope center line. (TP1 amplitude should be the same as TP4 if the phase shift amplifier is working properly.) Scope should confirm period is 1 ms. If phase shift is 90 degrees, the time between zero crossings of the two signals should be 250us. What is the time difference in actual fact? If other than 90 degrees, would you measure resistance between TP4 and U1 pin3?
I'm still optimistic about eventual success.
Sorry again for delay, the 339A on off switch went kaput so had to cut the cables and bridge it out. NOW, changed the probes on my scope for better ones came today AND NOW the 90deg phase shift is spot on, also the oscillator frequencies are more or less correct in the CAL position so it looks like the crappy probes were loading things adversely. A4 board, I will change U3 and U5 when the chips come and we shall see if anything changes. TP 3, 4 and 5 yes clean sinewaves. A3 board, should U3 pin 6 be the full null down ie 90-100dB or only the 16dB of the primary notch
Hi microx,
That makes sense now.
Since you have just received a set of new probes for your scope, take the time and use some epoxy to force the probe to be in the X10 mode all the time. I can't think of any brand of good scope probes that include a X1 position and that can be destructive to your scope input, or can load down the circuit you are measuring unexpectedly.
Best, Chris
That makes sense now.
Since you have just received a set of new probes for your scope, take the time and use some epoxy to force the probe to be in the X10 mode all the time. I can't think of any brand of good scope probes that include a X1 position and that can be destructive to your scope input, or can load down the circuit you are measuring unexpectedly.
Best, Chris
When everything is working correctly, U3 pin 6 is the fully notched ~100dB null. U3 pin3 will have the -16dB notch.
You can regard U3 as a differential amplifier that measures the difference between an unfiltered input (the R48, R49 path) and and a filtered sine wave appearing on U3 pin 3. The phase control path tweaks the tuning of the filter so that it exhibits no phase shift--- i.e it has pure 16dB attenuation with no attendant phase shift. The amplitude control adjusts the attenuation in the R48, R49 path to exactly match the loss in the Twin-T. The result is "perfect" cancellation of the fundamental at U3's output.
Elegant, no?
You can regard U3 as a differential amplifier that measures the difference between an unfiltered input (the R48, R49 path) and and a filtered sine wave appearing on U3 pin 3. The phase control path tweaks the tuning of the filter so that it exhibits no phase shift--- i.e it has pure 16dB attenuation with no attendant phase shift. The amplitude control adjusts the attenuation in the R48, R49 path to exactly match the loss in the Twin-T. The result is "perfect" cancellation of the fundamental at U3's output.
Elegant, no?
If you have the 339A with toggle on/off switch, there is a known problem with those units. The push button switch is the better unit. Richard did the evaluation on my unit also. He never would tell me exactly what or where the issue was, just that it was.
Also depending on your until when you start dialing in the meter's FS deflection by using the two pots designated in the manual, check the errata/change's at the back of the manual, their function and designation is reversed.
Sounds like you got past whatever it was at the time.
Good job, keep us posted please.
Cheers,
Also depending on your until when you start dialing in the meter's FS deflection by using the two pots designated in the manual, check the errata/change's at the back of the manual, their function and designation is reversed.
Sounds like you got past whatever it was at the time.
Good job, keep us posted please.
Cheers,
Sync, the switch is just the switch, job for later. Thanks for the heads up on meter adjustments again, job for later.
BSST, To confirm Board A3, TP4 and TP1 perfect same amplitude sinewaves with 90deg phase shift.
TP4 and U3 pin 3, external osc waveforms same phase at null.
TP4 and U3 pin 6 perfect 180deg null external osc at 863 Hz, 339A set at 1kHz. Amplitude at TP4 falls and at U3 pin6 rises as null is found.
The trace of pin 6 "slides" under TP4 trace until peak meets trough at null, as you say -elegant.
BSST, To confirm Board A3, TP4 and TP1 perfect same amplitude sinewaves with 90deg phase shift.
TP4 and U3 pin 3, external osc waveforms same phase at null.
TP4 and U3 pin 6 perfect 180deg null external osc at 863 Hz, 339A set at 1kHz. Amplitude at TP4 falls and at U3 pin6 rises as null is found.
The trace of pin 6 "slides" under TP4 trace until peak meets trough at null, as you say -elegant.
From post 87, "On A4, TP 3,4 and 5 sine wave similar amplitude. Tp1 garbage TP2 DC only. Circuits from Tp1 and TP2 drive E1 and E2 leds so DC would make more sense than garbage. Also the three trim pots do nothing as far as I can see "
I am waiting fir TL074 quadamps for A4 board. As for the single opamps, some seem unmarked .Dont know what I could sub them with.
I am waiting fir TL074 quadamps for A4 board. As for the single opamps, some seem unmarked .Dont know what I could sub them with.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.