Maybe someone here has a clue, I certainly don't. I have a Citation 17S preamp on the bench that sends DC to the output of one channel whenever the subsonic filter is switched in. It swings erratically from + to - with a magnitude of 2.5 VDC or so, gradually decaying to lower voltages. There is a small measureable pulse at the the output of the filter board when the subsonic filter is switched in, on the order of a few MV DC. I suspect that the output stage may be oscillating, but since my scope is dead, I cannot view the output waveform. I am using a DVM to check for DC. Here's a link to a schematic, any ideas are welcome: http://manuals.harman.com/hk/Service Manual/CITATION seventeen s sm.pdf
So far, I have changed all the electrolytics on the filter board, as well as the ceramic caps related to the subsonic filter, and the op amp also.
So far, I have changed all the electrolytics on the filter board, as well as the ceramic caps related to the subsonic filter, and the op amp also.
The problem was there before the opamp change, changing it made no difference. I removed the switch, took it apart, and burnished everything plus cleaned with Deoxit, the switch is now spotless, and the condition is the same. I believe I may have more than one problem- minor glitch on the filter board does something that results in the output board for that channel oscillating/swinging.
XEAGLEKEEPER said:
XEAGLEKEEPER said:
The unit does function with the subsonic filter disengaged; problem is in the RH channel. The mono switch has no effect, problem remains on the RH channel. With the Hi cut engaged, the problem changes- a single spike of around 70 to 90 mV DC appears at the RH output, quickly decays, does not swing or oscillate. I have -15.27 VDC at pin 4, and +15.46 VDC at pin 7. When I got this piece, it was a DOA/parts unit. It needed new regulators plus Q3 and Q4 on the power supply board for it to operate. Shortly after I got it operational, the subsonic filter problem appeared.
If the hi filter is engaged after the subsonic, the voltage swing is not halted by engaging the hi cut.
XEAGLEKEEPER said:Were you measuring oscillation off of pin 6 of opamp
Suppose you have already thoroughly inspected the filter board for cold solder joints especially around the connector J1, all grounds...possible hairline cracks in the PC board. Were running out of options here
Hmmm... on the bad channel, I have around .98 VDC at pin 6, and on the good channel, a few mV. I am going to go over the board for bad traces, bad grounds and see where that leads me.
the op amp is connected as a unity gain follower..... what is the dc voltage on pin 3? it should be zero, if it is, and you have .98 volts on the output, i'd suspect the op amp.... try a TL071 in it's place, you have no way of knowing exactly what type of op amp NTE used for that replacement. it could be a 741 for all you know.... what was the nte part number? a ca3100 is a cmos input op amp, and if static damages the inputs, it could cause offset, so try a jfet input amp.
XEAGLEKEEPER said:You could also increase the value of C22 from 470pf to 560pf or so.
Should have said earlier, this has the later opamp, C19 and C22 are not used.
Tube_Dude said:
R24 is good to ground; C20 was replaced with new.
unclejed613 said:
This unit is later spec, used a LF356N op amp, C19 and C22 are absent. The NTE937M was used to replace it, and the package states that it is a JFET input op amp. DC voltage on pin 3 is 0 V. I have a spare NTE937M which I could try, don't have a TL071 on hand at present.
All traces, grounds etc. seem OK, but I will double check all of the above. Looking at the circuit, the only way that I can see 1 VDC at pin 6 would be if the op amp were conducting/causing offset when it should not be.
Hi NH TomG
I think that your op amp is oscillating.
The NTE937 look to me similar to the LF357 Op amp , a very fast op amp (25 V/uS ) and not unity gain stable , need a minimum gain of 5.
http://www.nteinc.com/specs/900to999/pdf/nte937m.pdf
Adding to this , as you have take out C19 and C22,oscilation seems a very plausible explanation.
Put a LF356 and C19 and C22 and inspect if the issue is gone.
I think that your op amp is oscillating.
The NTE937 look to me similar to the LF357 Op amp , a very fast op amp (25 V/uS ) and not unity gain stable , need a minimum gain of 5.
http://www.nteinc.com/specs/900to999/pdf/nte937m.pdf
Adding to this , as you have take out C19 and C22,oscilation seems a very plausible explanation.
Put a LF356 and C19 and C22 and inspect if the issue is gone.
I agree that it may be oscillating, but the circuit configuration for the LF356N does not require C19 and C22; those were used with the CA3100 (earlier spec). Once I have rechecked all traces and grounds, I will replace the NTE937M with another that I already have, and see what happens. If that fails, LF356N's are around $0.55 each at Allied, maybe I'll buy a handful.Tube_Dude said:
Tube_Dude said:
I see your point- if the device requires a minimum gain of 5 to be stable, it won't be stable as a unity gain follower. Since my scope is dead, I can't observe that it is oscillating, but I can be certain that it has an offset problem. I will order a new LF356N from Allied and use that instead of the NTE937M.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.