weird octave overtones?

[hesamadman]

Not really sure whats going on here but I am getting some weird octave overtones with the pluck of each string. Pluck the A string, you can hear an octave lower in a sense. Its almost the sound you get when you have a very slightly loose connection somewhere. Anyone ever come across this?

[jjasilli]

Yes.  Another possible culprit is parasitic oscillation, which might be helped by a snubber cap across the PI output; or a compensating RC filter (conjunctive filter) after the power tube in an SE amp.

[hesamadman]

It's one of the 300 watt el84 builds I've taken apart and put back together. So many old solder joints. I don't know what else it could be since it's the same components I've always used. 

[hesamadman]

snubber cap across the PI output

What do you mean by this? I'll look into it.

[jojokeo]

snubber cap across the PI output

What do you mean by this? I'll look into it.

It's a common thing of connecting appx 100pF cap to each plate of the phase inverter tube.

You're having an oscillation going on with your amp at a lower frequency it appears. This can occur for a number of reasons. Have you tried swapping your OT primary wires going to the power tube's plates? Sometimes it doesn't SCREAM at you making it obvious. Disconnect NFB is first problem isn't it and see if this helps or not? Do you have feedback wire ground going to right location? Ground wire at output jack going back to reservoir/filter cap? Go through process of elimination...

[hesamadman]

What do you mean by this? I'll look into it.

It's a common thing of connecting appx 100pF cap to each plate of the phase inverter tube.

You're having an oscillation going on with your amp at a lower frequency it appears. This can occur for a number of reasons. Have you tried swapping your OT primary wires going to the power tube's plates? Sometimes it doesn't SCREAM at you making it obvious. Disconnect NFB is first problem isn't it and see if this helps or not? Do you have feedback wire ground going to right location? Ground wire at output jack going back to reservoir/filter cap? Go through process of elimination...



Swapped OT primary to no avail. No NFB in this amp. However, could my smaller PI CC be a problem. Design calls for .1 i have .022. I have his same setup in a very very similar amp though so i wouldnt think it is. Only other thing I did was moved filter caps from end of board and connected them in the preamp next to cathodes of each stage.

[jjasilli]

Yes, checkout for example the Traynor yba-1 schematic.  There's a 47pF cap across the outputs of the PI.  That's a common value, but could range form 10pF to 100pF or so, as posted above.  It might help.


Also, do check connections & components, especially the caps, in the Power Supply.  Also, I think excessive power supply filtering can cause harmonic ringing.  Maybe try less filtering - like maybe 8uF in the last filter stage.


NFB might also help.  True, this does not cure the root cause of the problem.  But if the root cause remains elusive, this might be worth trying.

[shooter]

Also, I think excessive power supply filtering can cause harmonic ringing

Do you know a source for more info, I generally build with 4 100uF caps, mostly SE/PSE/QSE and they all seem to have pleasing but *odd* overtones, since I'm not a real player I don't have much hearing experience, but all the real guitar guys comment on my "tremolo" sound
thx

[jjasilli]

Will look fr a source, but haven't found it.  This a recurrent topic which has come up before, but many of the really old threads are gone form the Forum.


Here's a thread, where Psychonoodler blames inadequate filtering:  http://el34world.com/Forum/index.php?topic=12181.msg112953#msg112953


This is supported by these threads which also blame inadequate filtering, i.e., low filter cap values; or bad filter caps. 


http://www.marshallforum.com/marshall-amps/60732-js-jvm-ghost-note.html


http://www.thegearpage.net/board/index.php?threads/ghosting-and-ghost-notes.236949/


Anyway, my memory (which cannot always be trusted!), is that so long as the early filter stages are large enough to provide good filtering, then 8uF in the last filter stage (1st preamp gain stage) can help here.  However this applies to vintage amps.  This may not be useful for modern, hi-gain amps in any event.

[shooter]

Thanks JJasilli for the links.  I now should be smarter, ya, NO

[hesamadman]

I believe the culprit was parasitic oscillation. I'm basing this off of recent information that I unvovered. I read that it could cause overheating of output tubes. I noticed my el84 were overheating quick. This amp was poorly put together. Was my first 30 watt. Not only was lead dress a problem, a poor layout on my board, as well as excessive distances from tubes to board. Leaving long lengths of wire from input grid and screen grid of my el84. The chassis used was incredibly large. With my new layout and smaller chassis, I could be able to eliminate these issues.

Can anyone point me to some info on parasitic oscillation? Maybe a section in merlins book?

[jjasilli]

google is your friend:  https://www.google.com/search?q=parasitic+oscillation&rlz=1C1RNBN_enUS433US433&oq=parasitic+oscillation&aqs=chrome..69i57&sourceid=chrome&es_sm=122&ie=UTF-8


E.g., it can be caused by "poor lead dress" = wires run in less than optimal places.   Moving signal wires away from other leads may help.  It is trial & error using educated guesses.  The problem is Gremlins:  given a bunch of identical amps, some can have issues that others don't.  As a DIY'er one can choose to isolate symptoms all day.  For production manufacturers it's a percentage thing, they can't spend time & effort on individual amps.  So they tend to use snubber caps, NFB, etc., to defeat such issues.