Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

slight ghost tones on SF Super Reverb?

3/31/2000 4:20 PM
slight ghost tones on SF Super Reverb?
Hello all,  
I have serviced the amp in question quite a few times and now I am running into a very slight lower ghost or harmonic tone on or around a higher fretted A note. I have practically rebuilt this amp over the coarse of the last year for this guy and I thought I had seen the last of it...nope.  
All the standards have been replaced:  
Checked all the tubes with known good stock  
Filter caps  
Coupling caps  
Any and all drifted resistors  
Cathode bypass caps/resistors  
The only thing to note is that someone before me used a Bassman UL OT transformer as a replacement.  
I am chalking this up to a grounding issue or a lead dress issue but I thought I would throw it out there for any ideas.  
Thanks guys,  
And now, a word from our sponsors:

3/31/2000 4:37 PM

What kind of speakers are you using, if I may ask
3/31/2000 4:50 PM
Ditto. Check with known-good speakers before digging any deeper into the amp.  
Crossover distortion sometimes sounds like ghost notes as well.
3/31/2000 6:19 PM
Thanks.......I'll keep you posted!
I have heard the ghost tone on his four 10" cab and on a single V30 in my shop.  
Thanks to all for the replies.:>)  
3/31/2000 5:00 PM
Re: slight ghost tones on SF Super Reverb?
I remember reading a long thread on alt.guitar.amps forum about ghost notes on a Super Reverb that would not go away, in spite of replacing many parts on the board. Try searching on "ghost notes" and "Super Reverb" if you're so inclined.  
In one of Weber's books, it mentions decreasing the value of the .1 coupling caps to the output tubes as a remedy that supposedly will not affect tone.  
I've been trying to get rid of ghost notes on my SF Princeton Reverb but no luck yet. I did everything you did but I haven't tried the Weber trick yet. The lead dress looks very sloppy in mine too, but aside from shortening the grid leads, I really don't know how to dress the other wires properly. I could really use a picture of the chassis layout of a BF model.  
Please keep us posted if you find a cure.  
Good luck!
3/31/2000 5:26 PM
I think the decreasing of the .1uf cap is a bandaid, ghost notes seem to be 1st PS cap ESR problems(plate supply). Sometimes new caps are bad! Thats one reason I check every cap I use in a TUBE AMP.
3/31/2000 10:16 PM

I have seen what Gus is talking about also.  
The B+ that supplies the plate(s) through the OPT is subject to some fluctuation when high power output signals are being swung by the power tubes. That swing is supposed to be confined to the plate end of the windings. Of course the voltage drop of the AC signal across the OPT primary inductance from there to the B+ DC at the first filter cap, is matched by the secondary winding to drive the speaker.  
(see footnote)  
So the high voltage DC B+ supply is also supposed to be at AC ground for this reason as well, this is accomplished by the same first cap that filters AC from the power transformer & rectifier. If this critical first filter cap has too high an Equivalent Series Resistance (ESR) to AC signals, then a portion of that unearthed AC will be reflected directly to the speaker, its frequency is from the main signal but it *comes from* the other end of the primary and it produces a ghost note.  
The ESR is variable depending on the manufacturing quality of the individual capacitor, and could amount to a few ohms at certain frequencies.  
Also this is one of the most audible locations for a cold solder joint, since if its resistance were to amount to an ohm or two the AC to be filtered will still have measurable resistance to ground even if the cap were perfect. I have fixed ghost notes just by remelting the connections applying to the first cap.  
Cold solder joints don't usually quite resist an additional ohm, so I wouldn't try to measure it for troubleshooting, this could be heard instead with greater sensitivity, or the solder viewed under magnification.  
I have found the Illinois caps sound like they usually have higher ESR than Nichicons, FWIW.  
To check for this carefully, making sure all connections are ideal:  
1) Add an extra cap in parallel to the first cap in place. This will lower the ESR just like paralleling two resistors lowers the working resistance.  
2) Temporarily remove the old one & replace with new, this is more to find a relatively bad cap, like Gus says he checks them all in real circuits, bad news for everybody that tubes are not the only components to audition, they're just the easiest to change :-(  
Some hifi amps use a different value smaller HV cap (I've seen 0.1MF's) of a different manufacture in parallel with the main filter, this does not add much capacitance in parallel but is supposed to allow lower ESR for the frequencies passed by the small, usually film type additional cap.  
A different source of not only ghost notes but unwanted distortion can come from soldering on a tube socket while the tube is still plugged in. Probably doesn't let the connection heat up or cool down as fast due to heat-sink action of the tube pins (remove tube & resolder), or maybe a little oxidation is more rapidly formed from the dissimilar metals under higher temperature conditions (reseat tube to get fresh contact), or even might lose tension on the heated socket contacts from expanding to fit the pin then relaxing from temperature stress (retension socket).  
If the effective inductance of the OPT primary is enough Henrys, it will resist very well the AC signal power applied by the tubes at their end of the winding, resulting in relatively little signal-derived AC making it through the coils and arriving at the first filter cap to begin with. With a replaced OPT it will obviously not have the exact same Henrys as the original, if less, you might expect more ghost notes regardless.  
Next assume a relatively ideal OPT. At the plate end of the primary there is, say, a 500VAC output swing from a power tube. At the B+ end there is no AC fluctuation since it is at AC ground through the first filter cap. If this OPT had a tap (like a UL OPT) perhaps halfway between the B+ & plate, and it was not connected to anything, you could measure on it a 250VAC swing in phase with the plate swing. This would be a high-amplitude/high energy potential source of interference, so lead dress might come into play even with the unused taps like it can for the plate leads.  
Hope this helps,  

  Page 1 of 2 Next> Last Page>>