My original PI tube must have been bad. Any suggestions for how high I can/should go on the plate voltage for these tubes (other than the max plate voltages in the data sheets)? I have an old Heathkit decade resistance box I can use to try stepped values for the resistor that feeds the 6sc7 and 6sl7. Any more headroom to be gained by increasing the 6l6 screen voltage?
Also, I still have ghost notes like crazy when playing two notes at a time. I have one more 20uf section in my 40/20/20/20 cap can I could bring into play. Any suggestions for adjusting lead dress? Should I be twisting ground and B+ leads together? Tino Z suggests adding a .1uf snubber cap across the output transformer as per the schematic below. Perhaps I'll try that but I'd rather fix things by adjusting lead dress.
I don't think lead dress is an issue my amp looks like a rats nest more then yours and it is very stable. I've always read where the 1m grid leak should be before the 68k grid on the tube pins. But not sure if that is a problem or not I think I had mine like yours originally but have since changed them. Not due to problems though just changing values, so I put the 1m and 10m on the input jacks. EDIT
I have my grid leaks at the tube socket like you do no problems
I know the 6SC7GT I got gave me problems so I used the NOS metal tube. I think they can handle voltage just fine. Keep the screens below the plates on the 6L6GC.
Make sure each component is the correct value I recently found my resistor draws mixed up and check each resistor before use now.
Make sure your wiring is correct and check to make sure you grounded all the tube pins Tino showed on those preamp tubes.
Thanks Mark, Getting ready to get back to work on the amp. Would you be willing to share the voltages you ended up with? I'm going to change the location of the 6sc7 grid leak resistors, then start working with the dropping resistor that feeds the 6l6 screen grids, shooting for around 390 on the screens.
I used my decade resistance box to determine that 1k for the 1st dropping resistor will give me 390v on the screens with 396v on the plates. That also gave me 219v on the plates of the 6sl7 and 157v on the plates of the 6sc7, which seem to be in the ballpark. To my ears the amp sounds much better for steel guitar with the higher 6l6 screen and preamp voltages.
However, I still have ghost notes under what I'm playing. They are low 3rds, like a frog singing bass under my two and three note chords. The ghosts are very noticeable with the tone control turned down. As the tone control is turned up the ghost notes gradually disappear.
I added the snubber cap across the output transformer as suggested by Tino Z and it sucked the life out of the amp. I immediately snipped it out.
I also added the extra 20 uf to the 6l6 screen supply. I noticed a slight stiffening of the power supply but no help with the ghost notes. I also tried a number of different 6sl7s and 6sc7s. I noticed definite tonal differences between the different manufacturers of 6sl7s but no tubes we more or less susceptible to ghosting. I double checked all my solder joints and reflowed some of them. All the things that need to be grounded are solidly grounded. Looks like i'm going to have to use my oscilloscope for something other than testing for the outer foil of caps.....
Got the scope hooked up. I have my signal generator set to 400hz and can see sine waves on all the plates of the different stages. Now, ummm, what exactly am I looking for? Can you guys help me interpret these?
You do not have the required permissions to view the files attached to this post.
Thanks Dave. Ken Fox over on the steel guitar forum pointed out that Hammonds sometimes used brown and blue leads for the opposite phase compared to Fender for the plate leads of the OT. I swapped the leads and it seemed to make the amp less brilliant (more dull) sounding. I still had the low frequency oscillation as well. I've been really hoping my problems are a result of a simple-to-fix error like that.
The amp wants to sound really good, and probably no one in my house would notice the oscillations very much except me, but they are there.
If my traces are nice and sharp until the output, what does that say about the preamp/pi section? I'm considering redoing the grounding system entirely. I'd move the filter cap and dropping resistor for the preamp/pi off the cap can and over to the other side of the amp and ground it on a buss with the stages it decouples. That bus would ground at one of the input jacks and everything on the 6sc7 and 6sl7 would ground to it in order, including the filter cap. I'll ground the cap can to where the cathodes of the power tubes ground.
Also, how do I know if I have the OT wired correctly? It works wired both ways. I guess I'll take the better sounding one.
Matt D wrote:TI still had the low frequency oscillation as well. I've been really hoping my problems are a result of a simple-to-fix error like that.
...
The amp wants to sound really good, and probably no one in my house would notice the oscillations very much except me, but they are there.
...
I'm considering redoing the grounding system entirely. I'd move the filter cap and dropping resistor for the preamp/pi off the cap can and over to the other side of the amp and ground it on a buss with the stages it decouples. That bus would ground at one of the input jacks and everything on the 6sc7 and 6sl7 would ground to it in order, including the filter cap. I'll ground the cap can to where the cathodes of the power tubes ground.
Also, how do I know if I have the OT wired correctly? It works wired both ways. I guess I'll take the better sounding one.
I completely agree about grounding changes, especially if you are getting motor boating. Make sure the preamp is completely decoupled. I would run any preamp grounds such as the input and cathodes back to the filter cap ground for that node. Don't have more than two nodes on the same cap. If you are getting high frequency oscillation that is not due to OT plate leads, check your lead dress to rule out induced parasitics with certain wire runs possibly interacting. Chopsticking may or may not help. Run the OT common back to B+1 only. Avoid grounding on the PT or OT mounting bolts. I like your buss idea. Some of the guys doing PTP layouts have done similarly (RP and someone else (?) have some nice 5F1 PTP builds). I have a very high gain amp that is PTP with a bus with each node draining back to it and it is dead quiet.
Aren't you following the TinoZ layout that Mark also used? You shouldn't have to change anything and IME these old, low gain circuits are usually quite forgiving, at least with novals. Maybe just a clean up of wiring and a recheck of everything? I would just try and make it as close to the TinoZ as possible before trying something new. IME also, Octals, especially metal 6SC7s, are troublesome, you need a good handful to find an adequate one, much less a good one. I had an oscillation on my 5C8 that I could radically affect with different 6SC7s. I've found ultrasonic oscillations to make the amp darker so on the OT the brighter might be the correct wiring. Maybe playing w/ the NFB, disconnecting and comparing or different tap / increasing it might help bring out the oscillation and help you determine which way to flip the wires.
I looked over your photos and nothing is popping out at me it appears to be as Tino showed.
Did you ground pin 1 of the 6SC7 preamp tube? Tino does like to ground the input jacks to the cathode on most of his projects but not this one we just tighten the switchcraft jacks down tight for grounding. The volume and tone pots do have their grounds connected to V1 cathode ground; however.
Play with the NFB resistor location and see if that has any impact.
Thanks for the replies and suggestions guys. Keep 'em coming!
Here is what I tried tonight:
1) Grounding the 27r resistor to the speaker jack.
2) Grounding the cathode of the power tubes to the same place as the can cap
3) Moving the negative feed back from the 4 ohm to the 8 ohm OT tap.
4) Changing the routing of the long wire that connects pin 6 of the phase inverter to the output section.
5) Changing the OT leads around until I found which way sounded best (the original way).
6) Clipping out the negative feedback.
7) Substituting in 12 different 6sc7s.
8. Four different 5u4s
9) Subbing some old Tung Sol 5881's for the the new JJs I had as power tubes.
10)Different Speakers
11)Different guitars
Throughout it all and in spite of everything, it still always sounds like someone is playing a jug in harmony with me whenever i play two notes or more at once.
Since you're following a known good layout I suspect there's got to be a mistake somewhere. I looked at the Zolota pages, it covers everything like a Dynakit or a Mullard guide so I'm thinking there's got to be a wiring mistake somewhere. He has a voltage chart, don't recall if you compared to that. Did you check and recheck all the colors/numbers on your resistors? It's possible you have a bad cap, electro or coupling but that's really rare.
I had a cook friend once say that what he loved about cooking was the wild swings of being lost deep in the weeds and then suddenly finding yourself running through the open green fields with the sun shining. Maybe that's why we build amps. Green fields await.
Last edited by rp on Wed Jan 23, 2013 8:07 pm, edited 1 time in total.
Thanks RP. All my voltages are ok. I did adjust the dropping string to raise the screen voltages on the 6l6's but before that my voltages were within 10% of Tino Z's chart. Maybe I should try screen grid resistors? I'm using 80's vintage Russian military PIO coupling caps. I tested them for leakage with my DMM ohmmeter and they show absolutely no continuity, but I don't have the gear to test them at high voltage. What does a leaky coupling or tone shunt cap sound like? All my resistor values have been double checked and are correct.
Not a lot of progress tonight but I did replace the PI coupling caps and change the output transformer ground from the jack to the power supply ground. No improvements.
