OK - dig this - all of my calculations have either put the EL34s at just below the very threshold of max dissipation OR ever so slightly above. Swapping out the shared 470R screen resistor for two individual 1.5K resistors seemed to get the dissipation to below max and therefore at "safe" levels. However, at points I have suspected that "over-dissipation" may have been part of the problem, if not possibly the problem entirely.
SO, I just disconnected one of the output tubes from it's cathode resistor and alligator clipped the two cathodes together so they'd be effectively sharing one 270R cathode resistor which would put output dissipation well below max and VOILA - the issue is GONE! Entirely gone, nothing but a nice sounding amplifier I thought the amp would sound quieter and more tame and while it may have thedifference was much less dramatic than I thought it would be. I am going to try and experiment with "splitting the difference" and using some higher value individual cathode bias resistors to retain the edginess. But overall, very surprising and encourging results!
...to elaborate further, this all begs the question WHY did this cause an issue? After all, the infamous Matchless Clubman is widely known as an EL34 "torture chamber" which chews through output tubes at an alarming rate, but I've never heard any complaints of it not sounding good or being plagued with issues. So, it's unlikely that over-dissipation alone has been my problem.
I suspect that it was the output of the EL34s combining for almost 50 watts being pushed through a 35 watt rated output transformer. As mentioned, the OT is an old Northlake made 35 watt wired up with 8 & 16 ohm outputs and a 6.6K primary. It is an exact match in specs and measurements to the Classictone "Trainwreck" output transformer, part no 40-18064. I don't think I damaged the OT but I do think the crackle was possibly it letting me know I was pushing it too far? Anyway, I think if I try and stay within the 35 watt combined output spec I should be ok. With the alligator-clipped shared 270R cathode resistor I'm getting approx 26 watts output combined. I might try to get it up to around 30 watts just to be safe. Time to order a bunch of different value 10 watt resistors to swap out until I hit the mark.
with 5AR4 rectifier, the new readings are, in VDC:
Thanks for all your help, everyone. There were some valid points made here that I'll be observing on future builds in any case. As always, the people here are a great help and I appreciate it!
Here's a brief sound clip of the amp with a pair of USA Tung Sol 5881s installed as output tubes. With the shared 270R cathode resistor I think I can now use these type of tubes "safely" and they actually blast out more power with this setup than EL34s do. By my (possibly screwy) calculations I am getting around 26-27 watts with a pair of EL34s (that's including factoring in the screen current) and the 5881s are putting out 33.5 watts combined (that's using the classic calculation which does not separate screen current).
Here are the voltage readings with 5AR4 rectifier and 5881 output tubes (in VDC of course):
Plate - 352
Screen - 352
Plate-to-cathode - 324
Cathode-to-ground - 27.4
Average 16.75 watts dissipation per tube x2 = 33.5 No crackle popcorn sound from amp when driven, at that's right below the OT's 35 watt rating. Amp is turned up to about 9 out of 10 volume, guitar is Telecaster re-issue. Guitar is cranked until about 0:27 in when guitar's volume pot is turned down to clean up sound briefly then turned back up at end of clip. This was done in a rush while both of our downstairs neighbors were away (usually one of them is always home!)
You do not have the required permissions to view the files attached to this post.
Here is another sound clip of the amp as it remains with the shared cathode resistor and 5881 output tubes. This one is a little more extensive, recorded with a ribbon mic about 3 feet away from 2x12 cab into garageband using an SG w/humbuckers
pompeiisneaks - how do I convert this to mp3 like you did??
You do not have the required permissions to view the files attached to this post.
You just click on the 'mp3' button on the top AFTER you've uploaded the file and saved it... and then get the link for the mp3. I should create a howto, but I'll just post it here for you.
Nice sounding clips... I wonder about the condition of the power tube cathode bypass caps... Any way to see if one of them in the former split-cathodes configuration was bad?
"Education is what you're left with after you have forgotten what you have learned" - Enzo
Ok, I'm gonna try this with another demo recording!!
This track doesn't feature an amp I've built...but I did play all the instruments... (ok, Michael Jackson didn't come over to my house to use the bathroom...but his sister did!)
The guitars were tracked using a JTM45 clone I owned for a while. Some lead guitar tracks were added later using a Bogen Challenger.
dorrisant - good eye. I have not tested, but will give it a look when I get the chance.
I think something was off though about my calculations and numbers adding up to around 50W output using the separate cathode bias resistors and EL34 output tubes. Everything I've read indicates you can't really expect more than around 30W output from cathode biased EL34s, so there probably was some kind of "overloading" somewhere. Plate & screen voltages were very conservative, and I used resistor values mentioned from many sources as being stock (or at least schematic stock) values found in Matchless Clubman amps which use way higher voltages. Hmm..maybe they get away with it by using an oversized output transformer? Even though the end result is not what I set out to achieve (a cathode biased EL34 amp) I'm extremely happy with how the amp sounds & behaves now. Maybe I'll take another stab at it sometime with a Classictone TW style PT I have handy and a 50 watt output transformer.
You do not have the required permissions to view the files attached to this post.
I thought the amp would sound quieter and more tame and while it may have thedifference was much less dramatic than I thought it would be. I am going to try and experiment with "splitting the difference" and using some higher value individual cathode bias resistors to retain the edginess. But overall, very surprising and encourging results!
