Not to hijack this thread but I've just finished a very similar project; a 1930s Gibson EH150 clone. This amp is using a 408-0-408 PT which is delivering +410VAC and 505VDC directly off of the 5U4 rectifier.
I have 502V on the 2x6L6 plates. I used 200R as a base cathode bias resistor and added a 1k 2W linear pot in series so to be able to dial in the cathode bias as needed. Originally, the amp was set up with 470R screen resistors and had about 495V on the screens. I was measuring about 55mA on a 1R/1% resistor between cathodes and bias resistor under these conditions. The amp is supposed to put out around 30W and has a very similar output section as the 5E5 (0.1uF coupling caps, 1k5 grid resistors). The OT is a Bassman clone and is about 3.6k primary.
I was told by a colleague that it would be best to lower the screens to no more than 370V and strive for a cathode current of no more than 40mA. I have achieved this using 1k screen resistors and a 47k dropping resistor which brought the screen B+ way down to around 360V. I am dropping 1V across the 1k screen resistors. I measured the combined resistance of the 1k pot and 200R and at 40mA it is 1.012k (right on). This 360V screen voltage seems rather conservative, but I don't have any references which show operating conditions for cathode bias AB push/pull under these conditions.
Does 360V on the screens seem right to you guys?? Too conservative? Just right? Complete nonsense? At 40mA, 40V on the cathodes, we have:
Colossal wrote:Not to hijack this thread but... I don't have any references which show operating conditions for cathode bias AB push/pull under these conditions...
That's why I'm here too... So hijack away... it's really annoying that there isn't any reference material - well none that I can find - about this very subject. I've kind of accepted that there are certain parameters within which you can make a very good sounding amp and that - well it would seem anyway - the further you stray from those the less firm the footing becomes...
When I started this build I knew in my heart of hearts that if it were possible to get a lot more power out of the 5E5 that either a.) someone had done it and it sounded awful or b.) it just couldn't be done... but it's been bugging me for a year now so I just thought: if I bolt the iron to the chassis and make a start I'd - at the very least - find out just how far it could be pushed if not anything else. I have a good few customers who are happy with the sound of the 28W 5E5 types I've built for them - but some, no doubt because they can easily be 'monstered' by other amps with less tone but more power, have asked if I can build something with the same tone but more power. So here I am - having a go
You would think there was some work-around but I wouldn't be surprised if there isn't - given what I've read so far. That is - however - what it's all about and if so - then I'm very happy to have gotten the 28W I've managed so far...
Any road - tomorrow's another day....
If I could find a road to get away it wouldn't be too soon...
Colossal - check out the Valvewizard site, it's great!
There are two reasons to lower our screen voltage:
1: To keep the screens from drawing excess current off the cathode at the point in your duty cycle where the plates are approaching zero volts. With a 3K6 OT (which is a relatively low primary Z) and your high idle voltage, your screens are very safe in this regard. The 0 volt grid line will cut off your tubes long before they draw excess screen current, even with the screens at 400 volts or more. By the way, you should measure the screen and plate voltages relative to the cathode, not ground, if you're not already.
2: The other reason is to not exceed the max idle screen voltage or dissipation. 6L6GC's are rated for 450 volts and 5 watts, so again, you should be fine.
Having said all that, I don't know your colleague's credentials and he may well know something I don't. But with a 3k6OT, I would run those screens hot. The only way to know things are really going to be alright is to hook your amp up to a signal generator and measure screen current as you crank 'er up.
Rob - if you really want the same amp with more power, get a higher current PT (250ma+), halve your primary Z and just run 4 5881's instead of 2 in your output section. You could also run a parallel rect tube to maintain your dynamics.
Hope that helps!
Life is a tale told by an idiot -- full of sound and fury, signifying nothing.
Alexo wrote:Colossal - check out the Valvewizard site, it's great!
Hi Alexo, yes, I've visited his site many times and read his preamp book about 6 times! Great information.
Alexo wrote:There are two reasons to lower our screen voltage:
1: To keep the screens from drawing excess current off the cathode at the point in your duty cycle where the plates are approaching zero volts. With a 3K6 OT (which is a relatively low primary Z) and your high idle voltage, your screens are very safe in this regard. The 0 volt grid line will cut off your tubes long before they draw excess screen current, even with the screens at 400 volts or more. By the way, you should measure the screen and plate voltages relative to the cathode, not ground, if you're not already.
2: The other reason is to not exceed the max idle screen voltage or dissipation. 6L6GC's are rated for 450 volts and 5 watts, so again, you should be fine.
Having said all that, I don't know your colleague's credentials and he may well know something I don't. But with a 3k6OT, I would run those screens hot. The only way to know things are really going to be alright is to hook your amp up to a signal generator and measure screen current as you crank 'er up.
I feel fairly comfortable with the calculations and I have Kuehnel's Power Amp book which is detailed in setting up the analysis. I agree with you about the safety margin as does a member here I've been communicating with on the matter. I wanted to run it by the group here to see what others think. My colleague is experienced for sure but these numbers struck me as very conservative and the output is running about 18W for a 2x6L6 amp! The original EH150 is running high plate and screen voltages and supposedly about 30W of output. His assertion was that today's power tube screens are not up to the task and that amps of old should not be cloned verbatim as their designs were based largely on some guesswork (I am quoting here). So I'm not trying to debunk or invalidate his comments, but merely check the math. The important thing is safety and reliability of course, but I would be inclined to run the screens hotter than they are just on inspection alone. This would result in higher cathode current and the assertion was that the amp should not exceed 40mA. This certainly limits the output power. It's not a huge deal, I am just looking to get it right and the tone is what ultimately matters. The amp is going to its new owner tomorrow so if he feels that it gets dirty to quickly, I'll up the screen voltage slightly for a bit more headroom.
VacuumVoodoo wrote:You're welcome, Rob. Very cool factory location. Me like!
While I'm at the keyboard: IMO common cathode resistor in PP AB1 PA is ok IF you don't expect to ever run your power tubes into overdrive and grid cut-off. Consider what happens when one tube is driven hard and the other at same time goes into cut-off. What happens to potential on cathode resistor and what does it do to the tube that is overdriven? You may want to consider separate cathode resistors with not too big decoupling capacitors. I'd choose these caps to give the PP tube pair same 3db LF limit as the OT. Simple and effective prevention of flatulent low end.
Good Luck!
Could you explain this more detailed to me ??
What are the benefits of seperate cathode resistors in overdriven pp stages ?
thank you
So hijack away... it's really annoying that there isn't any reference material - well none that I can find - about this very subject. I've kind of accepted that there are certain parameters within which you can make a very good sounding amp and that - well it would seem anyway - the further you stray from those the less firm the footing becomes...