Banet .I will get shot for this. Citroen GS was 4 cylinder.
Would a KT77 suit 300 V better?
I know this is off subject. I scares me.
http://www.mellowacoustics.com/articles/Tim_Mellow_25w_OTL_Tube_amplifier.pdf
Would a KT77 suit 300 V better?
I know this is off subject. I scares me.
http://www.mellowacoustics.com/articles/Tim_Mellow_25w_OTL_Tube_amplifier.pdf
To be honest, the only reason I pick 300V is because I thought it might be safer and simpler for my first attempt at building a valve amp. But I guess bumping the plate voltage up wouldn't make the design anymore complicated.
Do you reckon 300V is not suitable for a EL34? If so what would be your plate voltage recommendation for the EL34 without adding too much complication and hazard to the design? My design is a pull-push bias in class A.
Last edited:
I am so sorry my PP EL34 amp is in Nottingham and me in Oxford. If here I would have built it for you so as to get a start.
I am 99% certain 300V is fine. As said before valves are not the highest fidelity so why be too worried? If the output transformer is OK for 450V you can bump the supply. To be clear we don't really need better fidelity. Valves can have better micro detail in the modern way of saying it. A good amplifier will not have any obvious distortion and only be euphonic when we also are. That is when a bit too loud. Many low distortion transistor amps sound nothing like real music to me. A Quad 303 is excellent so don't take me to be a valve only person. 80% the other way.
The most exciting thing is to also try 4 x EL84 . They suit 43% UL quite well. It should give more power than thought as most speakers dip below their nominal resistance. I notice these days EL84 costs more than it did. Still worth considering. 43% if right is 3/7 winding ratio in the transformer and nothing mysterious. If separate bobbins that is so easy to do.
2 x EL34 300V triode would be fine. Same transformer also. I built a good quality fixed bias pentode amp. 48 watts pentode, 28 watts UL , 14 watts triode. The triode went louder through real speakers. By louder I mean you wanted to listen that loud. The pentode amp was like the cheaper transistor amps. The problem was this amp didn't like loop feedback. My tests usually are at 1 and 5 watts. An amp I built recently was Jean Hiraga at 5 watts and conventional wisdom valve amp at 1 watt. That is only 0.2% second harmonic shown. On both tests ideal. That amp had no loop feedback. It uses something like UL input to output. Old valves used in testing phase to be sure it will work regardless of tube age. It was set to standard load line example then moved away. EL34 SE UL80 % pentode input sounding better than the 300B amp being compared. Better being more lively and real. It was at 460V 60 mA with 33V 560R bias.
Sorry we said so much and yet so little. I searched for hours elsewhere to find an answer for you. If I had a curve tracer that would have been fun. I would have liked to test at 200V and see if that was impossible.
You could let us know when you start. I bet when you finish UL and 300V won't interest you too much even if that is the choice. EL 34, KT77, KT66, 6L6GC ( 6L6 ) , 5881,and I am sure many many more come to mind , 300 V should give you more choice. Even though pentode or beam tetrode they mostly can be exchanged. EL34 is a heptathlon competitor (multiply purpose). For that alone it is the winner. In my opinion it is the best device ever made and EL 84 even better if wanting UL. I have heard the famous Marantz Model 9 reconfigured to 300B against Sid Smiths the designers advice and overseen by him. The EL34 was better yet again. The 300B sounded dark and ponderous. When used as a driver valve it doesn't.
If you allow enough heater current KT88 and 6550A can be tried. To save you money only if you can borrow them. Used DC heaters only to protect the valves and no other reason. I only got 2 dB over hum bucked AC and the spectrum was worse at one important point. 2N3055 makes a cheap regulator and is tough. Zener voltage about 7V if 6.3 V heaters.
One of the most curious things is an EL34 mostly is an EL34 and not a KT88. When that is understood the differences between EL34's is surprisingly large when listening. One could be forgiven to say almost as large as between 34 and 88. This is when every attempt has been made to get rid of errors. The 88 mostly gives less distortion at the same power. It sounds to me less good as the spectrum is not quite to the Hiraga ideal. Nothing obvious and hard to pin down. At clipping the EL 34 maintains it's distortion curve right up to the point where it is 5% THD. I hate saying clips better as that is 0% what I am saying. The clipping is the sound somehow at lower levels. I would suggest we anticipate the curve by inferring it.
That's what we all need . Local friends valve bank. Anyone near Oxford got EL156?
I am 99% certain 300V is fine. As said before valves are not the highest fidelity so why be too worried? If the output transformer is OK for 450V you can bump the supply. To be clear we don't really need better fidelity. Valves can have better micro detail in the modern way of saying it. A good amplifier will not have any obvious distortion and only be euphonic when we also are. That is when a bit too loud. Many low distortion transistor amps sound nothing like real music to me. A Quad 303 is excellent so don't take me to be a valve only person. 80% the other way.
The most exciting thing is to also try 4 x EL84 . They suit 43% UL quite well. It should give more power than thought as most speakers dip below their nominal resistance. I notice these days EL84 costs more than it did. Still worth considering. 43% if right is 3/7 winding ratio in the transformer and nothing mysterious. If separate bobbins that is so easy to do.
2 x EL34 300V triode would be fine. Same transformer also. I built a good quality fixed bias pentode amp. 48 watts pentode, 28 watts UL , 14 watts triode. The triode went louder through real speakers. By louder I mean you wanted to listen that loud. The pentode amp was like the cheaper transistor amps. The problem was this amp didn't like loop feedback. My tests usually are at 1 and 5 watts. An amp I built recently was Jean Hiraga at 5 watts and conventional wisdom valve amp at 1 watt. That is only 0.2% second harmonic shown. On both tests ideal. That amp had no loop feedback. It uses something like UL input to output. Old valves used in testing phase to be sure it will work regardless of tube age. It was set to standard load line example then moved away. EL34 SE UL80 % pentode input sounding better than the 300B amp being compared. Better being more lively and real. It was at 460V 60 mA with 33V 560R bias.
Sorry we said so much and yet so little. I searched for hours elsewhere to find an answer for you. If I had a curve tracer that would have been fun. I would have liked to test at 200V and see if that was impossible.
You could let us know when you start. I bet when you finish UL and 300V won't interest you too much even if that is the choice. EL 34, KT77, KT66, 6L6GC ( 6L6 ) , 5881,and I am sure many many more come to mind , 300 V should give you more choice. Even though pentode or beam tetrode they mostly can be exchanged. EL34 is a heptathlon competitor (multiply purpose). For that alone it is the winner. In my opinion it is the best device ever made and EL 84 even better if wanting UL. I have heard the famous Marantz Model 9 reconfigured to 300B against Sid Smiths the designers advice and overseen by him. The EL34 was better yet again. The 300B sounded dark and ponderous. When used as a driver valve it doesn't.
If you allow enough heater current KT88 and 6550A can be tried. To save you money only if you can borrow them. Used DC heaters only to protect the valves and no other reason. I only got 2 dB over hum bucked AC and the spectrum was worse at one important point. 2N3055 makes a cheap regulator and is tough. Zener voltage about 7V if 6.3 V heaters.
One of the most curious things is an EL34 mostly is an EL34 and not a KT88. When that is understood the differences between EL34's is surprisingly large when listening. One could be forgiven to say almost as large as between 34 and 88. This is when every attempt has been made to get rid of errors. The 88 mostly gives less distortion at the same power. It sounds to me less good as the spectrum is not quite to the Hiraga ideal. Nothing obvious and hard to pin down. At clipping the EL 34 maintains it's distortion curve right up to the point where it is 5% THD. I hate saying clips better as that is 0% what I am saying. The clipping is the sound somehow at lower levels. I would suggest we anticipate the curve by inferring it.
That's what we all need . Local friends valve bank. Anyone near Oxford got EL156?
Maybe is good just to ad this graph for 2 x EL34 ( 6CA7) working UL in A1 class and supplied from 450VDC B+ , open loop characteristic of course without any GNFB applied .
What we can see is that Dtot is under 1% up to 30W output , where D2 dominate up to 12W , from that point D3 start to dominate up to around 22W , and than start to fall , where D2 takeover again up to max.out 37W but never exceed 0,8 % , finally D5 appear to be generated only from 14W up .
What we can see is that Dtot is under 1% up to 30W output , where D2 dominate up to 12W , from that point D3 start to dominate up to around 22W , and than start to fall , where D2 takeover again up to max.out 37W but never exceed 0,8 % , finally D5 appear to be generated only from 14W up .
Attachments
Last edited:
That's the sort of graph I like . Never seen d5 like that and it is an excellent way to get me to look. Must say my own designs have not asked it to appear. d3 opposite which I have seen. I use pre-distorion and make sure it is reliable by not asking too much. This must be why my amp sound so open. It is tempting to think pre-distortion is adding sugar. It isn't. It is just maths. UL is a similar thinking ( not the same). It is said d5 must never exceed - 70dB. d2- 30 dB. d3 must not be rejected. - 40dB would about right and - 50 db 4th.
FYI
"Mullard quote the power output for the EL34 tube as being the same for pentode and ultra-linear connection with 20% turns screen-taps, but for minimum distortion the screen-taps increase to 43% turns, which provides a power reduction of 15% (but distortion is halved)." - from Grimwood.
"Mullard quote the power output for the EL34 tube as being the same for pentode and ultra-linear connection with 20% turns screen-taps, but for minimum distortion the screen-taps increase to 43% turns, which provides a power reduction of 15% (but distortion is halved)." - from Grimwood.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.