Hi,
I bought a few days ago an old Push Pull tube amp from a console and I intend to build something fun with it.
The single PT uses SS diodes and provides a B+ of about 450-475Vdc
The OPT are rated 40W, 4K ohm with 4, 8 and 16 ohm output. These OPT do NOT have UL tap. The OPT used to drive 7027A tubes.
I don’t like the sound of pure penthode without global feedback so I guess I’m looking at a triode strapped scenario of something similar. I don’t mind using tubes that were not designed for audio, I have had good success with sweep tube and the like, 807, etc. I’m also open for suggestion with using Russian tubes either as output tubes and/or driver. I do enjoy the 6P3S-e in one of my SE amp.
I’m aware that the usual tubes for 4k push pull is EL34 but would there another tube I could use based on your experience ? Most of my tubes stuff are SE power amp except for a push pull class A EL84 which provides 10W/ch, I really enjoy it but would like power on my 4 ohm speakers.
What I’d like to build now is a more powerful amp, something with guts 😉
Thanks in advance for some recommandations.
Eric
I bought a few days ago an old Push Pull tube amp from a console and I intend to build something fun with it.
The single PT uses SS diodes and provides a B+ of about 450-475Vdc
The OPT are rated 40W, 4K ohm with 4, 8 and 16 ohm output. These OPT do NOT have UL tap. The OPT used to drive 7027A tubes.
I don’t like the sound of pure penthode without global feedback so I guess I’m looking at a triode strapped scenario of something similar. I don’t mind using tubes that were not designed for audio, I have had good success with sweep tube and the like, 807, etc. I’m also open for suggestion with using Russian tubes either as output tubes and/or driver. I do enjoy the 6P3S-e in one of my SE amp.
I’m aware that the usual tubes for 4k push pull is EL34 but would there another tube I could use based on your experience ? Most of my tubes stuff are SE power amp except for a push pull class A EL84 which provides 10W/ch, I really enjoy it but would like power on my 4 ohm speakers.
What I’d like to build now is a more powerful amp, something with guts 😉
Thanks in advance for some recommandations.
Eric
Is that 30 to 40 watts per channel, or total for the stereo amp?
What size is the power transformer? Hopefully it's huge. 🙂
What size is the power transformer? Hopefully it's huge. 🙂
Attached is an old marketing pamphlet intended to tout the benefits of RCA tubes, and this PDF also contains example schematics. Look at the 30 watt amplifier on pages 12 and 13. This circuit uses the 7027 output tube so it may work with the output transformer you have. The input/phase splitter tube is the now expensive and hard-to-get 7199 but you can substitute the 6GH8 for it. The 6GH8 has a different pin-out but if you are wiring point-to-point this won't matter much. Note that this design does use global negative feedback.
Maybe this at least gives you a place to start.
Maybe this at least gives you a place to start.
The 50-Watt Amplifier on pp. 14, 15 in that pamphlet has possibilities! It uses shunt (so-called "Schade") feedback (R21 and R24), as well as feedback from the 7027A plates to the driver stage 6CB6 cathodes (R22 and R23). Use 6EW6 or 6EJ7 instead of 6CB6 for more gain, and perhaps use a 1:1 input transformer to split phase instead of the 7199 cathodyne? The 7027As have 450V on the plates. The 400V screen supply using 0A2 and 6DR7 is probably more trouble than it's worth. Perhaps use a small solid-state regulator or 51V 1W zener diodes instead...
Is that 30 to 40 watts per channel, or total for the stereo amp?
Power transformer is huge.
Yep, 30-40 per channel 😀
Power transformer is huge.
Ok, so far thanks for the nice document and recommendations. Please note that the amp came without tubes therefore I don’t have any 7027A.
Thanks
Thanks
7027A was basically a beefed up 6L6GC. so you can use 6L6GC with no worries. You could also use Russian 6P3S-E, which is close to a 6L6GC.
Max plate dissipation for type 7027 is 25W. Max plate dissipation for 7027A is 35W. Max plate dissipation for 6L6GC is 30W. Max plate dissipation for 6P3S-E is shown on the Russian data sheet as 20.5W, but people regularly abuse that limitation and the tubes seem to do fine.
Back in 2010. Tubelab and chrish collaborated on a design for a Class AB2, push-pull 6L6GC amplifier. The schematic is in this post:
https://www.diyaudio.com/community/threads/6l6gc-ab2-amp.133034/page-23#post-2250447
(The thread is worth reading. There's a lot of good discussion that leads up to this design.)
It's a 3-stage design, with the first stage being a long-tailed pair phase splitter, DC-coupled to a differential pair driver stage, AC-coupled to MOSFET source followers which are DC-coupled to the push-pull 6L6GC Class AB2 output stage. The idea is to drive the output stage from a low impedance/high current stage so that the 6L6GCs can operate with their grids driven a bit positive.
They did use global NFB from the output transformer secondary to the input stage.
The B+ is 400V from a Maida regulator with one big honkin' heat sink on its pass device.
Tubelab (George) is adept at getting the maximum power out of output tubes. I'm sure this is a very well thought out design. Worth a look, I think.
Max plate dissipation for type 7027 is 25W. Max plate dissipation for 7027A is 35W. Max plate dissipation for 6L6GC is 30W. Max plate dissipation for 6P3S-E is shown on the Russian data sheet as 20.5W, but people regularly abuse that limitation and the tubes seem to do fine.
Back in 2010. Tubelab and chrish collaborated on a design for a Class AB2, push-pull 6L6GC amplifier. The schematic is in this post:
https://www.diyaudio.com/community/threads/6l6gc-ab2-amp.133034/page-23#post-2250447
(The thread is worth reading. There's a lot of good discussion that leads up to this design.)
It's a 3-stage design, with the first stage being a long-tailed pair phase splitter, DC-coupled to a differential pair driver stage, AC-coupled to MOSFET source followers which are DC-coupled to the push-pull 6L6GC Class AB2 output stage. The idea is to drive the output stage from a low impedance/high current stage so that the 6L6GCs can operate with their grids driven a bit positive.
They did use global NFB from the output transformer secondary to the input stage.
The B+ is 400V from a Maida regulator with one big honkin' heat sink on its pass device.
Tubelab (George) is adept at getting the maximum power out of output tubes. I'm sure this is a very well thought out design. Worth a look, I think.
Last edited:
I will read the thread, thanks.
I’ve read many threads by George, especially the one where he’s torturing some 6AV5GA. These Sylvania are tough.
The schematic is using UL tap but my OPT don’t have any so the sound won’t be the same but it’s seems like a very well designed power amp.
I’ve read many threads by George, especially the one where he’s torturing some 6AV5GA. These Sylvania are tough.
The schematic is using UL tap but my OPT don’t have any so the sound won’t be the same but it’s seems like a very well designed power amp.
The driver circuitry is of interest. I'll bet it would work well for triode-wired 6L6GCs.
Simpler versions of that driver have been used successfully. Like this one from the 1990s...
Simpler versions of that driver have been used successfully. Like this one from the 1990s...
But note that this design requires two negative voltage supplies (-60 and -80) in addition to the output stage bias voltage (-56). That introduces another level of complexity. It might be better to start with something more "mainstream" unless the OP is an experienced amplifier builder. Just a suggestion.
That's a good point, Ray.
I find that the more power you try to squeeze out of the tubes, the more complex things tend to get.
I wasn't suggesting building this exact design, just floating the idea of that basic topology (LTP phase splitter input stage > push-pull differential driver stage > push-pull output stage).
I find that the more power you try to squeeze out of the tubes, the more complex things tend to get.
I wasn't suggesting building this exact design, just floating the idea of that basic topology (LTP phase splitter input stage > push-pull differential driver stage > push-pull output stage).
The EL 36 is certainly possible. PP it delivers 44w at Va=300 and Vg2=150v (Google Philips schematic EL 6415). Your 180v ac is a little too high so you have to lower that.
You can use zeners here. Series connect them with your supply voltage to get the desired 150v and if you're uncertain about their wattage give each g2 it's own zener.
I assume your pt has a separate winding for Vg1; Adjust Ia by making -Vg1 variable. Connect 10 ohm resistors from cathode to ground. Bias them at 20ma (= 0,2v).
Va can be a lot higher than 300v ; these tubes can stand relatively high anode voltages so you don't have to worry about that. Anode current is mainly determined by Vg2 and not by Va.
You can use zeners here. Series connect them with your supply voltage to get the desired 150v and if you're uncertain about their wattage give each g2 it's own zener.
I assume your pt has a separate winding for Vg1; Adjust Ia by making -Vg1 variable. Connect 10 ohm resistors from cathode to ground. Bias them at 20ma (= 0,2v).
Va can be a lot higher than 300v ; these tubes can stand relatively high anode voltages so you don't have to worry about that. Anode current is mainly determined by Vg2 and not by Va.
If you can locate them, there were 6BG6GA tubes with 7027a guts. These were typically Philips/ECG, Sylvania, and I've seen some GE tubes as well. These are notable by large grey plates and large, black, screen grid radiators above the plates. The pin out is for the 6BG6GA.
Hi all
Thanks for the recommendation.
I’ve decided to go with the 30W from the RCA catalog provided by Ray. I like the fact that it’s simple, has the complete schematic with voltages that I have to aim for and also a BOM.
I have 2 questions.
1. Should I buy a quad of 6P3S-e or is their wattage to low for this amp..?
2. Any recommendations to maybe improve the front end ?
Besides the global feedback, I will probably add some local feedback à la Schade.
I’m adding a pic of an amp identical to the one I have. I believe the PT will be OK.
Thanks
Eric
Thanks for the recommendation.
I’ve decided to go with the 30W from the RCA catalog provided by Ray. I like the fact that it’s simple, has the complete schematic with voltages that I have to aim for and also a BOM.
I have 2 questions.
1. Should I buy a quad of 6P3S-e or is their wattage to low for this amp..?
2. Any recommendations to maybe improve the front end ?
Besides the global feedback, I will probably add some local feedback à la Schade.
I’m adding a pic of an amp identical to the one I have. I believe the PT will be OK.
Thanks
Eric
Attachments
Unless you are experienced with modifying existing amplifier designs, I would recommend starting by building this amp exactly as designed, with the 7027 output tube and no circuit changes other than substituting the 6GH8 tube for the 7199. The 7027 tube is available as both NOS and new production so it is easy to source, and the 6GH8 is a proven substitute for the 7199 and is commonly used to restore the Dynaco ST-70. As I mentioned in my first post, the pin-out is different between the 6GH8 and the 7199 but that is a difference that is easy to accommodate.
After you have built a design that is known to work, and which works properly for you, you can then experiment with different tubes, a different feedback scheme, and anything else. Walk before you run. 🙂
Good luck.
After you have built a design that is known to work, and which works properly for you, you can then experiment with different tubes, a different feedback scheme, and anything else. Walk before you run. 🙂
Good luck.
About question 1), yes, the 6P3S-E is powerful enough for a 30W push pull amp. I built one using those tubes, following the EICO HF-89, 405V B+, ultra-linear, fixed bias, I get 28W with the tubes biased at 50mA.
I don't think you have to buy a quad of current production 7027A, even if you build the 30 watt amp exactly as pictured. A current production 7027A is going to be as different from a ca. 1965 production 7027A as will be a current production 6L6GC from Electro-Harmonix, Svetlana or whomever. I agree the 6P3S-E will work well too.
If you compare the data sheets of 6L6GC and 7027A, you'll see how similar they are.
6L6GC RCA data: https://frank.pocnet.net/sheets/049/6/6L6GC.pdf
7027A RCA data: https://frank.pocnet.net/sheets/049/7/7027A.pdf
Typical pentode operating conditions for 6L6GC in push-pull class AB1:
Vp = 450V
Vg2 = 350V
Vg1 = -30V
Ip = 95mA (for two tubes)
Ig2 = 3.4mA (two tubes)
Load (p-p) = 6000 ohms
Typical pentode operating conditions for 7027A in push-pull class AB1:
Vp = 450V
Vg2 = 350V
Vg1 = -30V
Ip = 95mA (two tubes)
Ig2 = 3.4mA (two tubes)
Load (p-p) = 6000 ohms
Why, they're identical!
The difference is in the maximum ratings.
6L6GC Pdiss max = 30W
7027A Pdiss max = 35W
If you compare the data sheets of 6L6GC and 7027A, you'll see how similar they are.
6L6GC RCA data: https://frank.pocnet.net/sheets/049/6/6L6GC.pdf
7027A RCA data: https://frank.pocnet.net/sheets/049/7/7027A.pdf
Typical pentode operating conditions for 6L6GC in push-pull class AB1:
Vp = 450V
Vg2 = 350V
Vg1 = -30V
Ip = 95mA (for two tubes)
Ig2 = 3.4mA (two tubes)
Load (p-p) = 6000 ohms
Typical pentode operating conditions for 7027A in push-pull class AB1:
Vp = 450V
Vg2 = 350V
Vg1 = -30V
Ip = 95mA (two tubes)
Ig2 = 3.4mA (two tubes)
Load (p-p) = 6000 ohms
Why, they're identical!
The difference is in the maximum ratings.
6L6GC Pdiss max = 30W
7027A Pdiss max = 35W