Greetings everyone,
Given the tubes that I can use have roughly the same power output and plate voltage requirements, which would be wiser to build, P-P Triode (809) in Class B or P-P Pentode (6146B) in Class AB2? At this point I need to order the output xfmrs and my intention was to go with Edcor.
Thanks,
Ray
Given the tubes that I can use have roughly the same power output and plate voltage requirements, which would be wiser to build, P-P Triode (809) in Class B or P-P Pentode (6146B) in Class AB2? At this point I need to order the output xfmrs and my intention was to go with Edcor.
Thanks,
Ray
I have a few 809's. They are nice looking tubes, and I have a suspicion that they would work well in class A2 or AB2, but not at 100 watts. I have not had the time to play with them, but I did stuff one into a Tubelab SE a few years ago and got 5 or 6 watts. I also think that it will be hard to get a 100 watt class B amp sounding good at the 1 watt level, which is probably the average power level it will see unless you have bricks for speakers. SO...
That leaves sweep tubes or the previously mentioned 6146's. I think that 100 watts from a pair of 6146's is squeezeing them a bit too hard, but again I haven't tried it yet. I have been working on several big power experiments, and there are some 6146's sitting on my workbench, but I have been at work for 12 hour days, 7 days a week lately, so....no tube time.
I have found a few minutes here and there to play. I bought a bunch of NOS 35LR6's a few months ago, and I finally got around to stuffing 4 into the red board and setting the power supply on KILL. The result, 250 WPC. Yes thats 500 watts from Petes big red board, on lab type power supplies. I also found some real cheap power transformers on Ebay, so the obvious experiment is to connect two of them to the red board, stand behind something substantial and hit the switch. It will happen when there is time.
Seriously, sweep tubes make 100 watts relatively easy. Find out which flavors you have several of. You need 4 good ones. With used tubes I have needed as many as 10 to find 2 pairs that are even close enough to make work. I got about 50 NOS 35LR6's and it took 5 of them to find 4 that played nice with each other.
If you have tubes that are compatible with Petes board, that is the easiest way to get there. Some modification of the power supply circuitry is needed. I have been using lab type supplies, but now that I have cheap transformers, I will figure that out too. I still plan to design my own big power amp, but it won't happen soon.
http://www.diyaudio.com/forums/tubes-valves/169867-10-pounds-power-15-a.html
That leaves sweep tubes or the previously mentioned 6146's. I think that 100 watts from a pair of 6146's is squeezeing them a bit too hard, but again I haven't tried it yet. I have been working on several big power experiments, and there are some 6146's sitting on my workbench, but I have been at work for 12 hour days, 7 days a week lately, so....no tube time.
I have found a few minutes here and there to play. I bought a bunch of NOS 35LR6's a few months ago, and I finally got around to stuffing 4 into the red board and setting the power supply on KILL. The result, 250 WPC. Yes thats 500 watts from Petes big red board, on lab type power supplies. I also found some real cheap power transformers on Ebay, so the obvious experiment is to connect two of them to the red board, stand behind something substantial and hit the switch. It will happen when there is time.
Seriously, sweep tubes make 100 watts relatively easy. Find out which flavors you have several of. You need 4 good ones. With used tubes I have needed as many as 10 to find 2 pairs that are even close enough to make work. I got about 50 NOS 35LR6's and it took 5 of them to find 4 that played nice with each other.
If you have tubes that are compatible with Petes board, that is the easiest way to get there. Some modification of the power supply circuitry is needed. I have been using lab type supplies, but now that I have cheap transformers, I will figure that out too. I still plan to design my own big power amp, but it won't happen soon.
http://www.diyaudio.com/forums/tubes-valves/169867-10-pounds-power-15-a.html
PP 809s are rated at 100W+ for Class B operation. However, that's for AM modulator duty where fidelity is neither necessary nor legal. The cross over distortion that's inherent in Class B sounds really nasty. Also, these AM modulators frequently included an LPF between the modulator output and the AM xmtr for that very reason: to reject those nasty harmonics and to keep the AM signal within legal limits. It won't work at all well for an audio amp if you're looking for sonic performance and not just lotsawatts.
If you go with the 809s, you'd be better off accepting less power out, and improve the sonic performance.
As for 6146s, you have the same problem: intended for modulator useage, lotsawatts, and not fidelity. That means operating deep into Class AB territory, and, yuppers, more X-over distortion. Probably best to try new loadlines that get 'em more towards Class A operation, and accept the reduced power out in exchange for better sonics.
Sure. Pentodes with local feedback (last Forum fashion, code name Schade) may sound cleaner than triodes.
I have both in Pyramid-V (6N6P LTP driver), Pyramid-VII (6P15P LTP driver), Pyramid-8 (6J51P LTP driver), with GU50 outputs, all sound fantasiche, like SE on pianissimo and SS on fortissimo.
Miles, I disagree with your statements.
In AM broadcast service, fidelity is permitted and desirable. There are Gates, RCA, Continental and other brands in the <1kW class, fyi. (actually, the FCC just hacked the AM band down and restricted AM broadcasts to 5kHz, but historically many AM transformers were good to past 20Khz.)
In HAM Radio service, fidelity is permitted and is actually highly sought after, especially in 75M & 160M operation. For those interested, please check out The AM Forum - Index
The so-called "splatter filters" were mostly for rigs that used clipping in the earlier modulator stages. Modern test methods show that these filters (high level, at the modiron) actually often don't operate as expected and add unwanted harmonics... afaik.
Also, back in the day, it was possible to "hit the baseline" (modulate negative 100%) and THAT produced splatter... modern rigs (ham rigs) have a circuit that prevents ever hitting the baseline - thus no splatter filters... but they have nothing to do with "fidelity" (bandwidth) in this case.
And actually, xover distortion in a modulator that is running at ~90%+ level all the time is really not particularly audible after it is used to modulate the class C final stage... in hi-fi applicaitions, few tubes are in reality at dead-bang Class B, they're usually slightly into AB anyhow... like a typical 811A for example...
The 6146 does not produce much more power, if any at all, than a 6550, KT88 or 6CA7... but it can handle a bit more B+, but that implies a higher plate impedance, and then a different xfrmr...
If you want to run these tubes in Class A one does have to drop the B+ and run the current higher... but these are not great tubes for this application. One probably wants to look back at the 211 & 845 for Class A operation.
At this point in hi-fi history, I think just about every combination of every tube has been tried! You should be able to find amps all over the place with examples run in all sorts of classes of operation...
_-_-bear
PS. I do be a ham radio operator...
Last edited:
I have heard good things about the 6DQ5, so I started collecting them whenever I could get them at hamfests for $1. I have about 20 of them now, but haven't cranked them up yet. They are rated for 24 watts of plate dissipation.
The 21LG6 is a 28 watt rated sweep tube. I have not tested any of these (don't have any). It has a 12 pin compactron base and can possibly be operated in the red board.
The 6MG6 is another number for the 6LQ6/6JE6C. This is the big boy in your collection. They are rated for 30 watts and some versions have heat radiating fins welded to the plate for even more power. They are also quite valuable to owners of CB and ham radio amplifiers that eat tubes like candy.
By the book you would need a design that achieved 68% efficiency to get 100 watts out of a pair of 24 watt tubes. 68% is doable in P-P and I have measured in the low 70's on the red board. I have also cranked 6HJ5's to 100 WPC in the red board. They are 24 watt tubes. In reality you aren't going to run the thing at 100 watts continuously, so any of these tubes should do fine. I cranked the 6HJ5's till they started to glow and I was at 150 watts.
Many of the older transmitters used 833A's in the RF finals, and in the modulator. Even if the transmitter was designed to be HiFi (by AM standards) crossover distortion is not an issue. In order to be "louder" than the other guy's radio station the broadcasters employed a good bit of compression on the modulation to keep the average modulation percentage up. The modulator spends virtually no time at low power except during total silence.
A home HiFi set will see the opposite. The peak to average ratio for most music is at least 20db, usually more. This means that an amp rated for 100 watts operated such that it hits clipping on peaks will be putting out 1 watt on average. Make the "first watt" count. It's the one you will be listening to!
I agree that most "class B" amps run in class AB, and the data sheets point that out. I have also found that sweep tubes can run at lower idle currents than KT88's for the same distortion figures. That is the key to making an efficient amp.
Yeah, Ampeg had to learn this the hard way! I would not try to crank 100 watts from a pair of 6146's for more than a few seconds. It is fairly well known that most sweep tubes have rather conservative ratings. After all they run at full power all of the time, and many homes only had 1 TV back then. When was it NOT turned on? When no one was home.
So be I.
The 21LG6 is a 28 watt rated sweep tube. I have not tested any of these (don't have any). It has a 12 pin compactron base and can possibly be operated in the red board.
The 6MG6 is another number for the 6LQ6/6JE6C. This is the big boy in your collection. They are rated for 30 watts and some versions have heat radiating fins welded to the plate for even more power. They are also quite valuable to owners of CB and ham radio amplifiers that eat tubes like candy.
By the book you would need a design that achieved 68% efficiency to get 100 watts out of a pair of 24 watt tubes. 68% is doable in P-P and I have measured in the low 70's on the red board. I have also cranked 6HJ5's to 100 WPC in the red board. They are 24 watt tubes. In reality you aren't going to run the thing at 100 watts continuously, so any of these tubes should do fine. I cranked the 6HJ5's till they started to glow and I was at 150 watts.
Many of the older transmitters used 833A's in the RF finals, and in the modulator. Even if the transmitter was designed to be HiFi (by AM standards) crossover distortion is not an issue. In order to be "louder" than the other guy's radio station the broadcasters employed a good bit of compression on the modulation to keep the average modulation percentage up. The modulator spends virtually no time at low power except during total silence.
A home HiFi set will see the opposite. The peak to average ratio for most music is at least 20db, usually more. This means that an amp rated for 100 watts operated such that it hits clipping on peaks will be putting out 1 watt on average. Make the "first watt" count. It's the one you will be listening to!
I agree that most "class B" amps run in class AB, and the data sheets point that out. I have also found that sweep tubes can run at lower idle currents than KT88's for the same distortion figures. That is the key to making an efficient amp.
Yeah, Ampeg had to learn this the hard way! I would not try to crank 100 watts from a pair of 6146's for more than a few seconds. It is fairly well known that most sweep tubes have rather conservative ratings. After all they run at full power all of the time, and many homes only had 1 TV back then. When was it NOT turned on? When no one was home.
So be I.
At one time I had hoped to build a 6146A/b Amp as I got a couple from my dad before he passed on, but after researching it I found there are several issues with the 6146 variants, some of them like to self destruct in RF circuits. I suspect they aren't all that stable in AF circuits or there would have been more designs to be found.
I've scaled back to 6L6 class using Russian tubes.
I've scaled back to 6L6 class using Russian tubes.
Most of my red board sweep tube experiments have been run at 500 to 650 volts. The idle current has been 30 to 40 mA. the screen voltage is adjustable but 150 seems to be the best place for just about every tube I tried. Higher voltages require more negative bias, which can allow a bit more drive but distortion goes up and the tendency for the tube to runaway gets higher.
Most of my 100 watt experiments have been in the 550 to 600 volts of B+ with a 3300 ohm load. The exact tube choice doesn't seem too important. Of course the bigger tubes allow a bit more current and voltage without getting excited so things aren't stressed at 100 watts. I have used a 2500 ohm load and 650 volts of B+ (the limit of my power supply) to make 250 WPC with 35LR6's.
After everything is working try measuring the distortion at bias currents from 30 to 50 mA. Pick the point where the distortion levels off for best tube life. 48 mA and 600 volts add up to 29 watts. That may be a bit warm.
Most of my 100 watt experiments have been in the 550 to 600 volts of B+ with a 3300 ohm load. The exact tube choice doesn't seem too important. Of course the bigger tubes allow a bit more current and voltage without getting excited so things aren't stressed at 100 watts. I have used a 2500 ohm load and 650 volts of B+ (the limit of my power supply) to make 250 WPC with 35LR6's.
After everything is working try measuring the distortion at bias currents from 30 to 50 mA. Pick the point where the distortion levels off for best tube life. 48 mA and 600 volts add up to 29 watts. That may be a bit warm.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.