The EL36 is fully capable of 20 watts disipation. The surface of the anode is only somewhat smaller then that of the EL34 which is rated at 25 watts.
Moreover the cathode surface on the 36 is larger than that of the 34 which is unsurprising given the fact that the type was designed for pulse operation.
The datasheets for these tubes shows ratings for TV vertical or horizontal deflection use.
Example: 6DQ6A
http://www.mif.pg.gda.pl/homepages/frank/sheets/093/6/6DQ6A.pdf
Shows it is rated at 15 Watts maximum plate dissipation.
In Australia these were mostly made in the Sydney AWV Tube factory who also put out this application note on using them in a Class AB Audio Amp:
http://www.retrovox.com.au/6dq6aamp.pdf
Recommended operating point is at 20W dissipation with a note to not exceed 25W maximum.
This is typical of all this family of tubes and certainly applies to the 6CM5/EL36 which was the other common sweep tube used here in Oz.
Cheers,
Ian
Example: 6DQ6A
http://www.mif.pg.gda.pl/homepages/frank/sheets/093/6/6DQ6A.pdf
Shows it is rated at 15 Watts maximum plate dissipation.
In Australia these were mostly made in the Sydney AWV Tube factory who also put out this application note on using them in a Class AB Audio Amp:
http://www.retrovox.com.au/6dq6aamp.pdf
Recommended operating point is at 20W dissipation with a note to not exceed 25W maximum.
This is typical of all this family of tubes and certainly applies to the 6CM5/EL36 which was the other common sweep tube used here in Oz.
Cheers,
Ian
Thanks for the response, I have some pics but this furom is looking for a url to upload the pics, so I'll have to put them in my dropbox I guess.
More to come.
More to come.
RCA Victor 1952 television receiver
https://www.dropbox.com/sc/zmvkar9yzl2rjjy/4dzv0eUIjI
Just trying to upload pics
?
https://www.dropbox.com/sc/zmvkar9yzl2rjjy/4dzv0eUIjI
Just trying to upload pics
?
No real advantage. Regardless of whether using them as audio or RF finals, you're gonna get north of 100W easy. Any power increase from bumping up against that screen voltage spec isn't gonna make a noticeable difference.
Not really. If using conventional g1 drive, the input voltages are still in the same territory as most audio finals: ~30VP-P.
These TV HD pents don't draw very much grid current, and basically shouldn't be taken into Class *2 anyway. That's how they were designed: to pull bigamps at low VPK's. You can use 'em as Class C1 amps for RF, and 'AB1 amps for audio. Drive requirements aren't all that difficult, and you really don't need source followers for g1 drive. A cathode follower will do nicely for driving the input capacitance and to handle the occasional clip on peaks.
Music reproduction duty is a good deal less demanding than HD duty (or "brick on the key" RF modes like FM and/or packet). As for guitar use, you might want to be a bit more conservative so you don't poof something.
Well that's good news. I'd rather not abuse tubes if there's no point to it (other than the primal satisfaction of watching things melt, of course
). I've read that sweeps like a well-regulated screen supply. How well regulated does it have to be? Would a zener diode, a resistor,and a cap suffice, or is it better to go active? The guitar community seems to frown upon regulated supplies due to less perceived 'dynamics', but in what is likely going to be a high-powered amp which stays relatively clean in order to accommodate a high-gain preamp, I'm unsure of how important this would be in the grand scheme of things.Really now? That's actually quite surprising. I was figuring that the big boys would need at least as much drive as a KT88, and probably more. Still kinda want to try out a pentode LTP, but now more just to see how it sounds more than anything else. It was used to such great effect in the Red Board.
This just keeps getting better and better! Even though it's not absolutely necessary, I'm still leaning towards driving the grids directly with DC-coupled cathode followers just to eliminate blocking distortion, which typically sounds like ***. Plus, if I use a pentode LTP, I can make the entire driver stage out of 2 triode-pentodes, thus keeping with the TV tube theme (and there are so many to choose from!) Although in this case, a high-ish gm pentode and a low-mu, low rp, high gm triode in the same bottle would be ideal. Any suggestions?
Aww, but where's the fun in that?
Last edited:
If you're more interested in guitar amps, maybe ask at the "Instruments and Amps" sub-forum?
All pents that don't use Ultralinear sound best when the screen supply is regulated. Active regulators present the screen with a very low impedance, and this is what you want in a music reproduction amp, which is what I design. A Zener or gas tube regulator is better than none, and active is better still due to the reduced impedance.
A guitar amp is a whole 'nother beast where you like distortion, as "tone" is really distortion and makes the difference between "Marshall sound" or "Fender sound", or why you'd like Amp X if you're playing Metal, or Amp Y for jazz, and so forth. Guitar players, since they like distortion, are more likely to abuse the finals than those who listen to the recordings and want to hear them reproduced as faithfully as possible.
Small signal pents can make for excellent performing LTPs. Here's a loadline for the 6CB6 that could make a very clean LTP. Pents don't have to make lots of nasty high order harmonics. As for the HD finals, these really don't need a whole lot of drive, as they were designed for high sensitivity. This is accomplished by placing the grids quite close to enormous cathodes. This explains both the low screen voltages, and that these really don't need -- or like -- positive control grid voltages. Taking the control grids positive can result in excessive grid currents.
It's the RF finals (types like the 1624 or 814) that really need positive grid voltages as these are designed to pull small Vgk= 0 plate currents to assist in getting them into Class C. Same goes for RF power triodes: most are designed to operate either with no bias, or small bias voltages where the power comes from Class *2 operation.
These triode/pentode types are vertical deflection types like the 6LR8. Most HD combo tubes are a power pent with the damper diode in the same bottle.
All pents that don't use Ultralinear sound best when the screen supply is regulated. Active regulators present the screen with a very low impedance, and this is what you want in a music reproduction amp, which is what I design. A Zener or gas tube regulator is better than none, and active is better still due to the reduced impedance.
A guitar amp is a whole 'nother beast where you like distortion, as "tone" is really distortion and makes the difference between "Marshall sound" or "Fender sound", or why you'd like Amp X if you're playing Metal, or Amp Y for jazz, and so forth. Guitar players, since they like distortion, are more likely to abuse the finals than those who listen to the recordings and want to hear them reproduced as faithfully as possible.
Small signal pents can make for excellent performing LTPs. Here's a loadline for the 6CB6 that could make a very clean LTP. Pents don't have to make lots of nasty high order harmonics. As for the HD finals, these really don't need a whole lot of drive, as they were designed for high sensitivity. This is accomplished by placing the grids quite close to enormous cathodes. This explains both the low screen voltages, and that these really don't need -- or like -- positive control grid voltages. Taking the control grids positive can result in excessive grid currents.
It's the RF finals (types like the 1624 or 814) that really need positive grid voltages as these are designed to pull small Vgk= 0 plate currents to assist in getting them into Class C. Same goes for RF power triodes: most are designed to operate either with no bias, or small bias voltages where the power comes from Class *2 operation.
These triode/pentode types are vertical deflection types like the 6LR8. Most HD combo tubes are a power pent with the damper diode in the same bottle.
Attachments
@smoking-amp, can you look at the 829b, 4D32 and the QE08-200 for comparison with tv tubes wrt to screen grid drive?
http://www.r-type.org/pdfs/829b.pdf
http://www.r-type.org/pdfs/4d22.pdf
http://www.rell.com/filebase/en/src/Datasheets/7378_QE08_200.pdf
http://www.r-type.org/pdfs/829b.pdf
http://www.r-type.org/pdfs/4d22.pdf
http://www.rell.com/filebase/en/src/Datasheets/7378_QE08_200.pdf
For 829B I see on page 10 a graph with knee at about 160 mA at 200 V on g2, 0 V on g1. Using 3/2 power law: [160]*(150/200)^1.5 => 145 mA at 150 V on g2.
The high g2/g1 Mu of 9 is a pretty good clue that the screen grid is not very sensitive.
For 4D22 I see on page 8: 600 mA at 300 V Ecg (assuming that's Vg2) and 0V on g1.
So using 3/2 power law: [600]*(150/300)^1.5 => 476 mA at 150 V on g2.
For QE08-200 I see on bottom of page 8: 720mA at 200V g2 and 0V on g1. Again using 3/2 power law:
[720]*(150/200)^1.5 => 654 mA at 150 V on g2.
The high g2/g1 Mu of 9 is a pretty good clue that the screen grid is not very sensitive.
For 4D22 I see on page 8: 600 mA at 300 V Ecg (assuming that's Vg2) and 0V on g1.
So using 3/2 power law: [600]*(150/300)^1.5 => 476 mA at 150 V on g2.
For QE08-200 I see on bottom of page 8: 720mA at 200V g2 and 0V on g1. Again using 3/2 power law:
[720]*(150/200)^1.5 => 654 mA at 150 V on g2.
Last edited:
If you have the g2/g1 Mu factor, you can also compute gm2. Take gm1 (at the datasheet given current Io) and divide by Mu to get gm2. Then gm2 can be translated to a new selected current by gm2*(Inew/Io)^.33333 That new gm2 will help determine how much drive voltage is needed for an operating point.
Keep in mind that Mu typically varies with current also. But if you use it at the current where it is known, then the 0.33333 power rule for g2 gm is fairly accurate over the full range of currents. For sweep tubes, where gm1 is fairly maximized, then gm1 will vary as approx. 0.5 to 0.6666 power of current. That in turn makes Mu (g2/g1) vary around 0.3333 power of current (0.6666 -0.3333).
It is harder to compare different tubes using gm2, since they all have different DC max current ratings. One could pick a standard current I suppose, or some fraction of maximum DC current rating would be better, since bigger tubes will actually be operating at higher currents.
Keep in mind that Mu typically varies with current also. But if you use it at the current where it is known, then the 0.33333 power rule for g2 gm is fairly accurate over the full range of currents. For sweep tubes, where gm1 is fairly maximized, then gm1 will vary as approx. 0.5 to 0.6666 power of current. That in turn makes Mu (g2/g1) vary around 0.3333 power of current (0.6666 -0.3333).
It is harder to compare different tubes using gm2, since they all have different DC max current ratings. One could pick a standard current I suppose, or some fraction of maximum DC current rating would be better, since bigger tubes will actually be operating at higher currents.
Last edited:
guitar amps
With regard to regulation, I have found that beyond ANY doubt, regulating the preamp section is very desirable. The sound is much cleaner, the bass much firmer, and the highs much more "shimmery." It is worth the couple of bucks in parts. I use high voltage NPN transistors, you can also use MOSFETs. I use Zeners to establish the reference voltage and that is the signal to the base of the transistor. The collector is connected to B+, and the output is the emitter. The only problem I ever encountered was when i used a BIG transistor to regulate the B+ for the power tubes. The transistor drew enough current through the base that I had to make a driver for it. The MOSFET is superior in this regard as gate current is about zilch.
I did a screen drive with TV sweet tubes, the 6DQ6B, AKA 6GW6, and it sounded really nice. They seemed, to my very untrained eye, to be rather similar to the 6L6GC, so I used transformers from a 6L6GC (maybe it was GB). Anyway, sounds great. I wasn't shy about plate voltage, I think I had it around 500. I did stay in spec with the screens and, just to be safe, I did put a resistor on the screen. I think I slowly backed off the screen resistance. It didn't seem to change the sound much, if at all, I think I finally settled on 470 or 1K.
The guitar guys don't want the power tubes B+ regulated. They like the sound of the power tubes being overdriven into distortion. They like the sound you get from using tube rectifiers, choke ONLY for the preamp section, and not too much capacitance (tube rectifier limits that anyway). That way, the power supply "sags" when the tubes are overdriven, so they get their distortion easier that way.Well that's good news. I'd rather not abuse tubes if there's no point to it (other than the primal satisfaction of watching things melt, of course
Aww, but where's the fun in that?
With regard to regulation, I have found that beyond ANY doubt, regulating the preamp section is very desirable. The sound is much cleaner, the bass much firmer, and the highs much more "shimmery." It is worth the couple of bucks in parts. I use high voltage NPN transistors, you can also use MOSFETs. I use Zeners to establish the reference voltage and that is the signal to the base of the transistor. The collector is connected to B+, and the output is the emitter. The only problem I ever encountered was when i used a BIG transistor to regulate the B+ for the power tubes. The transistor drew enough current through the base that I had to make a driver for it. The MOSFET is superior in this regard as gate current is about zilch.
I did a screen drive with TV sweet tubes, the 6DQ6B, AKA 6GW6, and it sounded really nice. They seemed, to my very untrained eye, to be rather similar to the 6L6GC, so I used transformers from a 6L6GC (maybe it was GB). Anyway, sounds great. I wasn't shy about plate voltage, I think I had it around 500. I did stay in spec with the screens and, just to be safe, I did put a resistor on the screen. I think I slowly backed off the screen resistance. It didn't seem to change the sound much, if at all, I think I finally settled on 470 or 1K.
Do you have any experience running the 6HB6 in triode or ul push pull?
Thanks,
Scott
No, sorry. Just some traces on the curve tracer:
http://www.diyaudio.com/forums/tubes-valves/160240-suppresor-grid-used-feedback-5.html#post2072733
http://frank.pocnet.net/sheets/135/6/6HB6.pdf
http://www.diyaudio.com/forums/tubes-valves/160240-suppresor-grid-used-feedback-5.html#post2072733
http://frank.pocnet.net/sheets/135/6/6HB6.pdf
Last edited:
Guitar Amps - CONT'D
I note on the datasheets that both 6CM5 and 6DQ6 (The common sweep tubes in Oz) have unusually high Max Rg1 values of 1MOhm.
The guitar amp examples I've seen with these tubes run fixed bias with Rg1 values of 470K. That is a light load for the phase splitter.
Common Guitar Amp Tubes such as 6L6, 6V6 and even EL34, in fixed bias, would self destruct with those sorts of Rg1 values.
Is this general for this class of tubes or just the two examples with which I'm familiar?
Cheers,
Ian
I note on the datasheets that both 6CM5 and 6DQ6 (The common sweep tubes in Oz) have unusually high Max Rg1 values of 1MOhm.
The guitar amp examples I've seen with these tubes run fixed bias with Rg1 values of 470K. That is a light load for the phase splitter.
Common Guitar Amp Tubes such as 6L6, 6V6 and even EL34, in fixed bias, would self destruct with those sorts of Rg1 values.
Is this general for this class of tubes or just the two examples with which I'm familiar?
Cheers,
Ian
The TV sweeps typically are spec'd with 0.47 to 1 Meg Ohm max grid resistance. Some are spec'd even higher. But these are for TV switch mode use, where some drift doesn't make a whole lot of difference to a fully ON or OFF state. Analog useage likely will require lower resistance if you don't want it drifting around. Some are also spec'd for grid leak bias and some are spec'd for regulated feedback control, so I would always check the datasheet.
