I finally got my 6DQ6 push pull guitar amp fully restored and operating.
It runs a pair at B+ of +430V with +220V on the screens, -40V bias for 60 Watts Out (Raa of 2K7 for the Output tranny).
I had the local guitar gods (well 2 of them anyway) give it a workout on the weekend just gone. Both loved it and commented particularly on power amp compression which is superior in quality and quantity to anything they had used previously.
Is this a function of the low screen voltage or is it characteristic of sweep tubes? I know that you use low G2 voltages on preamp tubes like EF86 to get compression. I want to reproduce this behaviour in my next git. amp and would like to know if I can get that by just running low screen voltages or if I need to stick to 6DQ6, 6CM5 or similar.
Thanks (in anticipation)
Ian
It runs a pair at B+ of +430V with +220V on the screens, -40V bias for 60 Watts Out (Raa of 2K7 for the Output tranny).
I had the local guitar gods (well 2 of them anyway) give it a workout on the weekend just gone. Both loved it and commented particularly on power amp compression which is superior in quality and quantity to anything they had used previously.
Is this a function of the low screen voltage or is it characteristic of sweep tubes? I know that you use low G2 voltages on preamp tubes like EF86 to get compression. I want to reproduce this behaviour in my next git. amp and would like to know if I can get that by just running low screen voltages or if I need to stick to 6DQ6, 6CM5 or similar.
Thanks (in anticipation)
Ian
Junm My 36/40KD6 Sylvania and 18GB5 El menco (that is Japan made) arrived from ESRC. Any Ideas or suggestion for PP amp or screen driven amp that suits these TV tubes? I have 5K/8 Opt around 30w or more to use....what B+ is needed and bias current? [/QUOTE said:It seems there are some interests for amplifiers made with 'sweep' tubes. I have made a real workhorse for 30years ago. It happend because via my job as a TV & Radio repairman felle in love with PL519A and the exspired EL503. I thought that one day I would like to hear this big power tube play music and I have never regret that.
I have no schématic (only in my head) but it goes as simple like this...
Many different tubes can be used here except for the output that has to be PL519/509 or similar. The driver and input is ECC85 and I asume that most peoble know how to find a propriate plate resistor to any tube. I ALWAYS seperate the power supply for the small tubes from the big ones. I want a stable supply for the smal tubes. I even use a stabelized for the input.
The output tubes run on 300V (3000uF) and the bias is ca -83V (100mA)...that means that the driver should be able to deliver least 200Vpp. To do this it need 430V supply because it loses 90V over the cathode resistor. Its conected to the input without coupling capacitor because I found that gives a more stable amplifier at full throttle. The output transformer must have an load impedance between 1600 to 1900ohm and be able to handle 100W at 30 HZ
I make the 430V by a voltage doubler from a 160V winding which also supply the output tubes filament; 40V X 4 = 160V...but the new amplifier I am building now are using EL509JJ. Thast means the new power transformer has a 6,3V 10A winding.
I also want to try another driver ECC99....
Why I use Ultra Linear coupling
Because it gives an output impedance on 0,8ohm and a damping factor at 10...before feedback is applied. When..feedback which is only 12db is applyed...then the damping factor increase to 40 which I am very satisfied with. When/if tetrode coupling are used it would be impossibly to get this prober output impedance...not even with 25db feed back.
I am aware that PL509/519 are scarce by now...but EL509-JJ is a VERY good replacement and made for audio. I call my amplifier for; *THE FINAL AMP* because I have never wanted any other amplifier. My friends who also have one claim the same..
Oh yes...it delivers 85W at 0,3% distortion and can run ANY loudspeaker...no need for 4, 8, 16 ohm tap here...thats one of the reasons I created this amplifier ... no speaker is only 4 or 8ohm or anything else. Therefore, an amplifier should do it all unchallenged. This leads to the sound...
The sound
This amp's ability to produce energy is very strong. Drum rim shot causes you to blink ... chorus makes you cry ... it come out as if you were present when the music was created. This makes this amp special ...
If anyone plan to built this I reccoment hard wiring as these amplifiers tends to last loooong...curcuit boards don't.
Last edited:
Oh, the good old PL500, or, at least, an equivalent of it!
With two of those tubes, running at a plate voltage of 400 volts and 190 volts for the screens, biased at 30 mA, and your PP output xformr you can expect some good 50 watts of output power.
Your other tubes also appear to work with this arrangement, but are somewhat overdressed. Unfortunately I cannot find any plate curvesfor them @ 200 volts screen voltage.
Best regards!
Originally Posted by Junm
My 36/40KD6 Sylvania and 18GB5 El menco (that is Japan made) arrived from ESRC. Any Ideas or suggestion for PP amp or screen driven amp that suits these TV tubes?
I have 5K/8 Opt around 30w or more to use....what B+ is needed and bias current?
-------------------------------
You may want a bigger OT for those 36/40KD6s (and lower Zpri, like 2.5K to 3K). More like 200 Watts output likely by George (Tubelab) setup/ratings. 120 Watts out, conservative. Should be good for screen drive too, if that's what you want to try. Not a good idea to use them for UL, since the screen voltage likes to be lower (200V max g2, probably want to run them around 150V on g2).
Local Schade (plate) type feedback to lower output Z.
My 36/40KD6 Sylvania and 18GB5 El menco (that is Japan made) arrived from ESRC. Any Ideas or suggestion for PP amp or screen driven amp that suits these TV tubes?
I have 5K/8 Opt around 30w or more to use....what B+ is needed and bias current?
-------------------------------
You may want a bigger OT for those 36/40KD6s (and lower Zpri, like 2.5K to 3K). More like 200 Watts output likely by George (Tubelab) setup/ratings. 120 Watts out, conservative. Should be good for screen drive too, if that's what you want to try. Not a good idea to use them for UL, since the screen voltage likes to be lower (200V max g2, probably want to run them around 150V on g2).
Local Schade (plate) type feedback to lower output Z.
Last edited:
Mine works perfect with 300V on both screen and plate...never had any problem vith that.
The low B + provides a great opportunity to run the amplifier ultra linear ... so it should of course be done that way. It is very gentle at the tubes that easily holds 10-15 years
I know, of course, a famous designer who also use these tubes. His approach must have been considering the tube as a KT88 for his B + is 600V and it kills the tube. It is designed to work well under 300V as it does in the CTV sets. Don't mind the reverse voltage at 7KV...it has nothing to do with it.
The low B + provides a great opportunity to run the amplifier ultra linear ... so it should of course be done that way. It is very gentle at the tubes that easily holds 10-15 years
I know, of course, a famous designer who also use these tubes. His approach must have been considering the tube as a KT88 for his B + is 600V and it kills the tube. It is designed to work well under 300V as it does in the CTV sets. Don't mind the reverse voltage at 7KV...it has nothing to do with it.
Last edited:
You may want a bigger OT for those 36/40KD6s (and lower Zpri, like 2.5K to 3K). More like 200 Watts output likely by George (Tubelab) setup/ratings. 120 Watts out, conservative. Should be good for screen drive too, if that's what you want to try. Not a good idea to use them for UL, since the screen voltage likes to be lower (200V max g2, probably want to run them around 150V on g2).
Local Schade (plate) type feedback to lower output Z.[/QUOTE]
To all who give ideas thanks...
Yes I really plan a higher power amp using 36/40KD6...Just waiting for the bigger OPT and Power Trafo.
I wanna play first the 18GB5 (LL500 equiv)....my opt maybe good around 30 -50 watts .pentode mode and screen drive PP amp...no plan for UL pp since my opt does not have ul tap...
How about triode mode at 250V B+...anybody has reference triode curve?
Local Schade (plate) type feedback to lower output Z.[/QUOTE]
To all who give ideas thanks...
Yes I really plan a higher power amp using 36/40KD6...Just waiting for the bigger OPT and Power Trafo.
I wanna play first the 18GB5 (LL500 equiv)....my opt maybe good around 30 -50 watts .pentode mode and screen drive PP amp...no plan for UL pp since my opt does not have ul tap...
How about triode mode at 250V B+...anybody has reference triode curve?
The B+ in a color TV set may be in the 300 to 400 volt range but the Horizontal (line) Output Tube runs on the "boosted B+" or often just "boost" supply. This is a bootstrapped supply using the positive voltage developed by the damper diode (tube). Connecting a voltmeter to the plate of the HOT is usually a quick way to destroy a meter. You can measure the tap on the flyback transformer that has an electrolytic cap to ground or B+. This is the boost supply that runs the HOT and a few other circuits. It can run 400 to 600 volts depending on the size of the TV and it's deflection angle. The negative peak during retrace goes into the KV region, but the positive ramp on the plate can go above the AVERAGE boost voltage too. You can see this on a scope with a LARGE voltage divider, but accurate voltage readings were not possible with the equipment I had at the time.
Running a sweep tube in a circuit designed for a KT88 kills the tube because of the high screen voltage. A sweep tube wants 150 to 200 volts max on the screen. Some will run at a higher voltage in triode or UL, but beware a random runaway may happen as the tubes age. I found that 6LW6's will work and sound nice at 400 volts in triode, but a year down the road a tube will red plate and fry without warning.
I have run sweep tubes at 650 volts on the plate with no failures or shortened life in audio applications. I, and at least a dozen other builders assembled 125 WPC versions of Pete Millett's Engineers Amp running 600 to 650 volts of B+. All are still alive and running after 3 years. I have a design for a 13GB5 amp that uses combined G1 and G2 drive for over 100 WPC. It ran at 125 watts for 6 hours without issue during initial testing. The key to avoiding early tube death in a sweep tube is to keep within the peak cathode current rating, the max dissipation rating and stay well below the max screen grid rating. To get serious power within these constraints requires a fairly high B+. The minimum idle current to keep distortion low at milliwatt power levels imposes a maximum B+ value due to idle dissipation, which should be 1/2 to 2/3 of max rating.
Please don't tell that to my tubes...then they might blew up for first time in 30 year's. They are still doing fine in my amplifier and I suspect they will continue many years in the future.
That I have made 24 amplifiers that works on my way should convince everyone that it's okay to run the tubes in this way. If not, they would probably break in a short time, but they keep that said 10-15 years without problems. I would say they have a good life in my amplifier. If this is not convincing, I do not know what to do.
That I have made 24 amplifiers that works on my way should convince everyone that it's okay to run the tubes in this way. If not, they would probably break in a short time, but they keep that said 10-15 years without problems. I would say they have a good life in my amplifier. If this is not convincing, I do not know what to do.
Last edited:
Running a sweep tube in a circuit designed for a KT88 kills the tube because of the high screen voltage. A sweep tube wants 150 to 200 volts max on the screen. Some will run at a higher voltage in triode or UL, but beware a random runaway may happen as the tubes age. I found that 6LW6's will work and sound nice at 400 volts in triode, but a year down the road a tube will red plate and fry without warning.
I wasent thinking of running sreens that high...I know they can't. I was thinking of ear509...I have repared a lot of these with burnd curcuit board around the output tubes. They ar running the screens at propriate voltage and still fry the tubes.
I wasent thinking of running sreens that high...I know they can't. I was thinking of ear509...I have repared a lot of these with burnd curcuit board around the output tubes. They ar running the screens at propriate voltage and still fry the tubes.
6KG6/PL519 is rated 250 V max for g2, while 6KD6 is rated 200 V max on g2. One can run the 6KG6 plate around 120 V below the screen grid voltage before the screen current really picks up. (see curves linked)
So using 43% UL and 300 V B+, with the plate dropping to 180 V, the screen will have dropped to 249 V. Just squeaking by the 250 V max g2 rating where it starts to count (plate current going up).
With the plate dropping to 90 V, the screen will be at 209 V, just hitting the 120 V difference (Vg2-Vp) safety margin at the 90 Watts (peak) output level (before xfmr losses). (45 Watts average sine wave, but real audio is peaky)
(On the positive plate swing, the plate V is well above the g2, so the plate is soaking up most of the current, and the plate current is low anyway.)
So Tube mania's design is running within the safety area, avoiding heavy screen current. But just inside. An 85 Watt sine wave might smoke the tubes.
I would be careful with the 6KD6 or other lowish 200 Vg2 max sweeps in UL.
http://frank.pocnet.net/sheets/084/6/6KG6.pdf
http://frank.pocnet.net/sheets/123/6/6KD6.pdf
So using 43% UL and 300 V B+, with the plate dropping to 180 V, the screen will have dropped to 249 V. Just squeaking by the 250 V max g2 rating where it starts to count (plate current going up).
With the plate dropping to 90 V, the screen will be at 209 V, just hitting the 120 V difference (Vg2-Vp) safety margin at the 90 Watts (peak) output level (before xfmr losses). (45 Watts average sine wave, but real audio is peaky)
(On the positive plate swing, the plate V is well above the g2, so the plate is soaking up most of the current, and the plate current is low anyway.)
So Tube mania's design is running within the safety area, avoiding heavy screen current. But just inside. An 85 Watt sine wave might smoke the tubes.
I would be careful with the 6KD6 or other lowish 200 Vg2 max sweeps in UL.
http://frank.pocnet.net/sheets/084/6/6KG6.pdf
http://frank.pocnet.net/sheets/123/6/6KD6.pdf
Last edited:
I am not saying that your method doesn't work. It obviously does. Some sweep tubes can handle 300 volts in triode or UL and some can't. Sometimes one brand of a particularly type number can eat a lot more screen voltage than another brand of the same type number. The 6AV5GA and 6CD6GA are examples of this. I tested several brands of these in triode and found some stable at 350 volts while others showed a tendency to run away at 275 volts.
Many times this is due to the residual gas levels inside the tube, other times it's due to a different internal construction.
An 85 Watt sine wave might smoke the tubes.
No...I have done all things to this amp we normally aviod. Open output with signal overload, to low impedance (3ohm), sine wave (8ohm 85W) for an houer.etc- Never seen red plate or gloving screens as I often se on EL34. I use to say: to kill this amp you must drop it from a high place...it cannot be done the normal way.
Output tubes are Siemens PL519A.
Running 85W o,3% distortion
Noise flor -90db
No...I have done all things to this amp we normally aviod. Open output with signal overload, to low impedance (3ohm), sine wave (8ohm 85W) for an houer.etc- Never seen red plate or gloving screens as I often se on EL34. I use to say: to kill this amp you must drop it from a high place...it cannot be done the normal way.
Output tubes are Siemens PL519A.
Running 85W o,3% distortion
An externally hosted image should be here but it was not working when we last tested it.
Noise flor -90db
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Hi,
*imho* there ain't no need for screen drive in PP circuitries. This was a suggestion, maybe from Bob Danielak, in order to circumvent the crude, if not to say non-existing, linearity ot sweep tubes, which are designed for switching purposes. In a PP amplifier there's no relevance from that.
Best regards!
re: Tube mania:
"No...I have done all things to this amp we normally avoid. Open output with signal overload, too low impedance (3ohm), sine wave (8ohm 85W) for an hour.etc- Never seen red plate or gloving screens as I often see on EL34. "
OK, then you are safe as you said, at least with Siemens PL519A. You'll want to check again with the JJ tubes. The high screen current only occurs when the plate V drops below 90 V. This is a pulse like condition at peaks, even with the sine wave drive. The 6KG6/PL519 has a 7 Watt g2 rating that handles it then. The 6KD6 is a little less with a 5 Watt g2 rating, so one would want to check setup conditions more carefully.
--------------------------
re: Kay Pirinha:
"*imho* there ain't no need for screen drive in PP circuitries. This was a suggestion, maybe from Bob Danielak, in order to circumvent the crude, if not to say non-existing, linearity of sweep tubes, which are designed for switching purposes. In a PP amplifier there's no relevance from that."
So are you saying screen drive was an idea for SE amplifiers?
The P-P amps will cancel the even harmonics, but may still have high odd harmonics with sweep tubes. The sweep tubes do have maximized/optimized g1 gm, so tend to have more inherent distortion than low gm audio tubes. On the other hand, the large flat cathode of TV sweeps seems to make for a larger consistant gain area than the smaller curved audio cathodes, so some compensation possible there. The screen "kink" in the sweeps is more pronouced too, but is mostly out of the way of the loadline due to the lower g2 voltage. Going to screen drive will certainly linearize the gain, or more safely, use the recent Mu scaled g2/g1 drive scheme.
In any case (other than the screen kinks), these defects are gradual in nature, so easy to fix with a little local or global feedback. One just needs a good driver with some local feedback to it.
The more worrisome defect is crossover distortion for P-P amps. Its got high harmonic content. Either have to up the idle current (or class A, yuk!) or use some crossover correction scheme. I rather like the gm modulation scheme J Broskie mentioned (adapted for tube of course; I didn't care for his overly complex adapted version though. All you have to do is put some current sense resistors in the plate circuits, then take the local plate feedbacks back to a differential driver stage. Bias for class aB. The differential current FDBK component removes the gm doubling bump if adjusted correctly.):
"No...I have done all things to this amp we normally avoid. Open output with signal overload, too low impedance (3ohm), sine wave (8ohm 85W) for an hour.etc- Never seen red plate or gloving screens as I often see on EL34. "
OK, then you are safe as you said, at least with Siemens PL519A. You'll want to check again with the JJ tubes. The high screen current only occurs when the plate V drops below 90 V. This is a pulse like condition at peaks, even with the sine wave drive. The 6KG6/PL519 has a 7 Watt g2 rating that handles it then. The 6KD6 is a little less with a 5 Watt g2 rating, so one would want to check setup conditions more carefully.
--------------------------
re: Kay Pirinha:
"*imho* there ain't no need for screen drive in PP circuitries. This was a suggestion, maybe from Bob Danielak, in order to circumvent the crude, if not to say non-existing, linearity of sweep tubes, which are designed for switching purposes. In a PP amplifier there's no relevance from that."
So are you saying screen drive was an idea for SE amplifiers?
The P-P amps will cancel the even harmonics, but may still have high odd harmonics with sweep tubes. The sweep tubes do have maximized/optimized g1 gm, so tend to have more inherent distortion than low gm audio tubes. On the other hand, the large flat cathode of TV sweeps seems to make for a larger consistant gain area than the smaller curved audio cathodes, so some compensation possible there. The screen "kink" in the sweeps is more pronouced too, but is mostly out of the way of the loadline due to the lower g2 voltage. Going to screen drive will certainly linearize the gain, or more safely, use the recent Mu scaled g2/g1 drive scheme.
In any case (other than the screen kinks), these defects are gradual in nature, so easy to fix with a little local or global feedback. One just needs a good driver with some local feedback to it.
The more worrisome defect is crossover distortion for P-P amps. Its got high harmonic content. Either have to up the idle current (or class A, yuk!) or use some crossover correction scheme. I rather like the gm modulation scheme J Broskie mentioned (adapted for tube of course; I didn't care for his overly complex adapted version though. All you have to do is put some current sense resistors in the plate circuits, then take the local plate feedbacks back to a differential driver stage. Bias for class aB. The differential current FDBK component removes the gm doubling bump if adjusted correctly.):
Attachments
Last edited:
Not in my experience. The design I did using 6BQ6's had much less distortion than a project using 807's. With 6BQ6's, the harmonics was almost all h3. With the 807's, there was considerable higher order harmonics. Twin-T test with PP 6BQ6's was a residual that was sinusoidal at 3X the signal frequency. With PP 807's, the residual was more like a sawtooth. Listening tests agreed: open loop, 807's sound quite nasty. 6BQ6's sound overly "aggressive", but without that pentode nastiness. In that regard, 6BQ6s are more like high powered 6V6s.
The 6BQ6 project included gNFB only, as lNFB wasn't necessary. About 6db gNFB was all it needed to take the "edge" off. 13db tended towards a SS sound.
X-over is always present unless you go Class A. The HD power amp tubes are rated quite conservatively, given the intended purpose: max RMS power at all times. Audio is much less demanding, and so you can bust the PD spec quite a bit. The 6BQ6 is rated for 12W, but I bias them hotter: 17.5W and that definitely improves the sonic performance while not compromising service life. I've been running the same pair of 6BQ6s since Ought-7.
I was not familiar with the PL519, so I looked it up and found that it is a 40KG6, rated for 1.5 AMPS of peak cathode current, 40 watts of dissipation, and 275 screen volts in sweep applications. It is similar to my old friend the 6LW6 which does indeed work well at 300 volts in triode P-P AB1. It does not last long at 400 volts SE in class A.
I have a bunch of guitar amp quality 6600 ohm OPT's that do not have UL taps. I have also used them as 3300 ohm to 0-2-4-8 ohm OPT's. To extend their frequency response and power handling capability I want to drive them from a low impedance source. I find that a circuit similar to Pete Millett's engineers amp with plate to plate feedback works best. This uses conventional G1 drive. A pair of 13GB5 / XL500 operated in this manner will make up to 80 watts of continuous sine wave power provided the load impedance, plate voltage, and idle current are optimized to stay away from cathode current and dissipation limits.
Somewhere in Pete's Engineers Amp thread we explored operating the 13GB5 at low voltages and low plate load impedances in triode mode. If interested it could be found, but it is a long thread.
Screen drive has a valid use, with its advantages, but some serious limitations too. A well designed G1 driven amp will operate at 55 to 65% efficiency over a narrow power range near clipping. The efficiency drops off as the power output is reduced. Idle current is a significant portion of the plate dissipation and can be a limiting factor at high plate voltages.
Screen driven sweep tubes can often operate at low idle currents with very little distortion even at low power outputs. The minimum idle current depends on the load impedance, tube choice, and G1 VS G2 voltages used to get this current. Some tubes are happy with 5 mA of idle current or less, most need 5to 15 mA. This allows a higher efficiency over a wider range of output powers. Higher efficiency allows higher output power.
You don't get something for nothing. Screen drive results in a higher output impedance, and a new failure mode that results in lots of blown parts. When the screen gets driven positive the plate gets drawn near zero. At this instant the screen draws all the current. Continuous full power operation can over dissipate the screen causing it to glow red....or white!
On the very next half cycle that glowing screen will be driven negative while the plate will be at twice the B+ voltage. That glowing screen will emit electrons that will travel to the plate. This unwanted current will increase, further overheating the screen, rapidly leading to a tube arc which will blow up the driver circuitry. This failure mode can be avoided by limiting screen current, but this limits power output and causes premature clipping.
I have a working prototype of an amp that drives G1 and G2. It will make 100 watts at 2% distortion, and 125 watts at clipping from a pair of 13GB5's. The idle current is 8 to 10 mA for an idle dissipation of 5 to 7 watts per tube at 650 volts. The tube's heater produces more heat! At 100 watts output the tube dissipation is 20 watts per tube, which is over spec but doesn't result in red plate and will only be seen in transients in normal music. In fact operation with normal music will see power outputs below 10 watts 95% of the time. This is where the low idle current shines, since at 10 watts out the tubes dissipate about 10 watts each.
I have a bunch of guitar amp quality 6600 ohm OPT's that do not have UL taps. I have also used them as 3300 ohm to 0-2-4-8 ohm OPT's. To extend their frequency response and power handling capability I want to drive them from a low impedance source. I find that a circuit similar to Pete Millett's engineers amp with plate to plate feedback works best. This uses conventional G1 drive. A pair of 13GB5 / XL500 operated in this manner will make up to 80 watts of continuous sine wave power provided the load impedance, plate voltage, and idle current are optimized to stay away from cathode current and dissipation limits.
Somewhere in Pete's Engineers Amp thread we explored operating the 13GB5 at low voltages and low plate load impedances in triode mode. If interested it could be found, but it is a long thread.
Screen drive has a valid use, with its advantages, but some serious limitations too. A well designed G1 driven amp will operate at 55 to 65% efficiency over a narrow power range near clipping. The efficiency drops off as the power output is reduced. Idle current is a significant portion of the plate dissipation and can be a limiting factor at high plate voltages.
Screen driven sweep tubes can often operate at low idle currents with very little distortion even at low power outputs. The minimum idle current depends on the load impedance, tube choice, and G1 VS G2 voltages used to get this current. Some tubes are happy with 5 mA of idle current or less, most need 5to 15 mA. This allows a higher efficiency over a wider range of output powers. Higher efficiency allows higher output power.
You don't get something for nothing. Screen drive results in a higher output impedance, and a new failure mode that results in lots of blown parts. When the screen gets driven positive the plate gets drawn near zero. At this instant the screen draws all the current. Continuous full power operation can over dissipate the screen causing it to glow red....or white!
On the very next half cycle that glowing screen will be driven negative while the plate will be at twice the B+ voltage. That glowing screen will emit electrons that will travel to the plate. This unwanted current will increase, further overheating the screen, rapidly leading to a tube arc which will blow up the driver circuitry. This failure mode can be avoided by limiting screen current, but this limits power output and causes premature clipping.
I have a working prototype of an amp that drives G1 and G2. It will make 100 watts at 2% distortion, and 125 watts at clipping from a pair of 13GB5's. The idle current is 8 to 10 mA for an idle dissipation of 5 to 7 watts per tube at 650 volts. The tube's heater produces more heat! At 100 watts output the tube dissipation is 20 watts per tube, which is over spec but doesn't result in red plate and will only be seen in transients in normal music. In fact operation with normal music will see power outputs below 10 watts 95% of the time. This is where the low idle current shines, since at 10 watts out the tubes dissipate about 10 watts each.
Ah-ha, the EL519/PL519 is rated even higher than the 6KG6/PL509. 275 Vg2 versus 250 Vg2. 45 Watt design max versus 40 Watt design max. 9 Watt for the screen versus 7 Watt.
So PL519 with 300 V B+ in UL is close to nominal rated from the get go. 45 Watt/ 9 Watt/ 275 Vg2
No wonder it runs well under UL conditions.
While 6KD6 is 33 Watt /5 Watt/ 200 Vg2
Will be trouble in UL.
http://frank.pocnet.net/sheets/010/p/PL519.pdf
http://frank.pocnet.net/sheets/010/p/PL509.pdf
http://frank.pocnet.net/sheets/084/6/6KG6.pdf
http://frank.pocnet.net/sheets/123/6/6KD6.pdf
So PL519 with 300 V B+ in UL is close to nominal rated from the get go. 45 Watt/ 9 Watt/ 275 Vg2
No wonder it runs well under UL conditions.
While 6KD6 is 33 Watt /5 Watt/ 200 Vg2
Will be trouble in UL.
http://frank.pocnet.net/sheets/010/p/PL519.pdf
http://frank.pocnet.net/sheets/010/p/PL509.pdf
http://frank.pocnet.net/sheets/084/6/6KG6.pdf
http://frank.pocnet.net/sheets/123/6/6KD6.pdf
Another interesting observation.
For the bigger sweep tubes, the plate area does not keep up with the Watts rating versus the smaller sweeps. Instead, the max. rated operating temp. goes up.
For example, a 21HB5 18 Watt tube has a 34 mm long plate, and 220 deg. C envelope rating.
While a 6LF6, 6KG6, 6LW6 40 Watt tube has a 47 to 53 mm long plate, and 300 deg. C rating (well, 240 deg. for 6LW6, but it's got extra cooling fins)
That tells me that one is better off using two 21HB5 or similar 18 Watt 220 deg. C tubes (which were $1 each) in parallel, rather than using one expensive big sweep tube. (21HB5 each rated 230 mA DC max and 800 mA peak, ie total: 460 mA max DC and 1600 mA peak versus the 6LW6 rated 400 mA max DC and 1400 mA peak) (2 x gm a similar story) (total filament current higher though for paralleled tubes)
And also, the Watt rating of an 18 Watt 220 deg. C tube should be much more upgradeable using a fan cooling setup. Put a fan on a 21HB5 and it should be able to do 24 to 27 Watts region. (that's before doing any further sweep duty to audio rating upgrade, another 1.3X typically => 31 to 35 W) Although the current rating would not increase this way. But put a fan on two parallel 21HB5 tubes and you end up with a 62 to 70 Watt audio tube.
For the bigger sweep tubes, the plate area does not keep up with the Watts rating versus the smaller sweeps. Instead, the max. rated operating temp. goes up.
For example, a 21HB5 18 Watt tube has a 34 mm long plate, and 220 deg. C envelope rating.
While a 6LF6, 6KG6, 6LW6 40 Watt tube has a 47 to 53 mm long plate, and 300 deg. C rating (well, 240 deg. for 6LW6, but it's got extra cooling fins)
That tells me that one is better off using two 21HB5 or similar 18 Watt 220 deg. C tubes (which were $1 each) in parallel, rather than using one expensive big sweep tube. (21HB5 each rated 230 mA DC max and 800 mA peak, ie total: 460 mA max DC and 1600 mA peak versus the 6LW6 rated 400 mA max DC and 1400 mA peak) (2 x gm a similar story) (total filament current higher though for paralleled tubes)
And also, the Watt rating of an 18 Watt 220 deg. C tube should be much more upgradeable using a fan cooling setup. Put a fan on a 21HB5 and it should be able to do 24 to 27 Watts region. (that's before doing any further sweep duty to audio rating upgrade, another 1.3X typically => 31 to 35 W) Although the current rating would not increase this way. But put a fan on two parallel 21HB5 tubes and you end up with a 62 to 70 Watt audio tube.
Last edited:
I tried some EL 34 in a Chinese Sepparo KT88SE amp. The anodes glowed. They have UL and from what is said here that might be the reason? Perhaps the UL that is not 43%. An SE amp . The cathode resistor is 470 R which usually is OK for a short test ( I would guess 33 watt dissipation if so , this seemed far higher ) . HT is 450 V.
Not a bad amp that is nicely made. We replaced the ECC83 with Brimar and NOS GE 12BH74a to replace ECC82 SRPP stage. The KT88 Black treasure bought at a sensible price. The original 88's not too bad.
I am not a great fan of the 88 although we were born almost to the day together. Were it not for the valve base Gu50 would be an idea.
Rapid electronic UK have part number 48-0900 is in the clearance sale at about $1. It is an octal valve/relay base rated at 250VAC UL. Doubtless OK at 500 VDC. They have nice screw terminals. Ideal for hook up. If they flash over that's no different to some valve bases. Worth a try for quick tests of ideas? Just tried a 6L6GC , good fit. Ideal to try an RH 34 etc quickly.
Not a bad amp that is nicely made. We replaced the ECC83 with Brimar and NOS GE 12BH74a to replace ECC82 SRPP stage. The KT88 Black treasure bought at a sensible price. The original 88's not too bad.
I am not a great fan of the 88 although we were born almost to the day together. Were it not for the valve base Gu50 would be an idea.
Rapid electronic UK have part number 48-0900 is in the clearance sale at about $1. It is an octal valve/relay base rated at 250VAC UL. Doubtless OK at 500 VDC. They have nice screw terminals. Ideal for hook up. If they flash over that's no different to some valve bases. Worth a try for quick tests of ideas? Just tried a 6L6GC , good fit. Ideal to try an RH 34 etc quickly.