• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Maximum Plate/Screen Voltage 6550 in UL

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have two of John Atwood's One Electron BFT-1B PTs that I want to use in a pair of 6550 monoblocs. This PT has two HT windings, one is 770V CT, the other is 520V CT. The 770V tap translates to about 530V plate quiescent.

The 6550 datasheet seems to limit the applied voltage to 450V. Can I cheat on this up to 530V, and if so then I guess the OT has to be able to take it too. I'm looking at a 5K Edcor OT.

If not, then I'll probably use the 520V CT secondary winding as the screen supply, but this will degrade the performance some I think.

I've built several 50 to 100W guitar amps, lately mostly Dumbles, so I'm used to playing fast and loose with design maximum numbers. Do I need some re-education in this application?
 
KT88

Thank you Ruben. I know that the KT88 is generally considered to be tougher than the 6550, however, is a NOS Tung-Sol 6550 perhaps tougher than a current production KT88?

In a pinch I can rise to the cost of four NOS 6550, but certainly not to four NOS KT88, even if I could find them.
 
Thank you Ruben. I know that the KT88 is generally considered to be tougher than the 6550, however, is a NOS Tung-Sol 6550 perhaps tougher than a current production KT88?

In a pinch I can rise to the cost of four NOS 6550, but certainly not to four NOS KT88, even if I could find them.

i have been running two pairs of kt88-98's for about 6000 houres at 450 volt ultra linear with an output impedance of around 3kohm. i might have gotten an really good quad or they are strong.

but if the new production kt88's arent strong enough might aswell bump it up to an kt90.
 
...however, is a NOS Tung-Sol 6550 perhaps tougher than a current production KT88?
Quite possibily so. True NOS 6550s were reliabily manufactured. And they were used in tube regulated power supplies with higher plate voltages then that. I personally use GE Jan 6550As in ULPP with almost 600 volts on them with no problems. Yes, there may be an occasional NOS failure by fireworks, but these would probably have failed anyway even under lower voltage. I certainly would not feel comfortable using new production 6550/KT88s at this level. Can't speak for KT120s since I've never used them. I have a lot of NOS to use up first.
 
Thanx for your responses. I've been able to find some instances of +500V UL operation of KT88s so I'm going to proceed on that basis. Right now Golden Lion seems to be a well regarded choice in KT88s.

For 70W per channel I'm thinking 5K is about right for RLaa onthe OT. I'm also looking at using Patrick Turner's 50+50 power amp design as a basis, which uses 6K6 OTs. See pic attached.
 

Attachments

  • Integrated5050.pdf
    108.3 KB · Views: 101
I have used Tungsol Reissues 6550's at 560 volts as per the GEC circuit for multiple KT88's, they lasted for at least 10 years, but that is with 39K as recommended in the GEc circuit on the Grid using Fixed bias, not 100K. Current production tubes such as the KT120 recommend 50k not 100K. Don't try 560 volts with Psvane KT88's or Chinese tubes.
 
That's interesting. I saw somewhere that "Tung-Sol reissue 6550s don't like anything above 500V". Grid bias not specified. Shows you how misleading incomplete info can be!

Turner's design uses 68K grid bias at 520V and I'll be at 520V too, not 560.

I would think your comments on those current KT88s would be cause for concern even at 520V.
 
For 70W per channel I'm thinking 5K is about right for RLaa onthe OT.
I think 5K Rpp is slightly on the high side for 70 watts. Generally 4 to 4.3K is more typical. But it will also depend on power in. A general rule of thumb is higher Rpp gives slightly lower output along with lower distortion. Lower Rpp produces higher output and ligher distortion. All things being equal. Just keep in mind that every choice in output transformers is a tradeoff of some kind. But whether you actually get 65, 70 or 75 watts, you aren't going to "hear" the difference. Things also depend on how you drive the OPTs and whether you stay in class AB1 or reach AB2. For AB2 you'll need a good cathode (or FET) follower directly coupled driver.
 
Thanx again, I have seen at least one recommendation of 4K5. Fortunately I've not bought the OTs yet. I'm aware of the RLaa trade-off but have no personal experience with this particular application. I do have the PTs in hand so I don't want to stray from the 520V operating point.

I just looked at Turner's schem with 6K6 for 50W and eyeballed it, after looking at the 6550 plate curve for 520V, looks to be about 5K.

I have built one 100W AB2 guitar amp with a direct coupled cathode follower 12BZ7 driver (early Dumble Steel String Singer) which was one loud SOB with a quad of Mesa 415-STR 6L6GC at 530V, and is not much quieter with a different PT at 460V. So I kind of doubt I would need to go to a driver design for 70W with a single pair of 6550 or KT88. Is that a reasonable inference? AB2 does give great clean bass extension though.

These amps will be used for listening to dinosaur rock and blues, not a 92 piece orchestra with a pipe organ doing Beethoven's 9th, so I can tilt a bit towards the more power end of the RLaa spectrum.
 
I'm running Svetlana 6550=C= PP with 4.3K Raa 60 watt output iron on a B+ of 480V.
Switches between triode and UL mode. I use triode most of the time, if I just need it louder, party, go UL....
More volume than the 6.6K, but the 6.6k sounds sweeter, larger class A operating zone....
I've tried KT88's, just prefer the =C= tubes....
 
Have a look here.
http://www.diyaudio.com/forums/tubes-valves/186892-aronov-ls9100-any-info.html post #7
My Schematic Trace with measured voltages of the ARANOV 9100 which is 4 x 6550 in parallel push pull Utralinear. It runs B+ of 530V. When I did a repair of it I put Winged C Svetlana 6550C in it. Haven't heard back from the owner so must be still going fine.

The SV6550C data says Va max 600V and Vg2 max 400V but that is for normal pentode mode operation, the 400V Vg2 rating is to keep screen dissipation under control. In Ultralinear, as tube current goes up and anode voltage drops then because of the UL tap the screen voltage also drops and this helps keep screen dissipation under control. That is why you can run significantly higher B+ than the pentode Mode Screen max Vg2 rating.

The ARANOV above runs B+ at 530V with just 100 Ohm screen resistors. These resistors will be performing their "grid stop" function only with practically no extra screen protection (they would need to be 470 Ohms to 1K Ohms to give addition screen protection).

In summary: I would be quite happy with a B+ of 530V but suggest that you are toward the upper end of what is acceptable and would not want to go above say 550V. When I put that amp back together I biased the 6550C for 62mA idle each which is just under 33 Watts dissipation per tube, a reasonably conservative and hence RECOMMENDED bias point (I would not go above 65mA which is nearly 35 Watts per tube).

Cheers,
Ian
 
Last edited:
Ginger, thanx for that detail. Using the Turner schematic I am closer to 50mA cathode draw and at 220R on the screens. I'm more used to 470 on the screens myself.

I found a used pair of real old 1965 Tung-Sol 6550s in one of my back boxes so I'm going to try them first. If they are too far gone, which I doubt as my tester says they're good at 400V, I'll go for the current Golden Lion KT88s and change the bias to suit.

Chris, thanx for the thumbs up on the GLs.
 
Thanx tinnitus, there was an interview in Tonequest magazine with the owner of Tube Amp Doctor last year, and Tonequest ran some tests too, and the TAD tubes all got good reviews except the EL34.

I succumbed to the allure of NOS and got another pair of '65 Tung-Sols of Ebay, so I now have my quad!
 
Thanx Phil, I'm using Patrick Turner's 5050 Integrated 6550 circuit as a base, changing it to a pair of monoblocks but keeping 68K grid resistors on the 6550s. This shouldn't overstress them at about 525V I think.

I have previously put 150K grids at +500V on used Sylvania 6550s that were about 40 years old, and they didn't bark at me. How long they might last is another matter, as I disassembled that amp and did something else with it.

Interestingly Mr. Turner originally had 220K grids in his integrated 5050, but changed them to 68K "to give better DC stability since less Vdc is generated across the 68K as tubes age".
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.