• 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.

KT88 PP UL Design Help Needed

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Chuck Hollander is a name that comes to mind on that . It's very complicated . I did tell Sid off about it . It is in place of a big choke . The ripple was I guess about 15 V . The 37 V dynamic range of the LM317 copes with that . Problem is it must not swing further . Transient can kill it . I speculated that if the transformer was split somehow it only need see the ripple and not need the protection . LM317 is fast enough for a class A if wondering . It is a bit like making LP's from CD as sometimes happens this idea . Slightly wrong yet it works .
 
For 100W output from a pair of KT88's the data sheet gives a -75V (Vg1). Should I use this voltage or half of this voltage on the plate I vs E plots? The individual lines shown do not include -75V. Do I need a different plot? -37.5V gives a current up above 200mA this seems really high. If I extrapolate a line at -75V I would get a current of 0±mA at 560Va which seems too low. From what I see B+=560, Eg1=-58.6 to give Ia=50mA???

JW
 
Listening #111 | Stereophile.com

Didn't find what you wanted will ask friends . It says here hum was very good . No idea why he did it . Sid was a musician . It must have sounded better .He used Quad ESL 63 's if wondering .

I wasn't in the slightest bit interested in valves as we call them at the time ( about 1999 I think ) . I met Sid through a friend and had virtually no idea who he was . He talked me into believing capacitors mattered as he had just discovered it himself . I knew before him about it . I said nothing . He had a child s fascination about things . He wasn't high flung you know . He said he always needed other engineers to help .
 
I can never get my head around them . I was taught by Sid Smith the designer of Marantz amps . He gave up . He agreed my way works . Very nice man was Sid . Sid used LM317 for HT smoothing . It was arranged to work at 400 V and was a famous modification . I always would use cathode bias as it adjusts the KT88 as it ages . Sid always triode connected his amps at home . Sowter in UK make transformers about as good as SIds .
I had Art Audio Concerto triode/UL kt88 with variable feedback and cathode bias. One of their early units made in GB-terrible workmanship ,excellent sound I regret I did not copy schematic and manual .It was the best sounding KT88 PP amp I heard to date.
 
For 100W output from a pair of KT88's the data sheet gives a -75V (Vg1). Should I use this voltage or half of this voltage on the plate I vs E plots? The individual lines shown do not include -75V. Do I need a different plot? -37.5V gives a current up above 200mA this seems really high. If I extrapolate a line at -75V I would get a current of 0±mA at 560Va which seems too low. From what I see B+=560, Eg1=-58.6 to give Ia=50mA???

JW

With todays KT88's, keep Iquies low; or they won't last long at 560V and the business gets costly.

Decades ago, KT88 UL circuits were about that showed Iquies 60mA at 600V+, a non-survivable grilling for todays tubes. That's the reason I went to parallel push pull.

richy
 
This forum is great, my learning has taken leaps since I posted my questions.

I have decided that yes, 560 V is too high and am looking now at 460-480V depending on what I use for a driver tube, if I can use the same B+ for both it reduces the complexity and the part count. It also means I can use the 21st century Maida regulator. Tom has already done the heavy lifting and is selling kits.

So now I need to get my driver up to snuff. At present I have a single 6SN7 drawn as a LTP which will take a 3.5±V up to 124V or about 35 times. It has been suggested to me that I will need more gain when I add the NFB of between 14-20 dB. So now I am considering a pair of 6SN7's wired as a cascode LTP. Can I calculate the gain by finding the gain for a cascode then divide by two for putting it in a LTP?
 
ECC 81 and EF184 are not often thought of to drive KT88 ( 6550 ) . They are possible options . EF184 in Long tail pair perhaps . I don't get involved in PP so out of my depth after that . Look at Quad 11 and see if it can be scaled up . Over to someone who would know on that .

I am looking to buy some KT 88 . I don't want to pay silly money and would use EL156 if I did . JJ is out I think .

http://www.hifitubes.nl/weblog/wp-content/telefunken-el156.pdf
 
Last edited:
New Quad II forty by TdP which I was checking after it was blowing tubes runs them at 390-400V 100mA in cathode bias circuit . You need Penta Labs Kt88's with ~50W dissipation rate to run in High Power Circuit. Don't forget to confront your design with well regarded circuits. Personally I'd try Allen Wright (may he rest in peace) PP-2 amp.
 
KT 88 tester

K6KHP.jpg


I think this would work OK as a tester ? In a wood or plastic enclosure as it is not isolated . Various valves could be tested and data kept .

For USA I guess a capacitor fed doubler would give the same ?

Put a high grade input signal transformer on it and you would have an amplifier . Safety will be OK then if careful . NE 5534 op amp or boosted OPA 604 would drive it . I have tried something like this in the past built for me by my brother . I felt a mains transformer offered advantages sonically .

The 6V is a switch mode PSU for simplicity ( 6.3 V ) . The -6V bias switch is to lower the KT 88 into a lower bias . That would I guess make an ideal standby switch if an ideal voltage was found ( run from a comparator perhaps with a 1 hour flip flop action ) ? Please improve this idea if you can .

Stontronics 6V 2A Desktop Power Supply Unit T2319ST
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.