In an effort to avoid any accidents.
There's a bucket load of info on keeping 6550s alive, mostly around keeping the glass under 250C. An IR thermometer is suggested by some (and now cheap to hire as every electrician should own one).
The problem with this amp is threefold : Fender pushed the 6550s well past their official ratings back when US mains power was 110VAC. To the point that they're "specially designed and selected": that is, no longer standard 6550s (apparently GE 6550A is the factory part). I'm guessing they're idling at somewhere near 40W per tube, waay outside 6550 specs. (should be 600V/300V/50mA@-40VG1 not 700V/340V/???@-42.5 w/ ???>50mA! )
Next, you've got one, single pot to set the bias on all four tubes. If they're not all matched (or one screen resistor catches fire) you're screwed.
And now the mains in the US is often nearer 120VAC (by all reports). So your dissipation goes up again.
Did I mention carbon tracking yet? Your not planning on using this in a hot, sweaty bar by any chance?
No wonder there's a mythical 30 step plan for something that should take two minutes (once you've worked out whether the distortion switch causes more or less bias current).
Best thing to do with this amp is to reduce the B+ (and apply the immortal amp mods) or find some 50W "6550"s in a matched quad.
(ten minutes with my favourite search engine uncovered the factory test and set up procedures for the 400W version. Neither of which has 30 steps. But strangely on the source pages for the blown 6550 photo you used.)
Last edited:
Ran out of edit time
Over on talk bass they've discussed this amp, including with KT-88s and KT-90s in them
Over on talk bass they've discussed this amp, including with KT-88s and KT-90s in them
the amp's owner is resisting all suggestions for mod's to his amp and mention of reducing B+ almost made him have a fit.(seems to be after every ounce of output he can get)
as i've seen 600 and 700 volts listed as max plate voltage for KT88's not sure if the difference is how they are tested and spec'ed or manufactured?
i'm still unsure about what tubes to use due to operating parameters of the design and don't want to see a premature failure of some rather expensive components.
as i've seen 600 and 700 volts listed as max plate voltage for KT88's not sure if the difference is how they are tested and spec'ed or manufactured?
i'm still unsure about what tubes to use due to operating parameters of the design and don't want to see a premature failure of some rather expensive components.
I can understand that - if it's in good nick it's in danger of being "collectable".
Yup. Especially as there's many weird components in that amp that are pure unobtainium.
Perhaps the path of least resistance might be some time looking at KT-90 and KT-120 curves, given that genuine GE 6550A seem to be more expensive than NOS RCA 300Bs. (Especially in matched quads.)
A further challenge is going to be slipping in protective components (such as inline fuses) without buggering the vintage value. Any modifications need to be reversible.
Good thoughts - there's buckets of info on talkbass on setting up SVTs. And finding suitable tubes.
i'll have to check out "talkbass"
bb5000 if i bias it that level it wouldn't be making the 300 watts it supposed to, or is the 6v6 driving the x-former phase splitter making up the difference? sorry for the question this is a little unfamiliar territory for me as i've not worked on an amp like this one before so i'm perhaps overly cautious with everything.
bb5000 if i bias it that level it wouldn't be making the 300 watts it supposed to, or is the 6v6 driving the x-former phase splitter making up the difference? sorry for the question this is a little unfamiliar territory for me as i've not worked on an amp like this one before so i'm perhaps overly cautious with everything.
The 6V6 SE stage and the interstage transformer is there as a low impedance PI solution (I've built a similar solution in the Weber AB200 amp). I guess that the idea is providing class AB2 capability. I think that Ted Weber designed the AB200 to be used like that, but tragically he passed away just as the design was released, so there's no one to confirm it.
I think that you will get more than the standard 100W/pair even in class AB1. I've built a Hiwatt DR201 clone which has an output of 200W with 6xEL34 (720Vp, 390Vg2) in class AB1, this is more than the standard 50W/pair that the EL34 normally is used for in Marshalls etc..
The Reflector factory (Sovtek, ElectroHarmonix etc) manufactures a 6550 tube that seems to be very similar to a KT88 :
https://www.newsensor.com/pdf/electro-harmonix/6550eh.pdf
https://www.newsensor.com/pdf/sovtek/6550we-sovtek.pdf
I've serviced Ampeg SVTs with EH 6550's, so that should probably be a suitable tube, unless you opt for KT88s.
I think that you will get more than the standard 100W/pair even in class AB1. I've built a Hiwatt DR201 clone which has an output of 200W with 6xEL34 (720Vp, 390Vg2) in class AB1, this is more than the standard 50W/pair that the EL34 normally is used for in Marshalls etc..
The Reflector factory (Sovtek, ElectroHarmonix etc) manufactures a 6550 tube that seems to be very similar to a KT88 :
https://www.newsensor.com/pdf/electro-harmonix/6550eh.pdf
https://www.newsensor.com/pdf/sovtek/6550we-sovtek.pdf
I've serviced Ampeg SVTs with EH 6550's, so that should probably be a suitable tube, unless you opt for KT88s.
Last edited:
In an amp that reaches class B, bias setting has nearly no effect on MAXimum power.
Your car engine had an Idle Screw (now an ECU function). Twiddling that has no effect when you press the pedal to OFF-idle and to Wide Open.
Bias matters for heat and the first Watt. A cold idle bias will sound hoarse on soft sounds. Tubes don't do this as bad as transistors. A 300 Watt stage amp is rarely judged by its ability to whisper nicely.
well i guess i'm mistaken i thinking that improper bias in a push pull output stage can at one extreme can cause crossover distortion and reduced output and at the other extreme excess clipping and exceed plate dissipation and subsequent tube failure.
is this not correct?
also in the case of the PS300 the current balance setting is to compensate for asymmetrical clipping or do i have that wrong also?
is this not correct?
also in the case of the PS300 the current balance setting is to compensate for asymmetrical clipping or do i have that wrong also?
Last edited:
Are you referring to the bias balance pot ? Set level, then balance to achieve even idle current /minimize hum.. I think that the key here in biasing is to allow maximum voltage swing without redplating (hot) or class B operation (cold).. using a scope and a dummy load (a big one) would allow you to see what's happening and measure output power etc..
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.