Selenium Bridge Rectifier On Geloso Tube Amp

[ThermionicScott]

Oops i missed that too.
Yeap i have 19-20V on cathode. Great. 
And if redplating in future (don't think so as i play clean
mostly) i could raise the cathode resistor as Phoenix said.
Perfect.

Pedantic notes:  on a cathode-biased guitar amp, you're as likely to see red-plating at idle as any other time.  So whether you play clean or loud won't make a big difference. 

And the screens will account for about 10% of the cathode current at idle, so I'd reckon your idle plate dissipation at about 20.2 and 18.5 watts. 

[Rob Strand]

Pedantic notes:  on a cathode-biased guitar amp, you're as likely to see red-plating at idle as any other time.  So whether you play clean or loud won't make a big difference. 

And the screens will account for about 10% of the cathode current at idle, so I'd reckon your idle plate dissipation at about 20.2 and 18.5 watts. 

His measurements using the OT transformer resistance implied about 69mA plate current.   The disagreement between the plate and cathodes I was sweeping under the uncertainty floor.

At the end of the day the biasing is a probably little excessive.   I suspect if you lift the 1k  cathode resistor it will add a a bit more breathing space without any noticeable change in performance.  (Probably worth checking the power dissipation in the 160 ohm resistor, should be fine.)

[ThermionicScott]

Good catch, Rob.  I think I've measured the current across an OT winding once or twice out of curiosity, much prefer measuring it across a cathode resistor if present.   

[sarakisof]

Was taking some final measurements just for the record today, when i realised i get -22.1Vdc at preamps (I, II) heater voltage just after the heater choke. Shouldn't i supposed to get about -25Vdc there or isn't such a big deal?

[Phoenix]

Was taking some final measurements just for the record today, when i realised i get -22.1Vdc at preamps (I, II) heater voltage just after the heater choke. Shouldn't i supposed to get about -25Vdc there or isn't such a big deal?

Was that measurement when your wall voltage was nominal?

If so, yes, it's a little low. You haven't replaced the B60 selenium bridge rectifier, right? Its impedance has likely been increasing with age, a good reason to change it, though I'd always recommend changing any and all selenium rectifiers, as they're exceedingly nasty (very smelly and quite toxic) when they blow, and it's a matter of *when*, not if, they will blow.

Replace it with a silicon bridge, you may need to add some series resistance to drop some voltage. Between the bridge negative terminal and the 100uF reservoir cap is as good a place as any, or better, split the resistance in half and put the other half between the bridge positive terminal and ground (improved common-mode filtering).

If you'd already replaced it with silicon, the amp should work fine, that DC supply is only for the preamp tube heaters, which won't be starved for emission. Still a little low, but it's unlikely you'll suffer any negative consequences unless you use a tube with very poor emission anyway.

[sarakisof]

Ok, what i was thinking of, thanks once again Phoenix. 

[sarakisof]

Any inputs fender inspired mods for one channel guitar and the other for bass guys?

[Phoenix]

Any inputs fender inspired mods for one channel guitar and the other for bass guys?

Is there something wrong with the inputs now? If anything, maybe just throw a bright cap on one of the volume controls to give them a different flavour from each other.

But if you really want, you could convert the grid-leak-biased inputs to cathode bias, of the Bassman type.
Note that the Fender Bassman, while intended as a bass amp, makes a much better guitar than bass amp (at least that's the general consensus).

As far as different inputs to suit bass vs guitar, well, that really depends on the sound you're after. There's little to distinguish bass vs guitar amps, at least formally - the most important being extended bass response for the former. Much more important are the speakers/cabinet.
Early bass amps were often eq'd in exactly the same way as guitar amps, but evolved (in most cases) to a more "flat" response vs the classic mid-scooped eq of an FMV tonestack. Trying to cram both types of designs (flat/scooped) into one amp would really require two completely seperate channels, here you've really only got the two (instrument suited) inputs to play with. I say if you must, just go with a Bassman style input, forget about which is a "bass" or "guitar" input and just use what sounds best for what you're trying to do at the time.

Personally I vote to leave the grid-leak-biased inputs alone. Not everything needs to sound like a Fender, and there are already plenty of Fenders out there already. Why not let this be its own thing? The world is a richer place for it.

[antonis]

Any inputs fender inspired mods

How many 68k carbon resistors have you handy..??

[PRR]

The gridleak bias plan overloads too easily on modern (post-1957) pickups. Use Fender cathode bias on one of the two inputs. Since it is already built with 68k plate resistor, scale Rk down to 1k.

[amptramp]

3.3 megohms is a bit low to get good contact potential bias.  Radios that used the 6AV6 tube which has a triode with the same characteristics as a 12AX7 usually used 10 megohms but you would occasionally see 6.8 megohms.  The grid bias comes from electrons "boiled off" the cathode that impinge on the grid and the surplus of electrons drives the grid negative.  The preamp stage in my Trio KW-55 receiver has a 3.3 megohm grid resistor in one 12AX7 section and 1500 ohms in the cathode of the other.  The one with the 3.3 megohm grid resistor has a 68 K plate load and the plate runs at 85 volts from a 133 volt supply, so it almost matches your operating conditions.  It can be done.  The section with 1500 ohms on the cathode runs at 0.5 volt cathode bias with a 270 K plate load to the same 133 volt supply with the plate sitting at 65 volts.

A higher grid resistance would give more bass but if you want a dedicated bass channel, increase both coupling capacitors to 47 nF in that channel.

[amptramp]

As Paul says, you may get blocking or gulp distortion with 3.3 megohms.  Cathode bias is likely to resist overload better.

[PRR]

3.3 megohms is a bit low to get good contact potential bias....

This is for *microphone*. And work/school speech not Mick or Meatloaf. Mike levels may be VERY low.

Also in some places dynamic mikes were almost universal. Even if transformed to "50k" impedance these would drive a 12AX7 grid several tenths Volt positive without too-gross distortion.

Obviously this amplifier was not made for e-guitar. So changes are reasonable.

[ThermionicScott]

Fender used 5 Meg resistors for grid-leak bias, FWIW.

[sarakisof]

Any inputs fender inspired mods for one channel guitar and the other for bass guys?

Personally I vote to leave the grid-leak-biased inputs alone. Not everything needs to sound like a Fender, and there are already plenty of Fenders out there already. Why not let this be its own thing? The world is a richer place for it.

Yeap, totally agree. It sounds pretty nice as is right now. And in most of my old gear like this i prefer to leave things as is. "Does it sounds like a Bassman?" "No it is and sounds like a Geloso." Yeap, but knowing that this amp was designed for PA use

Obviously this amplifier was not made for e-guitar. So changes are reasonable.

, i will probably do changes on one of the two inputs, by using Fender cathode bias as Paul suggested.

How many 68k carbon resistors have you handy..??

haha 

[amptramp]

I have always been a fan of using PA amps for music.  They are designed for long life, plenty of derating and continuous or high duty-cycle use.  Many of them are used in stores where they operate 12/7 and are designed to need a minimum of care.  Compare this to guitar amps where they are trying to get the maximum power output from tubes and lifespan be damned.  It might matter when you are playing a gig in Snake Pit Junction and your overstressed amp finally decides it's time to roast a tube to death.

[sarakisof]

I'm also planning to experiment  using its 500ohm output for line level input on studio mixer, outboard gear, r2r tape etc.

Remember having read somewhere for this, that it would be also good to see if it can be wired floating with a third terminal for ground so to have a true balanced output option.
But i cannot understand clearly, what i have to check for this?

[PRR]

I'm also planning to experiment  using its 500ohm output for line level input on studio mixer, outboard gear, r2r tape etc. ...

NO!!

35 Watts at 500 Ohms is like 122 Volts !!

You do NOT need to drive lines with a "600 Ohm source". You want a "low impedance source", and even the 4 Ohm winding is plenty of Voltage for lines. Nearly 12V! You need an attenuator between any of these taps and line gear.

[sarakisof]

Woh i see Paul  thanks for this really appreciate.

You need an attenuator between any of these taps and line gear.

Is it better to build smthng like that (if it's possible) or get an output after the preamp stage.
In another push pull tube amp (EZ81, 3xecc83,2xUL84) of my Farfisa Foyer spinet organ, i am planning to get an output just right on the middle leg of master volume pot. But here it's another story i guess. If this is better, where should i do it in this amp?

[sarakisof]

I was always curious about how we can get a signal out from a tube amp. Understand that getting a pre-out is completely different than getting a whole circuit line out.
Was looking smthng like in 5e3 line out.



Could i do something similar here getting a "full" out?