What would you do with a CBS 70w Super Reverb?

[slorinczi]

I have my bandmate's 70W SFSR on the bench and I'm trying to do some gentle mods. Replaced electrolytics and phase inverter coupling caps (they no longer match, if they ever did), converted to adjustable bias, and did the simple BF phase inverter conversion.

Have I gone too far? I'm really not trying to make this into a different amp, just make it a little friendlier sounding. I'm not going to mess with the UL output, for instance. Thoughts, input, rages, whatever....

Thanks,
Seth

[Blair]

Ul Conversion

slorinczi
Let us know your thoughts on how the conversion works out. It could be interesting.
Cheers,
Blair

[lowell]

If I were you I'd try disconnecting the UL taps... it's only 2 wires and you can put them back on if you don't like it. Just be sure to put some heatshrink on the ends after removal so they don't short to something. You will then need to add 2 470ohm 2-5watt resistors to the screens and connect the screen supply to them instead of the UL taps.

I also recently saw something very interesting in a schematic... it looked as though it was a great way of reducing high end... which I find these SF amps need. The mod is to add a small value cap (150pf ?) and a resistor or pot across the phase inverter outputs AFTER the coupling caps. The out of phase signals via the cap, which has a high resistance to low frequency thus only passes highs, will cancel eachother out. I've never tried this but will get around to it. Let me know if you do.

[slorinczi]

Interesting...I wasn't planning on going that far but I see the argument. I'm a little confused about the screen grid B+ however; if I understand correctly, I'd have to tap it off somewhere, as it's not normally provided in an ultralinear circuit. (And then run it through 470 ohm dropping resistors).

Below is an excerpt from the schemo. If you have a moment, maybe you can enlighten me as to where the B+ is coming from?

Best,
Seth

[lowell]

Hi,
Oops.. sorry, for some reason I thought you had a 4-output-tube amp. This is even easier now.

Ok so disconnect the wires on pin 6 (I THINK!) - one side of the 470ohm resistors should be soldered there... leave the resistors where they are. Then you will be connecting that side of the resistors to the B+ which is the RED wire in the chassis. The RED B+ wire should be soldered on the eyelet board towards the end closest to the power tubes somewhere. On the schematic it is the point after the standby switch that states 500+ volts... see it? You will want to leave the 470ohm resistors mounted on the tube sockets and just run a new wire from the point on the eyelet board to the socket (pin 6 - I THINK!) where the old screen tap was soldered. Use wire that can handle this voltage/current. 18 gauge is good. Hope this helps, let me know!

[Steve Conner]

I remember a previous thread on the board where someone mentioned that they tried disconnecting the UL taps on a CBS Fender, and they said it made hardly any difference to the tone. Blackfacing the whole preamp would probably be more productive. The CBS amps run high voltages on the preamp tubes, and I heard that lowering them to blackface values mellows the amp out.

[hasserl]

Hi,
Oops.. sorry, for some reason I thought you had a 4-output-tube amp. This is even easier now.

Ok so disconnect the wires on pin 6 (I THINK!) - one side of the 470ohm resistors should be soldered there... leave the resistors where they are. Then you will be connecting that side of the resistors to the B+ which is the RED wire in the chassis. The RED B+ wire should be soldered on the eyelet board towards the end closest to the power tubes somewhere. On the schematic it is the point after the standby switch that states 500+ volts... see it? You will want to leave the 470ohm resistors mounted on the tube sockets and just run a new wire from the point on the eyelet board to the socket (pin 6 - I THINK!) where the old screen tap was soldered. Use wire that can handle this voltage/current. 18 gauge is good. Hope this helps, let me know!

Note that connecting the screen grids up like this will cause the tubes to run as triodes. You can try this for shits and giggles, but I wouldn't expect to find it very pleasing. The power will be reduced and the tone will be darker. Not sure what the 500+ volts will do for the screens though, may not be too good for them. My data sheet for 6L6GC's shows a maximum voltage for the plates/screen grids of 450 volts in triode.

[lowell]

No this is not making the output tubes into "simulated" triode mode. That would be if the screen resistors were connected to the plates of the output tubes. Here in this mod we are connecting the screen resistors to the B+ - just like any amp you'd see without an ultra-linear output transformer.

I agree that the preamp would be a good place to alter tone... however as in my previous post I did not touch the preamp in an amp that I converted to cathode-bias and removed UL taps and the amp sounded MUCH better. But as usual all this is subject to the "modder's" opinion and ears.

[slorinczi]

More UL...Plus Reverb!

Still unsure as to whether I'm going to try messing with the UL; it seems like the consensus (and my gut feeling) are that the amp just "is what it is."

That said, I'm still trying to figure out the mod. I understand that the screens need B+. Lowell, are you suggesting that the 500V supply feed BOTH the CT of the OT and the screens? That's where I'm hung up.

And while we're on the subject of mods, a tech told me to double up the value of the cap feeding the reverb. However, I'm not sure if me meant the cap feeding the driver or the return. Any thoughts? (Semi-handy diagram below.)

Thanks,
Seth

[lowell]

Yes exactly! The CT and pin 6 side of 470ohm should be one and the same - electronically speaking. Typically there is a choke b/t the CT and the screen supply filter cap but I don't think it's completely necessary.

If you increase the "in" coupling cap then the reverb will affect more low-end because a larger cap allows more lows through. A cap in series w/ the grid with a resistor to ground (as in the schematic) creates a high-pass/low cut filter. Cutoff frequency for the input to the reverb there is 320hertz. I'd try the "in" cap first and take a listen... The "out" cap has a cutoff frequency of 530hertz so the reverb return isn't allowing much below that back into the circuit. If you'd like to know what you're doing frequency-wise use this:

cutoff frequency=1/(6.28(R*C))
(6.28= 2*pi)

In your schematic the "in" cap is 500pf... or .000,000,000,5 farads and R is 1meg, or, 1,000,000 ohms.

Fc=1/(6.28(1,000,000*.000,000,000,5)) = 320hertz

hope this helps.

[slorinczi]

Yes perfectly, thanks much lowell!

[lowell]

Oops I just reread preamp basics in Kevin O'Connor's to be sure I was guiding you in the right direction. You actually also need to take the plate resistor into account when calculating frequency cutoff.

The plate resistor in a typical preamp stage is usually 100k. This value needs to be calculated in parallel w/ the 1meg of the high-pass filter. So 100k in parallel w/ 1meg = 1/(1/100k + 1/1,000,000)=91k.

Fc=1/(2*pi(91,000*.000,000,000,5)=3,500=3.5k hertz.

[hasserl]

Lowell, what you are describing IS triode operation. You're connecting the screen grids to the plate supply, the screens will operate at the same voltage as the plates and the tubes will operate as triodes. In a typical amp there is either a choke or a dropping resistor and a filter cap seperating the plates from the screen grids, this decouples the two stages. In what you're describing there is NO decoupling. It IS triode operation.

slorinczi, don't waste your time. Leave the UL alone. Try disconnecting the nfb as was suggested earlier. It is much easier and will sound much better. Reduce preamp voltage in the power supply as Steve C has suggested.

[lowell]

Ok you're being a little pushy and honestly quite rude here to just discount my suggestions and tell him not to try it. If he wants to try it let him try it. This forum is about people sharing their knowledge not discounting someones opinion, is it not?

Triode mode from what I know is connecting the screen resistor directly to the plate... this is not the same as connecting to the CT or the B+. One half of the output transformer is b/t the plate and screen if screens are tied to B+.

[hasserl]

Sorry about that, but the fact is that is triode operation. You're tieing the plate supply and the screen grids to the same power supply node, with no decoupling between the two.

Don't mean to be pushy or rude, but maybe you're being a little too thin skinned?

Try it, or don't. If you want to experiment, why the hell not? Except the 500 volts exceeds the max ratings for the screen grids. But what the hell, go for it. What's a couple of 6L6's anyway?