I plan to start messing around with a diy geetah preamp and wanted to try something different with the reverb circuit. It seems that most of the tube driven reverb units I see use a transformer, but that's driving a tank with an 8 ohm input coil. What about driving the tank directly with a tube? For example a type F with an input coil impedance of 1.4k @ 1kHz, should be around 230mH. In the accutronics literature they state to match the impedance and then say to use a current source with a source resistance of at least 5x the impedance at 1kHz. I was thinking of driving the type F with one section of a 12AU7 and AC couple it, set the high pass filter accordingly. Not too sure how to simulate a reverb tank but I was messing around with driving an inductance and the one triode of a 12AU7 seems to do just fine. I might give it a try but I was curious if anyone else has tried this out there. I read some stuff about people using a pentode as a driver, I have a bunch of 6AU6A's laying around.
-Bird
-Bird
Ampeg did this with some of their later guitar amps:
http://www.ampegv4.com/pdf/Ampeg_V4_1976.pdf
Take a look. 1/2 of a 6CG7 cap coupled to the input of a reverb tank. IIRC, the input impedance of the tank Ampeg used is around 2200-2300 ohms.
http://www.ampegv4.com/pdf/Ampeg_V4_1976.pdf
Take a look. 1/2 of a 6CG7 cap coupled to the input of a reverb tank. IIRC, the input impedance of the tank Ampeg used is around 2200-2300 ohms.
THANK YOU!!!
Nice to see it in a commercial application by Ampeg.
The 6CG7 looks electrically very similar to the 12AU7 too.
I did some poking around and it seems the replacement reverb tanks for the V4 is 4FB3D1B which the input coil is F, which is 1.4k @ 1kHz. It's the same input impedance tank I was looking at. Very cool.
Nice to see it in a commercial application by Ampeg.
The 6CG7 looks electrically very similar to the 12AU7 too.
I did some poking around and it seems the replacement reverb tanks for the V4 is 4FB3D1B which the input coil is F, which is 1.4k @ 1kHz. It's the same input impedance tank I was looking at. Very cool.
Hammond use tube plates DC blocked by capacitors for reverb. See H100, various 40's 50's tone cabinets on captain-foldback.com. I think the H100 used a 7247 which had a lot of drive current, there at the end of tube design.
I think the 40's stuff used a piezo crystal as a driver. They are rare as Rok's eggs on the parts market, but I've always wondered if a bar-b-que lighter piezo crystal would do the job for $2. I don't have a dead Hammond tone cabinet with crystal drive to experiment with, though.
I think the 40's stuff used a piezo crystal as a driver. They are rare as Rok's eggs on the parts market, but I've always wondered if a bar-b-que lighter piezo crystal would do the job for $2. I don't have a dead Hammond tone cabinet with crystal drive to experiment with, though.
This maybe of interest to you:
The Lamington Reverb – a better valve reverb unit | Valve Heaven
Cheers,
Ian
The Lamington Reverb – a better valve reverb unit | Valve Heaven
Cheers,
Ian
Do you recommend the follower to drive the reverb tank or the output after the recovery stage?
Interesting!! I will have a look at the Hammond schematics for more inspiration.
Good read and neat little inexpensive project, basically on par for what I was thinking of what I want to implement.
A follower whether tube, BJT or mosfet is used when you want to transform an rather high impedance (tube plate) into a low impedance (reverb tank).
Reverb tanks are available with input impedances from 8 ohms to a few K ohms. The usual choice for a tube amp is something in the 1 to 5 K ohm range, which can be driven from a (DC blocked) tube plate. A tube with a few mA of drive capability is the usual choice. The 12AU7, 12AT7, 6SN7 and equivalents are often chosen. Note that the 7247 (12DW7) previously mentioned is dual triode with a 12AX7 half and a 12AU7 half. The 12AU7 half would drive most reverb tanks without a transformer.
The 12AX7 is a poor choice since is usually runs at about 1 mA of plate current. A 12AX7 with a mosfet or tube follower would work though.
I have collected a few reverb tanks, all from solid state amps, but haven't had the time to mess with them yet. I love that 60's surf guitar sound, but until I have the time, I will have to rely on Overloud's TH3.
Reverb tanks are available with input impedances from 8 ohms to a few K ohms. The usual choice for a tube amp is something in the 1 to 5 K ohm range, which can be driven from a (DC blocked) tube plate. A tube with a few mA of drive capability is the usual choice. The 12AU7, 12AT7, 6SN7 and equivalents are often chosen. Note that the 7247 (12DW7) previously mentioned is dual triode with a 12AX7 half and a 12AU7 half. The 12AU7 half would drive most reverb tanks without a transformer.
The 12AX7 is a poor choice since is usually runs at about 1 mA of plate current. A 12AX7 with a mosfet or tube follower would work though.
I have collected a few reverb tanks, all from solid state amps, but haven't had the time to mess with them yet. I love that 60's surf guitar sound, but until I have the time, I will have to rely on Overloud's TH3.
Okay thank you for clarifying.
I intended to use a high impedance tank and drive it with AC coupled tube plate, specifically the 12AU7.
I just scratch my head to why fender being cheap would use a transformer to couple to an 8 ohm tank when they could have just used a different tank with higher impedance and drive it with one triode?
And yeah, 60's surf guitar drenched in Reverb is freakin great!!! I need a good spring reverb.
I intended to use a high impedance tank and drive it with AC coupled tube plate, specifically the 12AU7.
I just scratch my head to why fender being cheap would use a transformer to couple to an 8 ohm tank when they could have just used a different tank with higher impedance and drive it with one triode?
And yeah, 60's surf guitar drenched in Reverb is freakin great!!! I need a good spring reverb.
> why fender being cheap would use a transformer
Fender was not THAT cheap.
Also: who says the transformer was expensive?
Hammond drove reverbs with Push-Pull amp (6SN7?). Fender's SE 6F6 was already cheaper. And I seem to recall that many-many communications receivers used 250V with 6F6-like tube at 15K-25K loading for about 1 Watt. Such radios were going out of style. Leo may have found a Real Good price on a crate-load. Maybe the crate lasted longer than the supply of also obsolescent 6F6, because other tubes were used late in the run.
> can achieve a higher drive level
Same reason you bring a 100-Watt amp when you only need microWatts in each fans' ear. The room (pan) is a lossy thing. You do not want your precious tones to be sucked-down below the roar of the beer-coolers or the tube-hiss. You want to hit the spring HARD.
> 60's surf guitar drenched in Reverb is freakin great!!!
Then don't wimp out. Slap that thing silly. The 2*12AT7 driver in the combos is OK, but the Real Surf Reverb is the whole 6F6/6V6 most-of-a-Watt driver.
Fender was not THAT cheap.
Also: who says the transformer was expensive?
Hammond drove reverbs with Push-Pull amp (6SN7?). Fender's SE 6F6 was already cheaper. And I seem to recall that many-many communications receivers used 250V with 6F6-like tube at 15K-25K loading for about 1 Watt. Such radios were going out of style. Leo may have found a Real Good price on a crate-load. Maybe the crate lasted longer than the supply of also obsolescent 6F6, because other tubes were used late in the run.
> can achieve a higher drive level
Same reason you bring a 100-Watt amp when you only need microWatts in each fans' ear. The room (pan) is a lossy thing. You do not want your precious tones to be sucked-down below the roar of the beer-coolers or the tube-hiss. You want to hit the spring HARD.
> 60's surf guitar drenched in Reverb is freakin great!!!
Then don't wimp out. Slap that thing silly. The 2*12AT7 driver in the combos is OK, but the Real Surf Reverb is the whole 6F6/6V6 most-of-a-Watt driver.
Good points about back then transformers being inexpensive, and maybe he got a deal on them to boot.
I don't want to wimp out. My attempts at modeling some circuits showed me that the single 12AU7 triode drove a type F tank (1.4k @ 1Khz) better than the 2*12AT7 + transformer drove the type A tank (8 ohm @ 1kHz). I did not try the larger octal power tube + transformers in the model yet. I will and report back.
The 12AT7 transformer was 25k:8 and I believe the 6V6 transformers were 5.5k:8. The 12AT7 triodes in parallel would look like ~2 ohms on the secondary driving the 8 ohm tank. The 6V6 ran as a pentode looks like ~80 ohms on the secondary driving the 8 ohm tank. The 6V6 would act more like a current source than the 12AT7 looking more like a voltage source.
I don't want to wimp out. My attempts at modeling some circuits showed me that the single 12AU7 triode drove a type F tank (1.4k @ 1Khz) better than the 2*12AT7 + transformer drove the type A tank (8 ohm @ 1kHz). I did not try the larger octal power tube + transformers in the model yet. I will and report back.
The 12AT7 transformer was 25k:8 and I believe the 6V6 transformers were 5.5k:8. The 12AT7 triodes in parallel would look like ~2 ohms on the secondary driving the 8 ohm tank. The 6V6 ran as a pentode looks like ~80 ohms on the secondary driving the 8 ohm tank. The 6V6 would act more like a current source than the 12AT7 looking more like a voltage source.
> looks like ~80 ohms on the secondary driving the 8 ohm tank. ...more like a current source
Tanks are generally driven by current-source.
I dunno what these Ohms are proving. Hammond's push-pull lo-Mu triodes make about 1 Watt. Fender's transformer-coupled 6K6/6F6 will deliver near 1 Watt. The Fender 12AT7 scheme will approach 0.5 Watts.
Resistance-coupled amplifiers are much less efficient than transformer or direct coupled. The Ampeg scheme puts over 5 Watts in tube and resistor, but can only get 0.1W into the tank.
Opamps are also in the 0.1W range. Noteworthy is that an opamp drive is invariably teamed with an opamp recovery amp, which may have 6dB less hiss than a tube recovery amp. So to a first approximation the S/N is similar, 12AT7 or opamp.
Tanks are generally driven by current-source.
I dunno what these Ohms are proving. Hammond's push-pull lo-Mu triodes make about 1 Watt. Fender's transformer-coupled 6K6/6F6 will deliver near 1 Watt. The Fender 12AT7 scheme will approach 0.5 Watts.
Resistance-coupled amplifiers are much less efficient than transformer or direct coupled. The Ampeg scheme puts over 5 Watts in tube and resistor, but can only get 0.1W into the tank.
Opamps are also in the 0.1W range. Noteworthy is that an opamp drive is invariably teamed with an opamp recovery amp, which may have 6dB less hiss than a tube recovery amp. So to a first approximation the S/N is similar, 12AT7 or opamp.
Just trying to figure out why I hear a difference. I can hear a difference between the 6V6 or 6K6 type reverb drivers vs the 12AT7 drivers. I like the pentode much better. As you said tanks are often driven by a current source but I would think the 12AT7 type isn't one. Maybe thats why I hear a difference? Maybe it just less powerful? I am going to try and get my hands on an Ampeg and play through it, my buddy loves his V4.
"What PRR said" 
Short answer, you are analyzing "small signal" output impedance so pentodes look high, triodes look low, cathode followers look low impedance ... all that applies under small signal, small level conditions.
Once you drive them hard, and need power twisting those springs, power rules-
So transformer coupled Hammond 1W or Fender 1W better than transformer coupled Fender 0.5W way better than Ampeg/Traynor resistive coupled 0.1W or direct coupled TL072/LM1458 0.1W .
Want *excellent* SS reverb?, drive the hell out of them with a small power amp, adapted to reverb tank impedance:
As an example, check what "Twin Killer" Lab Series L5 did (BB King´s amplifier of choice):
Some details to notice:
* it is a full power amp, no kidding: Op Amp driving drivers and heatsinked power transistors.
* own +/- 25V supply, from own PT winding.
* tank is hard to drive 8 ohm nominal input (1 ohm DC )
* they add 56 ohms in series to turn voltage drive into curret drive.
Notice 56 ohm resistor is rated 12W
* mini power amp shows "up to 22VAC output" (which I find optimistic, wouldn´t expect more than 14 or 16V AC, in any case *a lot*)
*** reverb tank drive is given as current, 350mA at a given input test voltage.
Lab Series amplifiers were meant as "Twin Killers" and so boasted huge clean power output (actually some 200W RMS on the "100W" models), warm but punchy sound, big magnet speakers (many used Electro Voice speakers), very high quality construction, weighed almost as much as a tubed Twin ... excellent reverb ... sadly were phased out because to Norlin management "selling only 14000 in one year was not worth it".
Short answer, you are analyzing "small signal" output impedance so pentodes look high, triodes look low, cathode followers look low impedance ... all that applies under small signal, small level conditions.
Once you drive them hard, and need power twisting those springs, power rules-
So transformer coupled Hammond 1W or Fender 1W better than transformer coupled Fender 0.5W way better than Ampeg/Traynor resistive coupled 0.1W or direct coupled TL072/LM1458 0.1W .
Want *excellent* SS reverb?, drive the hell out of them with a small power amp, adapted to reverb tank impedance:
As an example, check what "Twin Killer" Lab Series L5 did (BB King´s amplifier of choice):
Some details to notice:
* it is a full power amp, no kidding: Op Amp driving drivers and heatsinked power transistors.
* own +/- 25V supply, from own PT winding.
* tank is hard to drive 8 ohm nominal input (1 ohm DC
)* they add 56 ohms in series to turn voltage drive into curret drive.
Notice 56 ohm resistor is rated 12W
* mini power amp shows "up to 22VAC output" (which I find optimistic, wouldn´t expect more than 14 or 16V AC, in any case *a lot*)
*** reverb tank drive is given as current, 350mA at a given input test voltage.
Lab Series amplifiers were meant as "Twin Killers" and so boasted huge clean power output (actually some 200W RMS on the "100W" models), warm but punchy sound, big magnet speakers (many used Electro Voice speakers), very high quality construction, weighed almost as much as a tubed Twin ... excellent reverb ... sadly were phased out because to Norlin management "selling only 14000 in one year was not worth it".
Last edited:
Or one of these:
Output Transformer for 6G15 Add-On Reverb Unit OT47605 MADE IN USA
FENDER CHAMP STYLE 5W OUTPUT TRANSFORMER TF103-48 Single Ended Ultra-Linear 4 / 8 Ohm MADE IN USA
I have tried the Champ replacement OPT, although mine is an older version with only an 8 ohm secondary. It's a really tiny transformer, but works pretty good in a Champ style amp or for driving a reverb tank. I have tried it with 6V6's, 6AQ5's and several other small tubes like 12L6's and 50C5's.
The smallest Edcor is probably a better quality transformer, but it takes forever to get them.
EDCOR - XSE10-5K
The tiny Hammond 125ASE is also an option since you have several taps to try for the best performance.
https://www.hammfg.com/files/parts/pdf/125ASE.pdf
I used the larger 125CSE as the OPT in my Turbo Champs back around 2000, when they were cheap.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.