Announcement

Collapse
No announcement yet.

Tube reverb driver circuits

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Tube reverb driver circuits

    Working on an older amp with a sub-par reverb circuit has led me to consider, purely theoretically, how it might be improved. The amp in question used a primitive solid state driver straight from the guitar input and returned the reverb signal via the cathode of the first 12AX7.

    Let's say that you have an amp with a power transformer that could supply the heaters and high voltage for an additional small signal tube like a 12DW7/7247. The reverb tank is the Type 4 1475/2250 Ohm variety.

    Is it possible to do an adequate job of driving the 1475 Ohm input coil with the AU7 half of the 7247 tube? I found the following discussion that suggests you can do it with both sections of a 12AU7 paralleled.

    The Valve Wizard

    Also, most reverb drive transformers I've seen have secondaries for 8 Ohm tanks, but what's the best approach for the higher impedance tank, assuming that you don't want to replace the tank? Do you use a transformer or drive the tank's input coil from the anode of the driver tube?

    A couple of vintage tube reverb drivers I've run across, the Fisher K-10 SpaceXpander and the Electro Tone Organ Mate drive the tank via half a 12AU7 wired as a cathode follower with an impedance matching transformer in the cathode circuit, but I'm not sure about the specs of the transformer they used. Comparing the two, it does look like Electro Tone "borrowed" the driver design from Fisher down to the specific component values.

    I know that you could add a small, lower-voltage supply for a solid-state driver, but, for the sake of this discussion, let's say that you want to see if you can do it within the limits of the existing power supply and with tubes.

    The AX7 half of the 7247 would be for recovery.

    David

    P.S. One of the best sounding tube reverbs I've run across was in a Supro amp that used a 6V6GT as the driver, but in this case, we're more limited.

  • #2
    Why make it a mystery, WHAT amp are we talking about? That way we can look at the circuit you have to work with and have some idea what parameters to consider.

    More importantly, just what are we trying to accomplish? Can;'t discuss solutions without knowing the problem. In what way is the reverb sub-par, and what do you want it to do differently? What does improved mean?

    I have seen people complain that their amp reverb was wimpy. Why? Because on the Fenders they have plaayed, if you turn the reverb up only as far as 4 or 5, it is already WAY more reverb than even surf music can use, but on their new amp, you can turn the reverb all the way up to 8 before it gets stupidly verbed. Seems to me the "wimpy" one is more useful.

    SO does yours sound tinny or dull or weak? Or it doesn;t ring out?

    And whatever the deficit might be, how will you decide it is a drive side problem rather than a recovery side problem?

    Tube reverb drives with transformers use the low Z input pans, the 4ABxxxx types. Yours is a 4EB or 4FB type, with high Z input. I am too lazy to look up which. SO you are not looking for transformer drives.

    One sample circuit to consider is the sort used by AMpeg in some later models, like in the V series.

    You mention that we could always add a low voltage supply for a solid state drive. But you started out saying there was already a solid state drive in there, so I would have to guess the low voltage supply already exists.

    Primitive or not, what is wrong with the SS drive already there?
    Education is what you're left with after you have forgotten what you have learned.

    Comment


    • #3
      How about a 6DR7? You get a reeeeeeally low Ra triode (900 ohms IIRC) and a triode that's a lot like a 12at7. The DR7 should be able to drive one of those tanks using a cap like an old magnatone circuit.

      Or...try driving the tank with a 12au7 and plan to use a series resistor like Merlin says- makes the tube appear to be a current source. Use a single LND150 as return- it's capable of a gain of over 100 without any trouble and uses no filament current.

      I've been toying with a circuit to add to amps that would use a 6dr7 high gain side to amplify an effects return, the low gain side to drive the verb (thru a cap) and the mosfet to recover the verb. A high value pot would connect in parallel with the PI input and the mosfet output would connect to the wiper of the pot allowing you to blend the reverb in at whatever level you prefer. Haven't even breadboarded it- just thought it'd be a good use for all the 6DR7's I have around.

      I've found that it's difficult to drive a tank that was designed for an op-amp with a tube and do it well. I haven't tried Merlin's SRPP circuit yet but I will one day- it looks pretty interesting. I'd assume it works well of he wouldn't have shared it!

      jamie

      Comment


      • #4
        Enzo,

        The amp was the Gregory Reverb 1500 I've been working on. The big problem with the stock reverb circuit was that its input impedance was far lower than the surrounding signal circuitry. When you switched it on--the on/off switch is in series with the signal input to the reverb drive--the overall volume plummeted and the tone went to hell. The highs disappeared due to high-pass filter at the input. It also had a nasty tendency to oscillate. I fixed it by sticking a JFET source follower in front of it, but it's still not very full-sounding. It's just noticeable if you have it all the way up. I'm quite sure it never worked right to begin with. According to a guy who has a Gregory amp site, this is the only existing example of this amp with its original circuit that he knows of. The only other one known to exist has been gutted by its current owner and used as a platform for a completely different amp.

        Still, this amp is as fixed as it's going to get. Client does not want to do a more extensive remodeling, but I'd like to think through it for the future.

        I measured the matching transformer (TMI DT-104) that Electro Tone used in the Organ Mate. With a 1.5k resistor on the secondary, my ESI impedance bridge gives me 7k on the primary at 1kHz, but this no longer seems to be a readily available transformer. I haven't had my Organ Mate reverb in use in a while, but I recall that it's a decent sounding unit. Still, no, this isn't the standard 15k : 8 /4 Ohm reverb driver transformer.

        I'll take a look at the other circuits you mentioned. That's really what I was asking for--examples of ways to work with this type of tank. I seem to have accumulated quite a few of them as Hammond spare parts.

        Comment


        • #5
          From my personal experience of playing around with reverb circuits in the past year or so, I found that I like the high impedance tanks. One because they don't need an extra transformer which saves money and space, and second because the circuit is so easy to implement. The example on the Valve Wizard site is great-- remember that he also mentions using a power triode in place of the paralleled ecc82. I have used a triode strapped el84, ecl82, ecl86 etc. and have been really happy with the results. If you're worried about extra filament current needed the ecl86 is a nice option, plus it's similar to an ecc83 with an el84 in one bottle. Easy enough to drive and recover.

          Comment


          • #6
            Thanks for the suggestions :-)

            Interesting you should mention the 6DR7. I ran across a tube reverb schematic that specified a 12DR7, but I couldn't find a datasheet for the type. I was thinking it was a misprint for 12DW7. I've never encountered one myself, but I looked at the datasheet for the 6DR7. The high-mu section seems to me to be more like the octal 6SL7 than the 12AT7, and you get the low plate resistance of the other half at the cost of the same heater current as a 6L6. Still, it might be a device to look at.

            Also, I'm not sure that the 1475 Ohm input coils were originally designed with op-amps in mind. They were certainly used in older gear. Hammond typically drove reverb tanks in tone cabinets like the PR-40 and its predecessors with a 6SN7 or 12BH7 in push-pull via a center-tapped output transformer--convenient for them because the console organs send a balanced line signal to the power amp. The tank specified was a Type 4K, and, embarrassingly, I can't locate the input/output impedances for that tank. Hammond used their own codes prior to spinning off Gibbs/Accutronics.

            The practice of driving reverb tanks via a dual triode wired push-pull appears to have been rediscovered by London Power, though it isn't cheap. It looks to be made by the "other" Hammond:

            Construction Supplies - - Push-Pull Reverb Driver Transformer

            I wonder if this is an off-the-shelf transformer marked up and sold as a boutique item.

            Comment


            • #7
              Originally posted by BiBi View Post
              If you're worried about extra filament current needed the ecl86 is a nice option, plus it's similar to an ecc83 with an el84 in one bottle. Easy enough to drive and recover.
              I'm familiar with the ECL86--they're used in the AO-44 reverb recovery amps found in Hammond A-100s. In the A-100, Hammond experimented with driving the reverb tank from the main power amp's output through a light-bulb compressor/limiter circuit. Unfortunately, they tend to sound muddy. Many people remove the AO-44 amps and install Trek II reverbs that interface with the organ's preamp. Trek II's RV-1 spring reverb also uses a simple compressor circuit, but it sounds great. It's my gold standard for spring reverb.

              Comment


              • #8
                Originally posted by Enzo View Post
                One sample circuit to consider is the sort used by Ampeg in some later models, like in the V series.
                Enzo, I looked at the Ampeg schematics. It looks like Ampeg is running half the 6CG7 at about 10% over its datasheet maximum both in terms of cathode current and plate dissipation. I ran a search and found at least one case where the 10k 5 Watt plate resistor had opened up.

                That's what I call smokin' reverb :-)

                David

                Comment


                • #9
                  Originally posted by Rhodesplyr View Post
                  Hammond typically drove reverb tanks in tone cabinets like the PR-40 and its predecessors with a 6SN7 or 12BH7 in push-pull via a center-tapped output transformer
                  I've done it too! It works fine. I used a 12AU7 in self-split, a low impedance tank and the smallest push-pull OT I could find, which was the size of a Fender Champ one and cost me the equivalent of $15.

                  I've never seen the London Power transformer before, and it could well be custom-made for them.
                  "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

                  Comment


                  • #10
                    Originally posted by Rhodesplyr View Post
                    I wonder if this is an off-the-shelf transformer marked up and sold as a boutique item.
                    I think I;ve read somewhere that it's a custom.
                    However, you can usually get a small PA line matching transformer to suit the task!
                    Or perhaps this (it is only $8!! Compare with LondonPower $62(CAD)- I am jealous of you Ameicans!) https://taweber.powweb.com/store/lopo_ot.jpg
                    Last edited by Merlinb; 04-17-2010, 09:57 PM.

                    Comment


                    • #11
                      Originally posted by Rhodesplyr View Post
                      Is it possible to do an adequate job of driving the 1475 Ohm input coil with the AU7 half of the 7247 tube? I found the following discussion that suggests you can do it with both sections of a 12AU7 paralleled.
                      One AU7 triode with 10k load and 10k in series with the tank should drive the tank pretty well yeah, if you run the AU7 with about 10mA quiescent current. In SIM I'm seeing up to 3mA RMS into the tank- nice!
                      Last edited by Merlinb; 04-17-2010, 10:12 PM.

                      Comment


                      • #12
                        Originally posted by Steve Conner View Post
                        I've done it too! It works fine. I used a 12AU7 in self-split, a low impedance tank and the smallest push-pull OT I could find, which was the size of a Fender Champ one and cost me the equivalent of $15.
                        Excellent! Of course, to keep things completely balanced, Hammond then used a balanced recovery circuit, employing both sections of a 6SC7. As you can tell, Hammond took a "maximum parts count" approach to design. I mean, why employ a shortcut like a crossover when you could have three separate output transformers on one amp chassis, one for bass, one for treble, and one for reverb?

                        Comment


                        • #13
                          Originally posted by Merlinb View Post
                          I am jealous of you Ameicans!)

                          I actually think of myself as Irish ; hence the "fine" intellect ; and no I am not going to give you any more free technology to publish in your book.

                          -g
                          ______________________________________
                          Gary Moore
                          Moore Amplifiication
                          mooreamps@hotmail.com

                          Comment


                          • #14
                            Originally posted by Merlinb View Post
                            One AU7 triode with 10k load and 10k in series with the tank should drive the tank pretty well yeah, if you run the AU7 with about 10mA quiescent current. In SIM I'm seeing up to 3mA RMS into the tank- nice!
                            What amplitude signal input does the AU7 need to achieve this drive level?

                            Comment


                            • #15
                              Originally posted by mooreamps View Post
                              I actually think of myself as Irish ; hence the "fine" intellect ; and no I am not going to give you any more free technology to publish in your book.

                              -g
                              Gary, I have never even seen you publish a schematic. Just where am I (or anyone else) supposed to be stealing these ideas from?? You haven't even said what ideas I have used which are supposedly yours! All you ever do is make 'half accusations' or 'implications'.

                              I suggest either you make a proper allegation, in which case I will explain exactly what alternative source I got the idea from, or else stop acting like child; you are embarrassing yourself, and I dare say liable to get chucked off yet another forum.

                              Comment

                              Working...
                              X