Announcement

Collapse
No announcement yet.

5E3 w/ LTPI

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

  • 5E3 w/ LTPI

    Anyone ever experiment with using both triodes of V2 for a long tail pair phase inverter? This would mean losing the second gain stage, but since the LTPI provides some gain where the split load PI does not, will the LTPI make up enough of the lost gain to drive the 6V6's?

  • #2
    Try the Paul C mod.It was given to me years ago by an old guy I knew who designed tube circuits for RCA back in the '40's and '50's,I forget what he called it then,but when I asked about increasing the headroom in the 5E3 he looked at the schem and then drew the mod for me.Years later I saw it on the web as the "Paul C" mod,I dont recall where it is,but if you do a search for "Paul C Mod" you'll find it.You dont lose any gain,but you will notice an improvement in headroom.

    Comment


    • #3
      I'm familiar with the Paul C mod. I've done this to my Princeton Reverb some time ago. I liked it there. But that's not what I was looking for here. I was just interested in what the sound would be like. It would be more like an 18watt style preamp, but I'm not sure if there would be enough voltage swing to drive the 6V6's. If there were it might be a cool sound, with possibly a smoother cleaner tone until the 6V6's were pushed to breakup; getting the breakup from the 6V's instead of from the preamp.

      Comment


      • #4
        I think that the phase inverter in the 5E3 is one of the topographical features that give the amp its charateristic sound. IME the LTPI is one of the key things that makes the bigger amps sound like bigger amps.

        Essentially, what you're proposing is to change the amp so that the PI/Power Section looks a bit like a Brown Deluxe. but the the Preamp section ends up losing one gain stage in the process. I haven't crunched the numbers, but you might end up having to redesign the gain of the preamp to give the voltage swings necessary to keep the PI/output stage happy.
        "Stand back, I'm holding a calculator." - chinrest

        "I happen to have an original 1955 Stratocaster! The neck and body have been replaced with top quality Warmoth parts, I upgraded the hardware and put in custom, hand wound pickups. It's fabulous. There's nothing like that vintage tone or owning an original." - Chuck H

        Comment


        • #5
          Your right, it would change the characteristic sound of the amp. For better or for worse, that is the question.

          One big difference between what I'm considering and the brown Deluxe is the brown Deluxe uses nfb and fixed bias. I'm looking at cathode bias and no nfb, which would help with some compression and a smoother transition into distortion (if the circuit is capable of pushing the 6V6's out of the linear range and into distortion). It obviously works with the Marshall 18 watt, but that uses EL84s, which need less voltage to drive to clipping. And the Vol/Tone control is a little different, maybe lower loss there.

          Just thinking out loud here, wondering if anyone has ever played with it before.

          Comment


          • #6
            You will lose too much gain that way,probably easier to bias the 6V6's hotter to get more power tube saturation,if that is what you are after.

            Comment


            • #7
              Have you ever heard a 5G9 Tremolux? It has the Long Tail PI and fixed bias as well as 4 20uf caps mounted in the doghouse, definitely a transitional model and just killer tone, one of the best tweeds I've heard.

              Comment


              • #8
                Nice suggestion there Clyde1. I have not heard the Tremolux before. I am familiar with the 5E3, can you compare the two?

                I looked at both the 5E9 and the 5G9 tremolux schem's. The 5E9 is cathode bioased, but the PI is a paraphase, not quite was I was thinking of. ANd the 5G9 uses a LTPI, but has fixed bias. What I'm thinking of would be a mixture of the two, LTPI with cathode bias.

                But, tell me, is the amp capable of driving the 6V6's to clipping? Any comparison you could make to a 5E3 would be great.

                Thanks,
                Hasse

                Comment


                • #9
                  It's been a few years since I've heard that amp, very, very similar to the 5E3 the owner had in his collection, I did restorations on both, P12Q's in each,
                  recone in the Tremolux. I just found the Tremolux to be sweeter sounding.
                  I just happen to be doing a build with cathode-biased 6V6's, LTPI, cathode follower-driven tone stack, no feedback, tweed to the bone. I'm just wondering if I should go with JJ 6V6's or dig into my RCA stash. Yes, the Tremolux clipped quite nicely.

                  Comment


                  • #10
                    Don't know about the 5E3 but...

                    Hasserl

                    I don't know, off the top of my head, how hard you'd be driving the LTPI with the 5E3 front end & wouldn't that be the key? I've had good luck driving 6V6s to saturation in top boost type circuits. Assuming you want to stick with the two preamp tubes you could do it with 2 'cascaded' triodes. You may even have to split the plate on the second one to tame it down a bit.

                    It wouldn't be much of a 5E3 anymore though. O'Conner had something like this in TUT 3 except he used it with 84's.

                    Think 220k plate, 1K5 KR/25u bypass cap on V1a into a .0022 coupling cap & (Brite cap optional.) a 1MA Vol pot feeding V1b via a 10k grid resistor. V1b gets the standard 100K 1K5/25u treatment although this is where I'd maybe throw a Komet type split plate switch (and maybe reduce the value of the cathode bypass cap with a 'fat switch' for adding a parallel cap back up to the 25u range).

                    Then a plate driven two knob FMV type tone stack feeding a .01 PI input cap (.1 on the other side). The PI would be basically voxish with a 1K2 bias set resistor & a 47k tail. .047s to the 1K5 grid Rs for the power tubes & 1K-2 watt screens I think he called for. No NFB & this would surely drive the 6V6s as well as I'm getting in the top boost set ups.

                    Then again maybe you'll get enough drive for the PI from the 5E3 front end...
                    Philip Morrison

                    Comment


                    • #11
                      And why not atacking the power tubes with a cascode phase inverter Big huge fckng grin

                      Comment


                      • #12
                        Now that youo mention it, I remember a thread from Wild Bill IIRC where he posted information about a couple of different cascode PI's. I believe it was on this new board, bnot the old one. I'll dop a search and see what I can find.

                        Satamax, if you've got any info or a link can you post it?

                        Thanks,
                        Hasse

                        Comment


                        • #13
                          Okay I found it. It was Ray Ivers, not Wild Bill (sorry about that Ray!).

                          I'm going to save both of those schematics and think about them. I'm assuming either one would provide more voltage swing than a LTPI, correct?

                          Comment


                          • #14
                            One other way to overdrive the 6V6s is to increase their gain through positive feedback. One way is to cross couple the screens and plates. Note this means that the power supply node will have to deliver plate and screen current so design accordingly.

                            Comment


                            • #15
                              Hi everybody!

                              Haserl, thoses cascode PI (well they have another name but can't remember!) are suposed to be able to go up to a 1000 in gain No preamp whatsoever then You could atack the power stage straight

                              Bye.

                              Max.

                              Comment

                              Working...
                              X