Announcement

Collapse
No announcement yet.

Preamp supply sag

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

  • #16
    I've also thought about setting up a fixed/cathode bias switch to see if cathode bias might get me a hint of a saggier feel, but I vaguely recall reading something relative to the cap-coupled bias setup I'm using indicating that it doesn't like to be run floating which it would be I suppose if put on a switch.

    Comment


    • #17
      The lack of any sag seems normal. Preamp tubes are voltage amplifiers operating in class A. So they're not doing any significant work amplifying voltage and even if they were the amount of work wouldn't change because they swing from the center of their bias (or mostly anyway). So the preamp tubes are already doing as much work just sittin' there as when they are amplifying. This is also why your 400 ohm resistor doesn't create more sag. The current in the preamp is fairly constant. Adding resistance might drop a little voltage but it can't make the steady state respond to what isn't happening. Preamp voltage sag is a consequence of how the power amp affects the voltage supplied to it. Without a power amp to do that any voltage sag analogous to signal level needs to be artificially induced.
      "Take two placebos, works twice as well." Enzo

      "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

      "If you're not interested in opinions and the experience of others, why even start a thread?
      You can't just expect consent." Helmholtz

      Comment


      • #18
        Right-o Chuck, thanks for responding. This I understand. But, this isn't a standalone preamp with it's own PS. The power amp is pulling off the midpoint of the totem filter cap setup and the preamp is pulling off the top for the higher (doubled) voltage. Neverthelss they're both operating off the same PT and the same tube rectifier. So as the power amp pulls current - idles @ 70 ma, rises to around 105 to 110 ma at max - and the voltage drops a little bit (maybe I need a saggier PT…), I would *expect* the preamp voltages to at the least sag the same voltage as the output is sagging. However, they're not. They're not sagging at all, and in fact, they actually bump up a few volts. This is what is not making sense to me and I'm wondering if the way I have the PS set up is the cause of this in a way that I don't understand. I also don;t understand why the power amp voltages were not affected basically at all by a big sag resistor, because whether the output is pulling off the midpoint or the top of the totem pole, I would think it all still has to come across that sag resistor. Maybe not? Maybe there is something going on with the PT center tap that I don't understand.

        Edit: if you look at the link I'm posting (completely unrelated topic, but the diagrams are what I'm referencing) and scroll down to the 4th drawing w/ the tube rectifier, this is the PS setup I'm using except I'm not running into two series cap/totem arrangements, just one as I didn't think two banks of series caps was necessary.

        Link: http://www.tubecad.com/2012/05/blog0230.htm

        Comment


        • #19
          I've been doing that since forever (almost 40 years now), to inject some "tubeyness" in my otherwise very stiff and predictable Op Amp based preamps.
          This is a customer demoing an old FAHEY guitar amp from 1989 (but that sound had been developed since '75/'76) .
          Notice very audible sag and compression (yes, I also drive a compressor with voltage drop ... exactly what happens when pentode screens drop) :


          My SS version of a JCM800 . Single channel and not a tube in sight.
          Clipping stage is an overloaded mini power amp, fed from a very saggy separate supply.
          Time constant is critical, about the same as a real tube supply sagging so it's more believable, some smoke and mirrors are involved.
          Real power amp, of course, is based around a couple 2N3055 .
          Today I use TIP142/147 or IRFP250 (might even use a chipamp in a low power version) but in any case they do not change sound, just boost what's fed them.
          Juan Manuel Fahey

          Comment


          • #20
            It's hard (for me) to say why your power amp current is increasing significantly but the voltage only drops a little. I will say that the way the power supply is arranged is very stiff. You basically have totem poled Pi filters. I can only guess that because the preamp is on top of the totem instead of in series with the rail that any capacitive reserve (which is part of their job after all) is able to maintain voltages. Why the small voltage sag in the power amp isn't having any affect on the other operating voltage at all is some kind of "totem Pi filter with CT reference" juju I don't understand. So, since this is the actual question, I'm sorry I couldn't help. But I've been peeking in on this thread anyway so now I'm "subscribed"

            FWIW the signal gated SS regulation is my horse in this race.
            "Take two placebos, works twice as well." Enzo

            "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

            "If you're not interested in opinions and the experience of others, why even start a thread?
            You can't just expect consent." Helmholtz

            Comment


            • #21
              Sag is for pants not amps!

              IMHO

              Click image for larger version

Name:	photo%u00252525202%252520%25252828%252529.JPG
Views:	1
Size:	111.8 KB
ID:	841211

              Peavey Triple XXX with 3x the PS filtering

              Don't tiny 20 uF filter caps and tube rectifiers give a nice sag?

              Comment


              • #22
                Here's an off the wall idea. Put the 400 Ohm sag resistor in the ground leg of the bridge, between the SS diodes and ground. That way both supplies will sag.

                Yes, 20uF will create plenty of sag, although if the current is only going from 70mA to 110mA, it won't sag much. Like I said above, amps designed for Metal players have stiff supplies and lots of filtering.
                WARNING! Musical Instrument amplifiers contain lethal voltages and can retain them even when unplugged. Refer service to qualified personnel.
                REMEMBER: Everybody knows that smokin' ain't allowed in school !

                Comment


                • #23
                  Quiero uno de esos amplificadores de Fahey!

                  Comment


                  • #24
                    I'll try lifting the diodes ground side and putting a resistor there - will be interesting to see what will happen. In a 'typical' PS setup, i.e. tube rectified, ct grounded, I've used a resistor on the CT on other occasions to back bias and drop voltage in that manner.

                    Comment


                    • #25
                      Well, it works pretty much exactly as a resistor on the center tap of a typical fw/ct transformer setup. It develops a negative dc voltage, however it seems to be essentially 6 of one/half dozen of the other. I can put a resistor between the reversed diodes and ground or I can put one between the 6CA4 cathode and first cap, the voltage drops at idle and under load are identical save one is positive and one is negative. Using a 100 ohm resistor, I get about 7V idle to 12V loaded. Not much of a drop, would need to get into the 300-400 ohm range to get any appreciable sag between idle/load, at which point I'm really dropping too much idle voltage. *sigh* What I really wish I could come up with is something like the reverse of an inrush thermistor, something that was dropping very little when lightly loaded and dropping much more when the current increases. Maybe I could put a 240V incandescent bulb in the PT center tap line (it's at @ 240V) or on the center tap of the OT and experiment with different bulb wattages. I know the Samamp guy has some kind of patent with light bulbs but I think he's using them on the power tube cathodes because they're standard 120V bulbs. Hmmm. I wonder if a light bulb would run on negative DC? How about a 100 ohm resistor off the reversed diodes and ground and then run a 12V or 24V bulb between the resistor and ground…. might exponentially increase the resistance and buy me some more sag under loading. I'm probably getting silly here.

                      Comment


                      • #26
                        Also made a little plug in adapter so I could try a 6X4 as opposed to the 6CA4. Ought to get me maybe a few more volts of sag (especially loaded, as 6CA4 is quite stiff)? Nope. Literally no difference, idle or loaded.

                        Aaaaaand, dropped the initial totem down from dual 32/32 stacked to dual 16/16 stacked, again no difference. Also lifted the 30 uf off the screen supply and popped in a 5 uF to see if I could get some sag there (don't want to increase the R value there yet) and of course, no difference at all.

                        I think it just may be an overly stiff PT for this application. Darn thing doesn't even get warm. Maybe I should pop in an extra pair of power tubes, make it a mini Superlead instead, and pull down the supply that way!
                        Last edited by EFK; 03-15-2016, 05:27 PM.

                        Comment


                        • #27
                          Originally posted by EFK View Post
                          Also made a little plug in adapter so I could try a 6X4 as opposed to the 6CA4. Ought to get me maybe a few more volts of sag (especially loaded, as 6CA4 is quite stiff)? Nope. Literally no difference, idle or loaded.

                          Aaaaaand, dropped the initial totem down from dual 32/32 stacked to dual 16/16 stacked, again no difference. Also lifted the 30 uf off the screen supply and popped in a 5 uF to see if I could get some sag there (don't want to increase the R value there yet) and of course, no difference at all.

                          I think it just may be an overly stiff PT for this application. Darn thing doesn't even get warm. Maybe I should pop in an extra pair of power tubes, make it a mini Superlead instead, and pull down the supply that way!
                          I'm "re" working a Champ on the bench right now. It has a PT secondary DCR of 450 ohms! I'm trying to use that to my advantage now. As long as it doesn't get too hot all is golden. But it sags with current like crazy.

                          So I thought... What's your bias voltage relative to output? If you're using a hot bias it's possible that the amp will only sag a little before the tube just has no more to give?..

                          if your HV DCR is low, and the mA rating is ample then a stiff power supply SHOULD result in very little sag. But that still doesn't explain why the PS sags for the power tubes with their node at the center of the totem, but the preamp doesn't get any of this sag. Like I alluded to above, perhaps it has something to do with the fact that there's a filter cap on top of the OT node in the totem arrangement and it's regulating voltage?
                          "Take two placebos, works twice as well." Enzo

                          "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

                          "If you're not interested in opinions and the experience of others, why even start a thread?
                          You can't just expect consent." Helmholtz

                          Comment


                          • #28
                            I wish I could figure out how to change the thread title, because I'd rather it reflect an attempt to cause the whole circuit to sag, not merely the preamp (as someone above questioned, and I believe he was correct). I put an extra big dropper to the preamp on a switch which essentially can toggle the preamp voltage between the early JTM45 voltages and early 70s 50W voltages (which are much lower), and frankly, aside from the bass getting a tiny bit mushier, you'd be very hard pressed to tell a difference because the damn PS is only sagging 8-10V max under load.

                            Chuck, the bias is fixed, set at the old standard 70%. Under full loading, the tubes will run up to about 110% plate dissipation, but they can take it. So there is definitely a swing there.

                            Most of my amps are 1950s types and you very often see a big 10K screen dropper used. So, I thought I'd give screen compression a try and see if - in this particular amp - it could give me some compression to help mask the lack of PS sag. I put it on a dpdt switch (I've got about 8 switches on this amp at this point) so that I could modify the bias ground resistor (off the adjustment pot) at the same time. Well, with a 10K dropper in there (and I tried 15K too), given the low voltage on the power tubes, I can actually run the bias up right to near 100% idling because even under load the current draw is so constricted I think I've basically biased it right into A. If I drop it back, I get some nasty sounding hash as the amp transitions from clean to drive that sounds like crossover distortion, which is why I biased it hotter in conjunction with the big screen dropper in the first place.

                            Next up I'm going to try running a small cap, like maybe 1uF to 5uF, as a bypass to a 15K screen R and see if it creates - as mighty Steve Conner called it in a very old thread - a "ghetto vary-mu compressor" if I can get it "pumping" (as he also termed it).

                            Comment


                            • #29
                              Well I backed off the bias, temporarily wired in a second pot so I could play the amp and adjust by ear to see if I could back off and get some compression back without causing the nasty crap to rear its ugly head. Managed to find a pretty good sound, measured pot and replaced with resistor and it seems to work ok when biased at around 80% (with the big screen R switched in circuit). This is with a 15K (!!!) screen dropping resistor and a 1uF bypass cap, which given the small value may not really be doing anything but I didn't want to "waste" a nice new full-length-lead 5uF electrolytic and I have a big pile of these 1uF/630V plastic caps that I use for a cap coupled reverb circuit.

                              This helps. It's not 100% what I'm after, but it's a help. Plus it knocked the volume back a touch more, and given my whole point has been to get a true Marshall 50W sound at a reduced volume, a bit more of a volume reduction is a good thing. So is having it on a switch.

                              I'd equate the amp, at this point, to a brand new piece of reproduction 18th cent furniture. One of the huge attractions of antique woodwork is the wear, the evident 'softness' of the edges, variations in finish aging etc. That would be the old 60s/70s Marshall. This amp is like a really good copy, but there is no patina, age or wear on it. In other words, the edges are too sharp and precise. It's too 'hard.' I need to come up with creative ways to soften the edges because I can't seem to do it the old fashioned way with a saggy PT, and all those old Marshall PT's are saggy.

                              Chuck I measured this one and it's only 50 ohms from each side of the secondary to the center tap.

                              Comment


                              • #30
                                Still here, still scratching my head. I've been playing with sag resistors again and something is seriously not making sense. I went up to a 450 ohm sag resistor right off the rectifier, which in any other amp would be a BIG mother of a sag resistor, but it only drops 4.6V. WTF? That works out to what, 10 or so mA? The power tubes alone are idling at around 35-36 mA each. How can this be? When I hit the amp with signal, the "sag" climbs to a whopping 8.2V total, so 18.2 mA. Meanwhile the tubes are pulling around 58-59 mA each at that point. It does not make any difference if I put the resistor right off the rectifier, or if I put it between the reversed diodes and ground. Same results either way.

                                What seems even weirder is that, **given the same wall voltage** between tests, with the rectifier straight to the first cap I'm getting around 496-ish vdc at the first cap. With the 450 ohm resistor in place, this drops down to around 482-ish on the rectifier side of the resistor, with the aforementioned 4.6V drop across the resistor. So where the hell is that initial 14V going?

                                Is my math wrong (would not be surprised) or is there something I'm missing here? It's not making sense to me.

                                Is it possible I need to somehow 'balance' things, i.e. maybe two sag resistors, one off the forward rectifier and one off the reversed diodes? And maybe another on the center tap?

                                I've not messed with this type of PS previously. I like what it does because it gives me perfect supply voltages for the preamp vs. power amp, but it doesn't quite seem to "act" the way I would expect.

                                Comment

                                Working...
                                X