Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: Wonder2121 on September 16, 2025, 11:31:49 am
-
Obviously need to do more reserch on cascading gain stages and input capacitance..
I have extra tube socket available so thinking about adding gain stage (Marsh type cold clipper) to this circuit,
1. Is the DPDT switch wired correctly
2. Best place. Between 5879/Cf or after tone stack before loop
3. Instead, would it be best/easier to just replace with high gain input jack
Remember viewing schematic on here with similar switchable gain stage but can't find it...But any help appreciated as always...
(http://el34world.com/Forum/index.php?action=dlattach;topic=33016.0;attach=122274;image)
-
Yes, the DPDT is wired correctly. I would highly suggest a potentiometer "inside" your extra gain stage prior to V2-1 (12AX7). Maybe a 250ka? I'd want the potentiometer in the circuit after the DPDT and prior to the V2-1. You can put the potentiometer on the back amp panel if needed and use it as a "set and forget" pot to control the gain going into V2.
With respect, Tubenit
-
If understood, I've replaced the 150k grid resistor into 12ax7 V2-1 with 250kA gain pot (As shown below). This makes sense, it can be dialed to set unity gain with the 5879 when switched right? As you say "set and forget"
(http://el34world.com/Forum/index.php?action=dlattach;topic=33016.0;attach=122278;image)
-
Typically, I'll add a grid resistor of maybe 10k-33k value in a situation like that. Not saying you should, but it seems like a helpful practice to me. Your potentiometer is somewhat serving that purpose anyway.
With respect, Tubenit
-
For the switchable gain circuit to operate, the original schematic required a grid leak resistor for its pin 2 control grid.
The gain pot of the revised circuit resolves that issue.
Just to note that the fx loop signal level and impedance seem far too high (assuming the goal is a transparent, non clipping loop)?
Dunno though, as a master volume after a loop somewhat negates the point of the loop being transparent, ie the LTP phase splitter is after the loop and, unless the master volume is turned up high / full, will be a significant clipping stage.
As the designer, it's your call, how do you want the amp to behave, what's your goal, eg overdrive preamp with transparent fx loop, or an amp that blends in clipping to the fx, as an overdriven power amp would do?
Whatever, it's pretty much necessary to reduce at least the signal level, and preferably its impedance too, on the fx loop send output.
So one of the triodes in your switchable circuit might be better used for a loop recovery stage?
-
I think that two stage switchable gain may be a real squealer. Also, if nothing is plugged into the passive fx loop jacks, the treble pot wiper will be grounded and no signal will be sent to the PI.
-
Just to note that the fx loop signal level and impedance seem far too high (assuming the goal is a transparent, non clipping loop)?
Dunno though, as a master volume after a loop somewhat negates the point of the loop being transparent, ie the LTP phase splitter is after the loop and, unless the master volume is turned up high / full, will be a significant clipping stage.
Good catch. At the least I would expect something like this:
-
... At the least I would expect something like this:
That would seem to disable the tone controls when using the loop?
-
That would seem to disable the tone controls when using the loop?
True, I was trying to keep it super simple. I guess this is the next-best thing; output impedance is kinda high but hey, simple circuits have simple shortcomings.
-
I have two amps at home where I've used a mosfet CF and then "borrowed" a triode in a psuedo kinda sorta active FX loop.
Every reverb (2) and delay (4) and my phaser I've tried in this has worked well enough that I'm happy with it and found nothing I've needed to change. Not saying this "should" be done. Simply sharing it's worked well for me.
Just food for thought.
with respect, Tubenit
-
what's your goal, eg overdrive preamp with transparent fx loop, or an amp that blends in clipping to the fx, as an overdriven power amp would do?
This amps built man, decided replacing it's tube driven effects loop with inline. The tube fx amplified the 5879 hiss when MV was dialed past 12. Couldn't find any solution. TBH I don't hear much difference between inline vs tube driven except for the inlines quieter...So using this spare tube now to add gain stage..
If you notice the split plate res off 5879 I'm dumping a lot of signal, if this helps any... I was also planning on lowering LTPI 1M grid leaks down to 330k to prevent clipping and maybe the tail down to 38k.
Merlin, I will implement your 100k send trim pot as per schematic above..
Also, if nothing is plugged into the passive fx loop jacks, the treble pot wiper will be grounded and no signal will be sent to the PI.
The return jack is switched, signal flows when nothings plugged in..It works at present, maybe my schematic is slightly off...
I have two amps at home where I've used a mosfet CF and then "borrowed" a triode in a psuedo kinda sorta active FX loop.
I'll be totally honest with you, I'm giving up on tube driven fx loops on this amp....Tried so many variations, the problems the 5879, it's a noisy little b*strd and theres no solving it.. :BangHead:
But I will add grid 30k stopper on v2-1 as you suggeested above..
-
what's your goal, eg overdrive preamp with transparent fx loop, or an amp that blends in clipping to the fx, as an overdriven power amp would do?
This amps built man, decided replacing it's tube driven effects loop with inline. The tube fx amplified the 5879 hiss when MV was dialed past 12. Couldn't find any solution.
The idea is to set a loop up to be unity gain. With that, input stage hiss level shouldn't be any different, whether the loop was engaged or not.
It seems to me the issue may have been due to the loop design / implementation / usage.
the problems the 5879, it's a noisy little b*strd and theres no solving it.. :BangHead:
But I will add grid 30k stopper on v2-1 as you suggeested above..
Why did you use a 33k grid stopper on a pentode? Compared to a triode, Miller capacitance is negligible, so it'll be adding hiss with no benefit.