I don't believe that will improve anything. The most significant reason you see imbalance is that the two ends have different stray capacitance to 'everything else'. Addition of damping resistance will not correct phase imbalance, and is at best only a partial improvement to taming magnitude differences. Trust me, I have tried with many different misbehaving transformers, lot of R-C combinations. If they misbehave, you are pretty much stuck with what you have.
I'm not saying you necessarily need to change anything; it is possible you might have found a good solution for your input transformer. My caution to you was that testing on the bench with tight controls vs in-circuit performance could vary, so you may not be reaching your goal with resistors. But, perhaps so - all you can do is try.
One last thought is to ask yourself why you need mag/phase balance for your circuit. If you are feeding a true differential stage, it really makes no difference what happens between each individual end; only what happens end-end. If you check input vs output differentially, you might be pleased with the result. If you are certain your circuit requires the balance, then it pays to really pursue it. For example, my bifilar interstages were desired so that my push-pull triodes 'see' equal swing on each valve, helping get optimal gain and bias conditions. Sure, it all gets combined eventually in the output transformer, so one could argue the result is still a differential output. But that does not optimize my hard driven output stage.
Sharing your circuit might help determine how important the issue actually is.
Thanks zigzagflux, i know what I'm thinking of for the final circuit, ill have to bump this thread next time with at least the basic idea drawn up. And some screen shots of the scope at 15k.
My use case is really pretty simple... A unit with both XLR antiphase and rca se input jacks switched for one at a time use. The XLR inputs will go direct to the driver stage then to the PP output stage, done. But the rca input will first go through the Transformer to create its antiphase then into the same driver as the XLR uses. I have 4 DACs all of which already have an XLR antiphase out, so I want to avoid a phase Inverter for those jacks when I dont need it. If the rca se jack is used I just want both the ground loop protection of an input Transformer and get the antiphase to boot as a bonus. Everything downstream is the same for either input jack type. Basically phase splitting right at the input or having the input come in already phase split. If I'm making sense.
https://www.jensen-transformers.com/wp-content/uploads/2014/08/as060.pdf
This is a solution. You can use a balanced stepped attenuator as R1 and R2 for a volume control as well. It accepts balanced and unbalanced inputs.
With an unbalanced input, the grids of the first stage will see ~(0.4 x Vinput) given the voltage divider in the secondary and transformer losses. The JT-11P4-1 (R1=R2=10K) has a 1:1.414 step up ratio, so you'll get ~(0.6 x Vinput) to drive the grids of the first stage. The JT-11P4-1 is working well in my push-pull 801A-45-1605 amp.
This is a solution. You can use a balanced stepped attenuator as R1 and R2 for a volume control as well. It accepts balanced and unbalanced inputs.
With an unbalanced input, the grids of the first stage will see ~(0.4 x Vinput) given the voltage divider in the secondary and transformer losses. The JT-11P4-1 (R1=R2=10K) has a 1:1.414 step up ratio, so you'll get ~(0.6 x Vinput) to drive the grids of the first stage. The JT-11P4-1 is working well in my push-pull 801A-45-1605 amp.
So your input stage looks like this?
Sounds like we are doing a very similar thing with the RCA and XLR inputs.
Next question is how your PP stage is actually built.
Sounds like we are doing a very similar thing with the RCA and XLR inputs.
Next question is how your PP stage is actually built.
Not exactly but close, yours really looks high end! Mine would be low budget. I'd eliminate the 46 tube stage and the monolith IT-02 stage. And I'd make the XLR and RCA just be part of a 3 position source selector so you can only use one at a time. Basically I'm thinking 3 stages:
1) RCA input transformer (or XLR) stage which is all passive.
2) Then a 6SN7 (or 6SL7) stage depending on how much gain I'll need (experimentally determined).
3) Then the PP power stage (looking like EL84 at this point based on what I have on hand).
I havent figured out if or where I'd put a volume control yet! As everything is antiphase end to end!
In these modern times we use a DAC in our living room system, our office desk, our bedroom and our basement rec room. So I dont need high gain inputs like in the olden days and if the signal is already antiphase XLR why include a cathodyne splitter? Also I like a quality transformer on the RCA because if you've ever tried plugging an output from your receiver into another amp with a TV and DVD etc. all hooked up too... The hum is impossible to not have. So basically with an inoput transformer I'm hoping to get 3 bonus features 1) Eliminate the hum if I want to take a source from my AV system 2) Eliminate the input capacitor 3) Eliminate the need for a cathodyne or other tube topology to do a phase split by doing it passively.
This is exactly what I'm doing. I posted this same schematic earlier in the thread! Good to see I'm not crazy thinking this! This the same Jensen unit I tested above. But I think you solved my volume control issue. Tubecad (John Broskie) sells a balanced attenuator board with volume and balance trimming at a very reasonable price.
Balanced Strereo Attenuator
Last edited:
The reason for avoiding selector switches was to provide a high impedance (and balanced) impedance network for the XLR to see. The input transformer secondary acts as both a high AC impedance balanced burden for the XLR source, and also acts as a low DC impedance grid leak for the SN7 stage. Best of both worlds IMO.
My volume control is within the fully differential preamp. It uses the resistor method for setting bias equal about circuit common and applies volume control to the middle of those resistors. Since the preamp then gets tossed into an output transformer, the resistor method works well. In a three stage amp like mine I would not pursue the resistor method. In a two stage amp like yours, you could perhaps be successful. Just watch what is happening at the EL84 grids to see how well it works.
My volume control is within the fully differential preamp. It uses the resistor method for setting bias equal about circuit common and applies volume control to the middle of those resistors. Since the preamp then gets tossed into an output transformer, the resistor method works well. In a three stage amp like mine I would not pursue the resistor method. In a two stage amp like yours, you could perhaps be successful. Just watch what is happening at the EL84 grids to see how well it works.
Come to think of it, if you haven't figured out your volume control yet then I would pursue the attached. Since you are determined to use an input transformer to address the RCA challenge (good choice) you are fully differential after. You don't need balanced attenuators - too much hardware. All you need is a single one in the right location. The 50k nodes referenced in the schematic is my single potentiometer (actually a Goldpoint).
This would be combined all into one right after your xfmr. If you need help drawing it out I can, but hopefully you get the idea.
XLR-balanced resistors-series resistors-pot-grids. Get it? The grids become a fully differential stage.
If you are truly differential, you need equal impedance for a balanced line, not equal signal.
This would be combined all into one right after your xfmr. If you need help drawing it out I can, but hopefully you get the idea.
XLR-balanced resistors-series resistors-pot-grids. Get it? The grids become a fully differential stage.
If you are truly differential, you need equal impedance for a balanced line, not equal signal.
I see, just shunt the whole thing PP? I've seen this setup across CT secondaries before as the volume control. Like this example too? (from another great diyaudio member site!):
A "Bad Hombre" is definitely on my to-do list, it looks like a fun little project!
El Bad Hombre “Balanced” Headphone Amplifier – wauwatosa tube factory
Last edited: