Passive low pass circuit on output transformer secondaries

[Brian Roth]

I sort of gathered that Paul's project intended for the transfo to be intentionally overdriven (if desired), so my scribble was based on that guess. IE, let the transfo live in it's own little world, then deal with level matching and filtering downstream.

While I've never personally messed with the zero-field design, I believe it is intended for a more high fidelity result vs. a "color" design Paul seems to be after. I may be wrong, though!

Bri

 

[ccaudle]

That's probably the most invisible way to transformer isolate

That is the exact opposite of what is wanted; Paul in an earlier post clarified that no isolation is needed, the transformer is being used to generate distortion by driving with high signal levels.

Note to Paul, you may want to put switchable resistors (or a pot) in the primary side, transformer distortion varies with driving impedance so increasing source resistance can give a different color than just hammering with high voltage.

 

[Voyager10]

In @Brian Roth's circuit, the B2 option (with a single resistor) is all you'll need. You can set any attenuation you need through the ratio of the B2 resistor to the feedback resistor. The 'A' components will then set the load on the transformer. (Typically the A resistor might be 1K or lower, and the B2 resistor can be 47K or more, so load and attenuation are essentially independent).

Also, if inverting the phase is going to be a problem, just reverse the connections to one side of the transformer.

 

[Brian Roth]

@Voyager10 It was quite late when I scribbled and posted that <g>. I had already drawn the 3 resistor attenuator when I realized it could be simplified; too lazy to start over....hehehe.

Bri

 

[RoadrunnerOZ]

@Voyager10 It was quite late when I scribbled and posted that <g>. I had already drawn the 3 resistor attenuator when I realized it could be simplified; too lazy to start over....hehehe.

Bri

You must have been tired - the date is last year

 

[Brian Roth]

You must have been tired - the date is last year

HA! Well, I'm still stuck in 2023. I went by my bank today to pay for my safety deposit box and take a cash draw. Damn, I dated that check as 1/11/2023! Typically it will be at least a month at the first of each year for me to get the year correct.... <g>.

Bri

 

[RoadrunnerOZ]

Hopefully there’s no penalty for pre-dating cheques - there is one for post-dating them here

 

[Brian Roth]

My bank for the past decade is a "small" regional company and has great customer service. To correct my "BrainPhart", all I had to do was use my pen to strike through the errant date, correct it, then add my initials. The bank's clerk was the one who suggested that vs. writing an entire new cheque. I don't write many cheques these days.....

Bri

 

[abbey road d enfer]

so quickest way I could think of is to use the Hi Cut circuit found in Vox amps,

This circuit works because it is applied to a high-impedance node, so you need either to scale the pot and cap for adapting them to the low impedance of teh secondary, which I don't recommand, or increase teh impedance by adding a resistor at teh secondary.
Now, this would mean the subsequent stage needs to be high impedance too. Clearly the 2k7/130r/2k7 U-pad is not high impedance. If it's an opamp, it(s quite easy to rearrange it for high impedance AND attenuation simultaneously.

as shown in the top drawing. I ran into the same issue I have when I implement this on various push/pull amps which is the useful range is only last 1/4 of the turn. So I lowered the value of the potentiometer to 2k but then encountered a new issue where the signal was high passed until the full amount of high pass is engaged. I've messed around with a couple more capacitor changes but ended back to 250k and 1uF for this circuit. I used reverse log, log, and linear pots with similar results.

There is always a compromise, because the treble-bleed is never completely off with a potentiometer, so the pot value is chosen so that when at minimum it will minimize the effect, so, unless using a custom taper pot, the result is the really active part of the rotaion is cramed on one direction.
An elegant solution is to use a switched pot that would completely disconnect it in fully CCW position. A separate switch is another possibility.

 

[Potato Cakes]

I tried connecting the secondary (-) leg and the primary (-) leg to ground and I did not like how it sounded. I went back to the original drawing as that is as the client was very pleased with this circuit layout. I think I'm just going to use another op amp to do a console style low pass filter.

Thanks!

Paul

 

[ruffrecords]

I tried connecting the secondary (-) leg and the primary (-) leg to ground and I did not like how it sounded. I went back to the original drawing as that is as the client was very pleased with this circuit layout. I think I'm just going to use another op amp to do a console style low pass filter.

Thanks!

Paul

This probably means there is something wrong with your grounding scheme.

Cheers

Ian

 

[abbey road d enfer]

I tried connecting the secondary (-) leg and the primary (-) leg to ground

Why grounding the primary? You should have left it as original.

 

[Potato Cakes]

I wanted to follow up on this topic's opening question with the solution I found for this application. Using a 2k 3k5 rev log pot (10k rev log with 5k1 resistor parallel) followed by 440nF (two 220nF's in parallel) across the secondaries gave the desired corner frequency as well as the effect being distributed much more evenly over rotation range of the potentiometer. Also, this flattened the high frequency shelf bump that was observed when the transformer was pushed well into saturation. I did try a LPF via an additional op amp which did work, but I found that I needed another buffer to accommodate something else in the circuit. Instead of adding another IC, I revisited the passive route about which I was able to think more clearly after stepping away from this for a couple of months. Very happy with the results and how few parts were needed.


Thanks for everyone's input.


Thanks!

Paul