Passive HPF volume loss

[snk]

Hello,
I have a tape echo that I am using on the AUX from a mixer. It works fine, but sometimes I feel that it would sound better if I send less bass content to it : therefore, I am considering building a small HiPass Filter device to remove the excessive mud before reaching the tape echo.
Of course, the obvious way would be to build a stompbox with a simple RC HPF tied to a unity gain buffer (or a baxandall eq, etc)... But as the unit wouldn't be on a pedalboard, and as the device would be placed next to the tape echo (where there is no 9V power cord), I am considering a passive device.

So, my question is the following : as passive devices create some volume loss, is there a way to know how much would my signal get attenuated?
I can dial in the AUX send volume from my mixer, and I can set the input volume from the tape echo, so i can live with volume loss if it is in the -3/-6dB volume loss range...
But I would find a bit awkward to have to boost so much the AUX sends and/or the tape echo input if the passive filter produce a -9/-12dB volume loss (In that case, I may re-consider my needs and might build a unity gain buffer with a bass cut switch)...

[snk]

Oh, and here are a couple more informations :
- At first, I had considered building a Klon buffer (which is obviously active), with a basscut on a switch. So I considered these RC values when "designing" my HiPass filter : 1.5nF cap & 1M resistor for a filter cut around 100Hz (the Klon buffer features 1M resistor and 100nF cap for a 1.6Hz hpf).
https://www.coda-effects.com/p/klon-centaur-circuit-analysis.html
https://www.electrosmash.com/klon-centaur-analysis
https://tagboardeffects.blogspot.com/2011/04/klon-buffer.html

[GibsonGM]

I think it's hard to say exactly how much attenuation you will experience because we can't see what's driving the AUX, nor can we see the tape echo input.

This page gives some info on the basic HPF you're thinking about:  https://www.electronics-tutorials.ws/filter/filter_3.html
There will always be some insertion loss.

It's so simple, I would suggest you simply try it, and see if you can drive it from the AUX and still keep an acceptable signal level.  If not, you could try a buffer...

[snk]

Hello, GibsonGM
Thank you for your answer.
Yes, you're right : it is very simple

From the technical specifications I have read on the mixer and the tape echo manual, this is what I have :
- The mixer AUX out impedance is 75ohm
- The tape echo input is 47kOhm

[GibsonGM]

The tape input seems pretty low.  But give it a try anyway!   I would be inclined to buffer it.   You could use a dual opamp and make the HPF variable...

[Elektrojänis]

The impedances look like passive solution might be realistic. Maybe not ideal, but could be made to work. Nothing in the real world is ideal anyway.

Some things to consider:

Passive circuit can only cut stuff out from the signal. Even with zero losses on tha passband (the part you want to keep) cutting out something will make the average signal level smaller. This holds true especially when you cut bass. It might not be real issue for the actual sound, but on the level meters (if the devices have those) it might look like you lost quite a bit of signal.

You might need to select a frequency for the filter that is a bit higher than you think because simple passive filter is not very steep. The gentle rolloff is probably good for keeping it sounding natural though.

If you look at the schematics on the page linked by GibsonGM, the volume loss on the passband is defined by the ratio of your output impedance of your aux out and the resistor of the filter. If you use 75 ohm resistor, the loss is 6dB. However the input impedance of your tape echo is in parallel with the filter resistor, so if you use 47k resistor in your filterbox, the actual resistor for the filter cutoff calculation is actually 23.5 kohm. Two same value resistances in parallel combines to half of that same value. If you want to calculate other parallel resistances, you can find equations and calculators for that.

It would probably be good to aim for resistance in your filter that's somewhere between the output impedance of your aux output and the input of the tape echo.

[antonis]

Generally speaking, to calculate 1st order passive filter's attenuation, preceding impedance must be more than 10 times lower and succeding one more than 10 times higher..

[ElectricDruid]

Given that the filter will see the 47K input impedance in parallel with the filter's R, I'd be inclined to try and keep the R as large as possible, and adjust the C to get the cutoff where it's needed. In that situation, the "R" in the filter is basically the 47K input impedance.

470K in parallel with 47K gives about 43K. 43K/47n gives about 80Hz, so you could start with 470K/47n and then try reducing the cap value (22n, 10n, 4n7, 2n2 is octaves going up) to see how much bass you need to remove.

The losses in such a thing are pretty much negligible since you're not altering the situation much from the 75R/47K that you've got currently.

[antonis]

Given that the filter will see the 47K input impedance in parallel with the filter's R, I'd be inclined to try and keep the R as large as possible, and adjust the C to get the cutoff where it's needed.

I'd be inclined to completely delete filter's R..

[snk]

Thank you all !

[Elektrojänis]

I'd be inclined to completely delete filter's R..

That's one solution. Given the nature of this, a very custom box for very specific situation, it might be best. It would be the simplest.

If you keep the R, you could make it variable. Say 10 kohm pot in series with 1 kohm resistor and use 0.33 µF cap. That would be variable between about 44 Hz - 480 Hz.

[PRR]

0.01uFd spliced into a patchcord.

(350Hz is common in reverb drive.)

[snk]

Hello,
Sorry for the lack of answer : I just found the time to build the little circuit 
A 33nF capacitor and a 470K resistor did the trick perfectly : it gently cuts the bass, and the volume drop is not too strong, so it should be perfect to clean the mud before the echo layers
Thank you all for the help, and have a nice weekend!

[ElectricDruid]

Glad it worked out for you!

[Processaurus]

0.01uFd spliced into a patchcord.

(350Hz is common in reverb drive.)

And DON'T label the patch cord