Intel on mic pre low-cut circuits!

[djmiggymigz]

Hoping that someone can either point me in the right direction via an existing post or help me out here. Currently working on a 312 preamp clone and I'm hoping/wondering if it'd be possible to include a switchable (on/off) low cut filter prior to the opamp in the circuit. I have an audient 8pre with lowcuts on every channel and I love being able to clean up a bit while tracking. Currently laying out the design in KiCad and the 312 circuit is so simple that I have a ton of empty space on the pcb and would love to implement this.

Definitely worth noting that I am certainly a beginner when it comes to circuit design. I'm thinking that the filter would be added before the opamp but after the input transformer? That's more or less just a guess though.

 

[pasarski]

Definitely. If your fine with a 1st order high pass filter you only need three components, including the switch (or two if the preamp already has some kind of a high pass filter at the input). Search "RC filter".

 

[djmiggymigz]

Definitely. If your fine with a 1st order high pass filter you only need three components, including the switch (or two if the preamp already has some kind of a high pass filter at the input). Search "RC filter".

Hmm okay so it looks like that circuit would live directly prior to the OpAmp stage right?

 

[My3gger]

Yes. One example of what you need is here:
https://rockscience.net/capi-vp312-preamp-build-and-mod/

 

[djmiggymigz]

Yes. One example of what you need is here:
https://rockscience.net/capi-vp312-preamp-build-and-mod/

wowwww this is so helpful thank you thank you!!

it's really as simple as inserting a couple of caps into the mix to roll off low frequencies??

 

[My3gger]

You are welcome.

 

[djmiggymigz]

hm so that being the case, if I followed the idea of inserting the caps inbetween the load resistor and the op-amp, I could also probably b uild out a circuit of different caps that are selectable with a rotary switch for different rolloffs yeah? Not sure if the rolloff selection would be worth the work/cost of the rotary switch but, just an idea

 

[djmiggymigz]

I suppose what rockscience did is probably best and easiest tho. just two caps on a switch for 2 different rolloff options. Probably easiest too.

Also, just realized what I think is an error in my design! KiCad clocked the R3 and In+ pin of the 2520 as grounded thus there was no connection between them.

Edit: Went thru and found several errors! Inadvertent success,, thank you!

 

[ruffrecords]

hm so that being the case, if I followed the idea of inserting the caps inbetween the load resistor and the op-amp, I could also probably b uild out a circuit of different caps that are selectable with a rotary switch for different rolloffs yeah? Not sure if the rolloff selection would be worth the work/cost of the rotary switch but, just an idea

A word of warning. Yes, it is as simple as switching in a suitable value capacitor, but this is a high impedance point in the circuit which means two things. First it will easily pick up external interference so you need to keep lead lengths to things like switches as short as possible. Also, the high impedance means any stray parallel capacitance caused by leads going to things like switches could potentially cause an unwanted HF roll off so they need to be kept as short as possible.

Bottom line, the leads to any switch connected to this part of the circuit need to be as short as possible.

Cheers

Ian

 

[Newmarket]

Whilst I appreciate that this implementation may be 'standard' - I would raise the issue of whether the hpf function would be better placed after a buffer stage eg post op amp - such that the tx secondary sees a constant load impedance (notwithstanding the Zobel) and the filter is fed from a low impedance source ?

 

[djmiggymigz]

A word of warning. Yes, it is as simple as switching in a suitable value capacitor, but this is a high impedance point in the circuit which means two things. First it will easily pick up external interference so you need to keep lead lengths to things like switches as short as possible. Also, the high impedance means any stray parallel capacitance caused by leads going to things like switches could potentially cause an unwanted HF roll off so they need to be kept as short as possible.

Bottom line, the leads to any switch connected to this part of the circuit need to be as short as possible.

Cheers

Ian

Hmm okay thanks for the heads up on that. Is there a way to remedy that within the circuit or is it simply a matter of accounting for that when laying out the pcb? That shouldn't be to hard to account for. Is it mainly the leads from pcb->switch that need to be kept short or also the tracks on the pcb where this addition is spliced in?

Whilst I appreciate that this implementation may be 'standard' - I would raise the issue of whether the hpf function would be better placed after a buffer stage eg post op amp - such that the tx secondary sees a constant load impedance (notwithstanding the Zobel) and the filter is fed from a low impedance source ?

This is a question that is beyond me! Hopefully someone else has an answer!

 

[djmiggymigz]

If someone's willing to humor me, I dont quite fully understand the schematic drawn out from Rockscience. I'm attaching 2 schematics to this. One's an exact copy of rockscience's idea and then what I think I want to do. Now for the questions.

So the capacitors are wired in series with eachother. Pin 1 on the switch bypasses the caps and this just recompletes the original circuit leading to the loading resistor. Pin 3 introduces both of the caps into the circuit which when in series, about double the rolloff of the filter. Now what I dont understand is how does pin 2 result in a rolloff? From what I can tell it seems as tho pin 2 doesnt complete a circuit to either of the caps. So Rockscience used an ON-OFF-ON switch. So I'm not really getting how the middle position created a rolloff caused by one of caps.

Here's the link again to what I'm referencing in case more info is needed.

Yes. One example of what you need is here:
https://rockscience.net/capi-vp312-preamp-build-and-mod/

Regardless, I dont think I necessarily need or want to have the multiple rolloff option, rather just one at 80Hz that's either on or off. Would the circuit on the right accomplish this?

 

[djmiggymigz]

Another question while I'm at it: why does this circuit create a filter?? So at it's simplest we're just inserting caps into the circuit. BUT, there's caps all of audio circuits that don't create filters so why does it happen in this case?? Is it specific to where it's at in the circuit (just before a loading resistor)? Not really wrapping my head around why (at least rockscience's circuit) filters the input signal.

 

[My3gger]

That's how RC filter works and creates a problem Newmarket mentioned. Would inserting a cap between this two points solve the problem like it does when used between zobel and load resistor, or create one more?

 

[ruffrecords]

Hmm okay thanks for the heads up on that. Is there a way to remedy that within the circuit or is it simply a matter of accounting for that when laying out the pcb? That shouldn't be to hard to account for. Is it mainly the leads from pcb->switch that need to be kept short or also the tracks on the pcb where this addition is spliced in?

The entirety of the connection from the transformer vias the filter to to the input of the op amp should be as short as possible

This is a question that is beyond me! Hopefully someone else has an answer!

What Newmarket means is that this solution is a bit of a kludge - not really good engineering practice and it could lead to unexpected changes in frequency response. There are better ways to do it but they would probably involve additional active circuits.

It would help f you posted a schematic of exactly what you are planning to build.
Cheers

Ian

 

[djmiggymigz]

Okay noted, will keep everything as short as I can!

I’m away from my computer right now so can’t post what I have right now but I’ve essentially used the Capi 312 schematic as my base while also referencing the original 312 schematic from the waltzing bear archive. The original doesn’t have the load resistor but I stuck with capi’s layout for the input transformer -> op amp section. I was thinking of trying both with the load resistor and without (by shorting where it would be) just to see what difference it makes

 

[ruffrecords]

Another question while I'm at it: why does this circuit create a filter?? So at it's simplest we're just inserting caps into the circuit. BUT, there's caps all of audio circuits that don't create filters so why does it happen in this case?? Is it specific to where it's at in the circuit (just before a loading resistor)? Not really wrapping my head around why (at least rockscience's circuit) filters the input signal.

Actually, all capacitors in series with the audio path create high pass filters. It's just that the frequency at which they start to cut is often very low (0ften less than 1Hz).

Cheers

Ian