Where to insert HPF and LPF.

Are there any disadvantages of inserting a HPF and a LPF as shown below?
Simulation seems to work OK (with about 5.5db/oct.) but I'd like your opinion before I breadboard it.

thanks
w.

Depends what the triangles are supposed to represent, but if they are meant to be inverting op amps then the answer is yes they will cause problems because both of them open the NFB loop at dc.

Cheers

Ian

warpie said:

Are there any disadvantages of inserting a HPF and a LPF as shown below?
Simulation seems to work OK (with about 5.5db/oct.) but I'd like your opinion before I breadboard it.

thanks
w.

Click to expand...

The way your basic info is presented is not understandable. If you have simmed that, you must have some kind of real schemo, mustn't you?
Not the only oddity, but I see your HP and LP filters in a parallel arrangement. I don't see how it would work (in a useful manner).

Schematic updated...

Unless I'm missing something obvious, I can't see any reason why High Pass would affect the circuit.

I'm mainly concerned about the Low Pass.

What would be the best place/point to place LP on this circuit?

thanks again for your help,
w.

The LP would work, somewhat. The HP wouldn't always work, because the LC may play games with it.
You'd better use C84 to control your HP frequency.
But it would be better to use C82 and C81 for the LP.
But I don't understand what you're trying to achieve; the slopes are ultimately 6dB/octave. I think a proper HPF should be of a higher-order (at least 12dB/octave) and variable. As to an LP filter, I can live without any.

I was just thinking of using 1st order filters but I might change it to a higher order (maybe 2nd).
The filters (red caps) in fact will be switchable caps with 6 different Fc.

I'm just trying to insert them to the current circuit but I'm not sure where and how...

abbey road d enfer said:

You'd better use C84 to control your HP frequency.
But it would be better to use C82 and C81 for the LP.

Click to expand...

How can I use C84 for the HP?
Also, doesn't the input impedance affect the LP?

warpie said:

I was just thinking of using 1st order filters but I might change it to a higher order (maybe 2nd).
The filters (red caps) in fact will be switchable caps with 6 different Fc.

I'm just trying to insert them to the current circuit but I'm not sure where and how...

abbey road d enfer said:

You'd better use C84 to control your HP frequency.
But it would be better to use C82 and C81 for the LP.

Click to expand...

How can I use C84 for the HP?

Click to expand...

Just like you've done with your cap marked HP... But as I said, better use C82. C84 is under the influence of the LC circuit too, which may make it unpredictable. OTOH, with C82, the cut-off frequency is given by F=1/[2pi.(R25+R26).C82] 

Also, doesn't the input impedance affect the LP?

Click to expand...

Not much, unless you have very high source impedance like phono cartridge or crystal microphone. The LP frequency is given by F=1/[2pi.(R25.R26/R25+R26).C81]

Your updated schematic will probably be OK as long as the HP and LP frequencies are far enough apart not to interfere with each other.

Cheers

Ian

This is the active part in an EMI TG12412 four band EQ, Nice vintage EQ
Wish I had the inductor values.

abbey road d enfer, newbie question but can I just replace the lytic (C82) with a film cap?

Ian, that shouldn't be a problem. For HP I'll probably go as high as 300Hz and for LP maybe down to 3-4K.

Also, is it possible to make 2nd order filters with such configuration?

You see, I can breadboard it and test it as I said but I'm not sure what should I look for. Frequency response? NSR? THD?
In other words, how can I make sure that the filters don't mesh up with the circuit?

Tekay, the inductor values go down to a couple of mH. Sent you a pm.

warpie said:

abbey road d enfer, newbie question but can I just replace the lytic (C82) with a film cap?

Click to expand...

Yes. If you want to set your LF response there, you need stable capacitors that do not distort under voltage, which pretty much rules out electrolytics.

warpie said:

Ian, that shouldn't be a problem. For HP I'll probably go as high as 300Hz and for LP maybe down to 3-4K.

Click to expand...

That should be OK as they are an order of magnitude apart. By the way I just noticed a problem with your HP. As drawn it is inside the LC feedback loop which you definitely do not want. Better to place it immediately after C84.

Also, is it possible to make 2nd order filters with such configuration?

Click to expand...

The HP you can make 2nd order fairly easily. You current HP is formed by the HP cap and the 47K input impedance of the stage it feeds. The easiest way to make it 2nd order is to add another HP cap of the same value in series with the first and take the junction of the twocaps  to ground via a 47K. This makes it 2nd order but not with coincident poles i.e the turnover frequencies are close but not identical. You see this circuit in many pro mixers. LP is a little harder. I'll give it some thought.

Cheers

Ian

ruffrecords said:

The HP you can make 2nd order fairly easily. You current HP is formed by the HP cap and the 47K input impedance of the stage it feeds. The easiest way to make it 2nd order is to add another HP cap of the same value in series with the first and take the junction of the twocaps  to ground via a 47K.

Click to expand...

I've already tried(simmed) this but it doesn't seem to work...

The problem is we don't know exactly what's in the LC boxes; I suspect ther's at least a pot (or rotary switch). The response of the stage is governed by these external components. Hence my suggestion to work on the very first stage.

Thanks for the schemo. Very interesting. Somewhat reminiscent of the K+H UE1000.
The string of resistors account for 1953 ohms, which in parallels with the 47k res establish the input impedance at a tad under 1900ohms. That would compute to 0.9uF for 100Hz cut-off frequency. But in fact, if some LF boost was applied, that would react with the cap and create unexpected results. So I stand to my recommendation of doing it at the first stage - in addition, the higher input impedance allows the use of much smaller caps (about 25 times).

OK, so by replacing C82 with a 20nF film, Fc is around 200Hz, but is this sufficient for DC blocking?
As far as I understand C82 (2.2uF) is for DC isolation.

Any capacitor is DC blocking as long as its working voltage is higher than the voltage to which it is submitted.
In your case, the voltage is less than 1 volt and the working voltage of any film cap is > 50V, so you don't have to worry.

thanks abbey, I'll try that. 

Do you believe it is possible to incorporate a second order filter to the current circuit?
I think a 2nd order LP will be possible by switching C81 and C83 simultaneously?
But not sure about the HP...

thanks again
Michael

warpie said:

Do you believe it is possible to incorporate a second order filter to the current circuit?
I think a 2nd order LP will be possible by switching C81 and C83 simultaneously?

Click to expand...

That wouldn't work because C83 is inside the closed-loop.
Anyway, I think a LPF doesn't need to be second-order, unless you have very nasty noise, in which case you need a much more powerful tool than a simple LPF. Very often I wish I would have a half-order LPF (3dB/octave).

But not sure about the HP...

Click to expand...

You could switch C97 in the output stage. 100nF there gives you 100Hz turnover frequency.
But pls note that the response would not be maximally flat, as a real purpose-built 2nd-order filter would. Not a big deal but I need to mention that for the purists.

excellent! thank you.

time for some testing..