Toggle between a filter and a buffer, most clever way?

[Vallhagen]

Hi.

As part of a slightly larger circuit, i like to implement a lowpass filter, which i can switch/toggle on/off with a DPDT. And, when off, i want the signal through a buffer (as in "why not use it, there are two OP-amps in a TL072 case..."). Fairly simple task, which i have solved this way (and yes, it works):



Question: Is there anything unneccessary in there, or a more clever way to do the switching? Do i need R5 and R6?

Or put the question this way: I think yes, i need R5 and R6, because its a good thing to keep the input biased even if the stage is not used. Am i correct?

cheers

[DutchMF]

I say: "if it works, and you like it, leave it alone!!!!" Then again, with a 3PDT and a bi-color LED added, you could have a visual indication if the circuit is in Lowpass or Buffer mode, or switch the input of either option to ground. With a 4PDT you could do both!

Paul

[midwayfair]

You can use the same Vb instead of biasing the inputs separately. That would save two parts.

Did you mean C2_2 to go to pin 1 with C2_1, or was it supposed to go to ground to make R2/C1 a second order filter?

Also, 10nF is a REALLY steep bypass. Is this for a bass?

[Vallhagen]

You can use the same Vb instead of biasing the inputs separately. That would save two parts.

Did you mean C2_2 to go to pin 1 with C2_1, or was it supposed to go to ground to make R2/C1 a second order filter?

Also, 10nF is a REALLY steep bypass. Is this for a bass?

Hi guys. Thanx for replying.

Good question about the C2_2 and C2_1. Its really just for convenience. I need C2=20nF but i dont have that. I have 2pcs of 10nF. So yea, they shall be in parallell, the posted schematic is correct.

...if i read you right: I can connect the switch SW1 2 to the R3/R4 junction, and then SW1 1 direct to IC pin 5, SW1 3 to R1, and just leave R5 R6 out? My concern was; If i do that, it will leave pin 5 on the IC floating when not connected. But that maybe is ok?

Or... do i miss something obvious?

...and nah... it aint THAT steep? 

[midwayfair]

...if i read you right: I can connect the switch SW1 2 to the R3/R4 junction, and then SW1 1 direct to IC pin 5, SW1 3 to R1, and just leave R5 R6 out? My concern was; If i do that, it will leave pin 5 on the IC floating when not connected. But that maybe is ok?

Pin 5 is connected to Vb. It's never floating.

R1 is connected to Vb.

Your switch shunts the audio connection at both ends, so it's not like it matters that parts of them are connected together. The same bias voltage is used for all op amps in tons of designs that might switch out some part of the effect.

[Vallhagen]

...if i read you right: I can connect the switch SW1 2 to the R3/R4 junction, and then SW1 1 direct to IC pin 5, SW1 3 to R1, and just leave R5 R6 out? My concern was; If i do that, it will leave pin 5 on the IC floating when not connected. But that maybe is ok?

Pin 5 is connected to Vb. It's never floating.

R1 is connected to Vb.

Your switch shunts the audio connection at both ends, so it's not like it matters that parts of them are connected together. The same bias voltage is used for all op amps in tons of designs that might switch out some part of the effect.

Thanx. Great answer.

But hey, this means that i dont need SW1 at all, right? With "your" (or rather everybodys..?) design, its hardly necessary with a switch that shifts from VB to... VB.

ill be back with a picture, soon

[Vallhagen]

you mean like this, Jon?



... im still learning stuff daily... obviously:)

[midwayfair]

Put the switch back from the input -- just move it after R1 instead of before. The bias is only connected to one pin at a time that way.

[Vallhagen]

rite... rather after the junction R3-R4, and before R1... hold on...

[Vallhagen]

This should be it, then.



...but im not reeeeally convinced this is correct "common sense" when it comes to OP-handling? As it leaves a stage without knowing what he's doing, possibly causing parasite noise?  I had to dive down back into googling, found this page; scroll down to fig 3A: http://www.electronicproducts.com/Analog_Mixed_Signal_ICs/Amplifiers/Properly_terminating_an_unused_op_amp.aspx.

[midwayfair]

This should be it, then.



...but im not reeeeally convinced this is correct "common sense" when it comes to OP-handling? As it leaves a stage without knowing what he's doing, possibly causing parasite noise?  I had to dive down back into googling, found this page; scroll down to fig 3A: http://www.electronicproducts.com/Analog_Mixed_Signal_ICs/Amplifiers/Properly_terminating_an_unused_op_amp.aspx.

The only way around that is a 1M to Vb from pin 5 instead of directly biasing. That's more common, but I assumed there was a reason you weren't using it.

[Vallhagen]

... but I assumed there was a reason you weren't using it.

No specific reason really. I guess im doing everything simultanously; playing, testing, simulating, (re-)investigating, surfing... and hopefully thinking. In this case I just added what i thought was the simplest possible buffer next to the filter.

In fact all pictured schematics above do work.

Thanx for good replies, Jon. It helps, really:)

Cheers

[RobA]

The NJM2121D is pretty much designed for this task. It might be useful here. The spec sheet has examples of how to do this kind of switching with different gain structures in the op amp.

http://semicon.njr.co.jp/eng/PDF/NJM2121_E.pdf

I did a bit of playing with them and they did work well. Mouser carries them, but I don't know of others that do.

[Vallhagen]

Thanx RobA.

I didn't know that such a circuit existed:). I am getting ideas just by lookin at it. Fun.

I just spent a few minutes browsing around though, and as you say, it seems like its only Mouser that carries them.

Cheers

[RobA]

I didn't know they existed either. I was trying to search for info on channel switching so I could figure out a way to do presets. I came across a discussion about circuits using op amps like these in Marshall amps so I did a web search and found these at Mouser. There at least have been others made in the past, but I can't find them. I have seen that these are used to replace the original op amps in some Marshall circuits, so maybe an amp parts place would have them.

I think they'd be useful for doing programmed or MIDI controlled loopers. It's too bad that they don't seem to be readily available.