Problems with opamp biasing

[rackham]

So I get the theory, the problem is applying it sometimes

I'm trying to breadboard this:

http://rubidium.dyndns.org/~magnus/synths/schematics/svf1.pdf

Only problem is biasing the opamps. I subbed the CA3140s for LM741s as the 3140 datasheet says it's pretty much a replacement. Also I've ditched the notch output part along with R9 and R10.

As I'm breadboarding this I've not bothered with the buffered outputs either.

My problem is running it off a single 9v battery, I'm totally stumped as to where the 4.5V needs to go on each opamp to bias it properly?

I'd be really grateful if anyone can figure this out

Cheers

Jon

[mike_a]

It's actually quite simple once you get the idea:
when adapting an opamp design that uses bipolar power to use "stompbox power", you need to substitute V- with NEW-ground and old-ground with bias-voltage.
Where bias voltage (or Vb) is usually a voltage divider between V+ and ground (the NEW ground).

[The Tone God]

Take a look at the "opamp-eration" article at my site. It may help you out.

Andrew

[rackham]

Thanks guys

Andrew, I've had a skim through that article and I think I'm getting there, would the input cap value have any bearing on the resistor values?

I think one of my problems with the schem is that there's no V- shown...

I think I need to go through the article again...

[The Tone God]

would the input cap value have any bearing on the resistor values?

Generally I don't think so. In most situations you can keep the cap value and just rebias.

Andrew

[darron]

give this article a shot. it covers opamp biasing and has pretty pics and everything.

cook your own distortion @ GGG:
http://www.generalguitargadgets.com/richardo/distortion/index.html

[R.G.]

Your problem is that the schematic needs a bipolar power supply, probably +/- 12V or +/-15V, and it's not shown.

To make that schematic work, you need to connect each pin 7 to +12 (or15) and each pin 4 to -12 (or 15).

The bias voltage is shown - it's ground (0V).

[rackham]

Ah, I was concerned that might be the case.

Can I mod it to run off 9v?

[Minion]

Just make a Voltage divider from the +9v to ground useing 2 say 100K resistors and connect the center point of the Voltage divider to the Input of the opamp after the input cap ,now the input will be referance to Half supply...You willa slo need an output cap as there will be a Bias voltage on the output which must be blocked with a Capacitor....

Actually Looking at the schematic these Opamp are in an Inverting config so you should put the Bias Voltage into the Non-inverting input of the Opampsas opposed to connection the Non-inverting input to Ground...This needs to be done with all of the Opamps....

[R.G.]

Yes. Wire all pins 7 to +9V. Wire all pins 4 to 0V. Make a resistor/resistor/capacitor bias voltage with equal resistors - see the article on GEOFEX about designing bias networks, as that's your next question, I think -  and AC couple the inputs with a series cap. The places which are shown as "ground" in the schematic now go to the bias voltage.

[rackham]

Superb, that's encouraging then.

I've spent today reading up on voltage dividers (rather than doing work ) so I'll give it a go and let you know how I get on and if it's any good.

Cheers

Jon.

[rackham]

OK, an update:

Thanks to everyone's help, this thing works

I still need to mess about with the Q control though because it's oscillating like mad at higher settings. It appears that cats don't like oscillating filters because one just bit me.

Not sure on the bandwidth either, but I can alter the cap and resistor values until it's right (probably).

[rackham]

Now I'm getting somewhere.

Just a bit of oscillation at one end of the frequency pot.

Is there a way of setting a minimum resistance for the pot above zero? I measured that if it drops below 1.35k on the top half and .02k on the bottom of the dual pot then I get oscillation. I've tried soldering resistors of similar values to those above across the lugs but that doesn't help. I think I'm missing something Ohms-law-related here?

Otherwise it's a bit good. I've altered the design a bit from the above schematic and it's nearly good to go.

Any suggestions?

[mike_a]

It makes sense that you get oscillations on the minimum settings of the dual pot.

A resistor across the lugs only makes the MAXIMUM resistance smaller.
Just stick a 2k resistor in SERIES with each part of the dual pot.

[rackham]

It makes sense that you get oscillations on the minimum settings of the dual pot.

A resistor across the lugs only makes the MAXIMUM resistance smaller.
Just stick a 2k resistor in SERIES with each part of the dual pot.

Nice one Mike, problem solved.

I now have a fully functioning multi-mode filter.

Just needs a little bit of 'voicing' now and it'll be perfect.

I'll post a finished schem if anyones interested?

Big thanks for everyone's help!

Jon.

[mike_a]

Glad I could be of assistance 

I would definitely like to see the schem of the finished product.
Sound samples would be great too.

[GibsonGM]

Sure, we'd like a look )

[rackham]

Better late than never, here you go:



I'll do some mp3s once I've built it and boxed it this weekend. I'm still doing some some last minute tinkering.

Sound-wise, it's just what I had in my head, decent strength Qs and a good range on it.

Hope it's of use.

Jon.