opamp problems

[seanm]

Last night I whumped up the front end of the Voodoo Labs Sparkle Drive. For the purposes of this problem, I only built the input buffer and a simplified first stage of the 4558.

FIXED: Fix shown in red. See  The Tone God's reply below



When I first started it, it would work. But then the output would die off. If I played a note I would get a staticy sound that would quickly die off.

The signal after the input buffer at pin 5 is always good. I tested all the caps but it made no difference. I then replaced the opamp with another new one. Same problem.

I then tried a TL072 and it worked.

Any ideas? My only thought is that the 4558 wants a stronger battery. The battery was at about 8.4 volts.

[The Tone God]

Just by looking at the circuit without the original schematic (which you should post a link to for more assitance) I would venture to say that you need to bias your input signal to 1/2 V+ after the cap (C2). Put a 100K or so resistor from 1/2 V+ to just after the cap (C2) right on the opamp input. Also make a habit of decoupling your output too.

Andrew

[seanm]

Just by looking at the circuit without the original schematic (which you should post a link to for more assitance) I would venture to say that you need to bias your input signal to 1/2 V+ after the cap (C2). Put a 100K or so resistor from 1/2 V+ to just after the cap (C2) right on the opamp input. Also make a habit of decoupling your output too.

Andrew

Thanks I think you are right, I left too much out. The clean blend has a link to Vr. I like to build projects in stages. Usually it works, this time I screwed up!

Interesting that the TL072 worked though..... I think the TL072 has allowed me to get away with a lot of biasing errors.

I used an audio probe to test, so I had an implicit output decoupling cap.

Original schematic here: http://home.hetnet.nl/~chrisdus/download/VLSD.jpg

[davebungo]

What's the reason for using precise resistor values like 49.9K and 4.75K etc.  I've seen this in circuits before which I wouldn't have thought required such accuracy - is there a simple reason for using these?  Are these 0.1% or similar?

[seanm]

What's the reason for using precise resistor values like 49.9K and 4.75K etc.  I've seen this in circuits before which I wouldn't have thought required such accuracy - is there a simple reason for using these?  Are these 0.1% or similar?

No 1% metal. 499k is a standard 1% value, 500k is not.

I used 5% resistors .... I am evil. Plus the 99k resistor used in the "fix" is actually a 100k. Easier to erase the 4 from 499k in the cut and paste

[ninoman123]

Well its probably...around a 100k...Resistor values arent always dead on...so even if you did use a 99k resistor, it prolly wouldnt even be that. Unless you bought some expensive precision resistors.