Ok, so I want to use my old PC as a signal generator. But I can only get a .1V output from my soundcard (guess the PC speakers were amplified internally or something).

Anyway, I know I'll need to amplify that to get a 1V signal for use with my o'scope. Thinking about an opamp with bipolar supply as sort of a simple 'instrumentation amplifier'.

For grins and ease of construction, I'd like to split a 12VDC wall wart with two 5K R's, and use the junction as ground. Maybe toss in a couple of caps if it's noisy.

Question: will this 'floating ground' be an issue if I then connect my scope probes to it when measuring the output signal? Or do I need to bond this to earth ground to assure everything is at the same potential? Don't wanna blow anything up. Perhaps later I'll do it properly with two regulators, but will this work for now?

Thanks!

 

Anyway, I know I'll need to amplify that to get a 1V signal for use with my o'scope. Thinking about an opamp with bipolar supply as sort of a simple 'instrumentation amplifier'.

Don't use the expression "instrumentation amplifier". This is used for a special opamp configuration that gives very high CMRR (Common Mode Rejection Ratio).

For grins and ease of construction, I'd like to split a 12VDC wall wart with two 5K R's, and use the junction as ground. Maybe toss in a couple of caps if it's noisy.

This is not an option. You must install decoupling capacitors.

Question: will this 'floating ground' be an issue if I then connect my scope probes to it when measuring the output signal? Or do I need to bond this to earth ground to assure everything is at the same potential? Don't wanna blow anything up. Perhaps later I'll do it properly with two regulators, but will this work for now?

This is not a problem. Make sure your 12VDC wall adapter is "floating". Then connect your new "COMMON" to your PC ground.

 

0.1 volt from a sound card sounds very low to me. The primary soundcard output are made to drive 32 ohms headsets. to at least +/- 1 volt. What kind of Os do you use. I would also highly recommend use a dual output power supply. It is quite easy to make a dual power supply from to single output supplies. In the long run this setup will make life much more easy for you.

 

If you want to play with a resistor made fake ground, you have to consider that the resistors are included in the output circuit. It gets irritating to have to think about that for every load. Much better to do split 15's and use a TL071. It's a nice, quiet, well behaved chip, as long as you give it wide supplies. I've had them running for decades in stuff I designed for an audiophool that thinks, "If a house fly lands on an aircraft carrier, it will influence how you hear the sound of the ships engines". He thinks there are no transistors in my (vacuum tube) amplifiers, and he also thinks I'm a God.

That's my 2 cents.

 

Yeah, it does sound low, but I am getting .184 VAC at maximum output. I'm running "Visual Analyser", a scope program that also has a signal gen. function. Card and all settings are maxed; so I do need an amplifier. This is an old 766 from 2000 with an ESS Allegro card...


I have a 1980s Tektronix 2213 I just got from Ebay; cheap, and seems like a good first scope. I do 99.9% audio effects/amp work, so 60MHz should be fine.
It works well, so I don't want to wreck the inputs by connecting a sig. generator wrong...

I will build a bipolar supply when I can put an order together, which will take a few weeks. Just wanted to do some 'math' by finding the values of some unknown inductances. Some may have wah potential.

Seems like it would be good idea to tie the grounds together (and connect the COMMON to it as well)!

 

If you want to play with a resistor made fake ground, you have to consider that the resistors are included in the output circuit. It gets irritating to have to think about that for every load. Much better to do split 15's and use a TL071. It's a nice, quiet, well behaved chip, as long as you give it wide supplies. I've had them running for decades in stuff I designed for an audiophool that thinks, "If a house fly lands on an aircraft carrier, it will influence how you hear the sound of the ships engines". He thinks there are no transistors in my (vacuum tube) amplifiers, and he also thinks I'm a God.

That's my 2 cents.

My friends consider me the same way (ha ha!), but it's clear how little I actually know about pure theory I just want the fake ground to 'test the idea' of calculating a few unknown inductors. I'll do the real deal soon.

Yes, the card IS putting out that voltage, and the old speakers have an amplifier in them. That late 90's technology!

 

Isn't the audio out of a sound card capacitively-coupled. If so, why do you even need split supplies to power the opamp? Wouldn't a simple gain of ten (or so) gain stage with a single supply work?

 

I think the idea is to build something versatile, as if this one, lame, sound card isn't going to be the last device he uses to learn about electronics.

If it is, forget the power supply, he can use this, adjust for about 1 ma ilde current, and get 400+ hours out of a 9V battery.

 

Did you measure the sound card output with a scope, or the Visual Analyser tool? The output from a sound card does not follow any standard, nor does the input. Use a proper scope, or set the output to around 50Hz sine wave. Then use a DMM in AC mode. Remember here that the DMM will show the RMS value of the sine wave. So you have to multiply with 1.41 to get the peak value

 

I think the idea is to build something versatile, as if this one, lame, sound card isn't going to be the last device he uses to learn about electronics.

If it is, forget the power supply, he can use this, adjust for about 1 ma ilde current, and get 400+ hours out of a 9V battery.

Ding ding, correct. Trying to move from "I know how to do this because I saw the schematic" to "I know how to do this because I know what the components are actually doing". To do so I need to actually see them (software is cool, but the real thing is better for how I learn...). Yeah, I could (and probably will) just use an opamp with a reference voltage, same as we do with audio effects, and skip the negative voltage for now.


T06: "Did you measure the sound card output with a scope, or the Visual Analyser tool? The output from a sound card does not follow any standard, nor does the input. Use a proper scope, or set the output to around 50Hz sine wave. Then use a DMM in AC mode. Remember here that the DMM will show the RMS value of the sine wave. So you have to multiply with 1.41 to get the peak value"

Ha ha, thanks for the push...yup... .6V RMS at 60Hz. Being self-taught, I didn't consider the capacitive output coupling of the card with the input impedance of my DMM and how HF would roll off. I was measuring the output at 10 to 20KHz! Funny, if this had been a distortion pedal I'd have given that more thought...that kind of rolloff is GOOD in that case, and usually desirable, depending on how you've treated the signal prior to output.

Anyway, some of what I'm doing requires a 20KHz sine wave at 1V (yes, .707 VAC on the DMM). So I have to build the amplifier anyway. But I will keep it simple and use a TL071 biased to 1/2 supply so I can keep a common ground for all of this.

 

But I will keep it simple and use a TL071 biased to 1/2 supply so I can keep a common ground for all of this.

Why are you so stubborn on this matter This is a bad solution that will get you in trouble rather sooner than later. What you can do if you insist on using a single supply. Is to use some small output single supply audio amplifier. This will give a robust output for the audio frequency range

 

There will be a "common ground", and that is the sleeve on 1/8" sound-card jack, which is internally tied to the PC's chassis and to building earth ground through the green wire in the PC's line cord.