Author |
Message |
CHRISKELLY
Joined: Apr 08, 2018 Posts: 103 Location: England
Audio files: 3
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Sun Jan 13, 2019 7:14 am Post subject:
|
|
|
The 4.5 volts isn't really a bias, it's a reference. I know a lot of people do this because it looks simple and less expensive, but it's just a way to make a fake bipolar supply by "cheating". The 4.5 volts is called a "virtual ground". Virtual in that it's not real and ground in that it pretends to be the same kind of ground you'd find in a real dual supply.
If you look at inverting opamp circuits on the web you'll most often see the non-inverting input tied directly to ground. It ought to work with the 100K you have, but try it without, it may improve.
Virtual grounds are tricky to use properly. The limitations are that can't source or sink much current from them. Too many experimenters have tried to connect too much "stuff" to it and it begins to wander, bob and weave (which we call noise). Another issue with this 9volt method is that opamps like TL072 cannot drive the output all the way to the rails. There's a good volt or more that can't be used on both sides of the rail. This means your output voltage headroom is lower. A much better battery operated system with a good ground is to use two 9v batteries in series to make 18 volts. The point where the batteries are connected together is ground - a real ground.
Another problem with these circuits (single 9v batt with 4.5v virtual ground) is too often the signal ground is taken as the negative terminal of the 9v battery (and it appears you are doing this in your circuit). This is wrong. The correct point for signal ground is (in this case) the virtual ground. Using the negative battery terminal as signal ground causes the circuit to operate asymmetrically which can cause distortion (clipping). If you look at opamp circuits on the web, you'll see that when a dual supply is used, the real ground point (the connection between the + and - ends of the supply) is used for signal ground, not the negative supply rail which is what too many 9v single battery circuits do.
With all that said, if you really need to do this with a single battery, then replace the 100K resistor feeding pin 5 with a wire and use that also as signal ground for both input and output. Obviously I'm looking at part of an experiment you're doing and that likely more opamps and circuits are intended to be added. If that is true, then be careful just how much stuff you add. The more things that depend on the virtual ground, the more likely the virtual ground will become a problem. The virtual ground can be "bolstered" with an opamp voltage follower. Essentially, you make a voltage follower (output connected to inverting input) and connect the center point of the two 10K resistors to the non-inverting input. Then use the voltage follower's output as the virtual ground everywhere else in the circuit. The voltage follower amplifier then allows the virtual ground to sink/source more current. This is, however, not a magical solution. It only somewhat extends the usefulness of the virtual ground. It does not create a ground that is as good as a real dual supply.
So my recommendations (which you are fully free to ignore) are:
1) replace the 100K resistor with a wire
2) use an 18 volt supply using two 9 volt batteries or at least think about using a real dual supply - they can be made with an AC wallwart.
3) if a single 9 volt battery is absolutely necessary, make sure that you don't overtax the virtual ground and try the bolster amp if needed.
My primary advice is this: Unless you truly understand virtual grounds (you can google for a lot more info) don't try to design circuits with them. Building known-to-work circuits is one thing, but seeing the virtual ground in them and trying to expand on it is really a bad idea. The best thing you can do for yourself is to build a real dual supply. It won't disappoint you with noisy circuits or circuits that work weird or sometimes not at all. _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
CHRISKELLY
Joined: Apr 08, 2018 Posts: 103 Location: England
Audio files: 3
|
Posted: Sun Jan 13, 2019 7:53 am Post subject:
|
|
|
Hi Jovian
Thanks (again!) for such a detailed reply.
I'll remove the 100k to pin 5 from the Virtual Ground. I'll also try using a buffer for it like U2A is doing in this schematic
http://www.runoffgroove.com/tri-vibe.html
Hopefully this will help. The Virtual Ground is only being used on this one Op Amp so it might work
Thanks
Chris |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Sun Jan 13, 2019 8:15 am Post subject:
|
|
|
That sounds like a good idea. Given your intentions to keep it to just this one opamp, you may not need the opamp that bolsters the virtual ground, but it won't hurt if it's not needed.
Good luck - and above all - have fun! _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
CHRISKELLY
Joined: Apr 08, 2018 Posts: 103 Location: England
Audio files: 3
|
Posted: Sun Jan 13, 2019 9:12 am Post subject:
|
|
|
It worked!!! Thankyou |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Sun Jan 13, 2019 9:48 am Post subject:
|
|
|
Awesome! That means you're having fun! _________________ FPGA, dsPIC and Fatman Synth Stuff
Time flies like a banana. Fruit flies when you're having fun. BTW, Do these genes make my ass look fat? corruptio optimi pessima
|
|
Back to top
|
|
|
|