Opamp summing amp help.

[scottyd]

I'm sorry its not a pedal, but I guess it could be? I am building a sum amp to allow me to have an little onboard mixer on my bass that will allow me to mix a piezo with magnetic pup. Or possibly buffer a set of low impendence pups and allow individual input settings for each... I found a simple sum amp schematic changed the resistors to variable to allow me to control input levels and output level. When converted to pcb using DIPTRACE which i guess is a language I can better understand, it makes no sense and the circuit just appears WRONG to me. Here's what I got so far, forgive my ignorance this is uncharted territory for me

The schematic, which I found online, copied and changed the resistors. I am sure there are some issues in this.. probably on the variable resistors.


Initial PCB board layout which for the life of me I cant see it being correct. This is what diptrace said that schematic should be.



This is what I think would be correct... It just makes more sense to me.. in the previous one the wipers on the pot wouldn't even be wired correctly?? BUT then again I dont know if there is some opamp voodoo that would make it work. The thin blue lines are Diptrace still trying to show what should be connected, disregard them. ALSO I think there should be a jumper from 3 to 4 on the opamp but dont have it on the pcb just in case its not needed. IF it is it'd be an easy solder.


Any input is greatly appreciated.

[scottyd]

I think I have found at least one problem, used pot symbols instead of variable res.

[PBE6]

It seems odd to allow both the line resistances and the feedback resistance to be variable. Here's an alternative method from the sound.westhost website that I've used without incident:

[scottyd]

The idea was so that the output could have gain control and be ran into an onboard active equalizer and turned down to keep from peaking the input of the eq. From what I read supposedly changing the value of R3 will effect the overall output.

[PBE6]

Increasing Rf will attenuate the output, but the more attenuation you want the bigger Rf will have to be which will add noise. I'd suggest keeping Rf constant and using the individual line pots to set the overall volume, should be simpler and quieter.

[scottyd]

Ok, will do. I'm working with the schematic that you posted to transfer it to a pcb now.

[GibsonGM]

You may want to try it out on breadboard first, to be sure it's what you want??

[scottyd]

I intend to, I want to lay them out on a pcb as well because I want to see how small the circuit will actually be, to also see if that's what I want. If I can get the board small enough I can mount it directly on a pickup. At this point I'm tossing around lots of different ideas. It only took about half an hour to use the schematic PBE6 posted to make this.


Which is a little bulky at 1"x 1" If I can ever figure out what the variables need to be I can get it smaller by using non-variable resistors. This thing will eventually serve a few different purposes for me. 

I ordered all the parts to put it on breadboard, I got a tl702 op amp instead of LM107...

[scottyd]

The original schematic not only did i use the wrong symbols for the variable resistors,  I took the non-used lugs to the negative on the symbol of opamp, which is an input channel (?) instead of to ground..I'm not sure how that would have worked.. Probably just wouldnt! I'm learning   

[scottyd]

Modified version of PEB6' posted schematic, I just used less inputs and put them on a J1 terminal.

[PBE6]

I'm pretty bad at reading PCBs, just wanted to make sure that you adapt the diagram properly to a single supply (battery) system. Here's another schematic using a single supply you can follow:
http://www.muzique.com/schem/mixer.gif

EDIT: just saw your post above, if you're using two batteries you can use the Westhost diagram straight up. You will need coupling caps at the input and output to prevent your DC bias from interfering with the pickups and subsequent circuitry though, check the AMZ diagram I just linked to.

[scottyd]

You rock man, thx for all the help

[GibsonGM]

If you can breadboard it with the '72, you will find TONS of help "on the fly"...I'd hesitate to order any boards til after then.  Once you know it works, it will be simple to post the schematic AND your PCB diagram for checking, and we'll all know what's going on!

Looks like it is coming along though   Takes time to make it right - you are about 3/4 of the way there!

[scottyd]

Thanks Gibson, will do.

Is there any reason why this circuit wouldn't work off a single 9v? I realise 2@18v would be better but I'm just curious. Also what size caps for the input and output I see different values on what would appear to be similar projects

[GibsonGM]

Ok, for that, you should look up "opamp single supply" on Google etc., find a tutorial.   Briefly, as shown, that opamp can amplify signals from zero volts up to near the supply voltage, 9V or whatever.  Some is used to 'run' it, so maybe it will only go to 8V. Fine.  If you were amplifying DC, and mixing them together, you're ok here as long as your signals aren't so hot as to swamp the opamp.

We input an AC signal tho, since that is what we're working with in audio... let's just say 6 volts peak to peak, maybe from another gain stage or something.     That's 0 to 3VAC,  and then 0 to MINUS 3VAC, swinging around the zero point.

What happens when your AC audio signal goes from zero to MINUS 3V??   Your ground on the "+" pin has set the reference for the opamp - it can do the "+" swing, but not the "-". Can't go below that 'ground' point. So it clips off the negative portion of the cycle.  It's a rectifier.  Distortion results.

Adding a bias voltage (usually 4.5V) to that pin (+ input) thru a resistor "tells" the opamp "this is the mid point to work around".   So it can go above AND below that reference point.

Running on 9V is fine, your bias would be 4.5V, and you can swing maybe 3V each way.  If you want more 'headroom' - more amplification, and to work with bigger signals - you up the power supply to maybe 18V.   That gives you near 8V swing in EITHER direction.

Here's one to start: http://www.ti.com/lit/an/sloa030a/sloa030a.pdf

Look up Tube Screamer and other opamp-based pedals to see how this is applied in the real world...same principle for what you want to do...