How do I get power up the back of this one

Started by Timjag, May 20, 2017, 05:44:50 PM

Previous topic - Next topic

Timjag

I'm only just got to getting my head around op amps mainly because I want to build a state variable filter in a pedal, but for the life of me I can't get this to work, I'm trying to figure out how to get a 4.5v ref onto the main differential amp to no avail. I really don't want to use two batteries and creating an artificial split voltage of +\- 4.5v just puts dc on the ground.

I'm really just a hack I haven't touched electronic design since I was young. I recently had a severe work accident and am off sick so I decided to kill the time by building some guitar pedals, done all the fuzzes etc but wanted to build a state variable eq pedal - oh ho ho - now I'm scratching me head!!

Anyhow - here's a fairly standard SV schematic, it's almost identical to the ones found in the Soundcraft 8000 series - thing is that uses +\- 17v - I've only got 9volts - fine for the filter op amps - the + is grounded so I can stick 4.5v up there, but the differential amp is causing me consternation. I just can't get it to power up. I've been using tl072s because I have loads, but I understand there are lots of new ones that are better

Any help would be muchos gracias


Timjag

* I should add that I've added a high imprudence inverter buffer at the start to make guitar impedance acceptable

Timjag


EBK

#3
 Autocorrect makes me smile sometimes.  Once, I was trying to text my wife, saying that I've finished cooking the pancakes.  Whatever typo I put into pancakes caused it to autocorrect to pancreas.  :icon_lol:

But, I digress, and I've squandered the time I had for a better answer for now, but I at least have time to blurt out:

TL072 op amps are fine. 
  • SUPPORTER
Technical difficulties.  Please stand by.

EBK

#4
Ok, so to double check what you've done so far:
You've connected pin 4 of each TL072 chip to ground and pin 8 of each chip to +9V, right?
And, every place you see ground on the schematic, you have connected it to 4.5V instead, right?
Your 4.5V is made from a resistive voltage divider?  Did you use any capacitors as well for that?
Finally (for now), did you measure your voltages?

By the way, welcome to the forum!   :icon_smile:
  • SUPPORTER
Technical difficulties.  Please stand by.

Kipper4

What is the intended use of the SVF?
What are you plugging it into?
Do you mean you can't get U1 to power up?
Did you use a dual 100k pot VR2 A and B?

And if you remove U1.
Do you get a bandpass output at U3?
Have you audio probed it?

Voltages are going to tell you more.
What's your volts?
Ma throats as dry as an overcooked kipper.


Smoke me a Kipper. I'll be back for breakfast.

Grey Paper.
http://www.aronnelson.com/DIYFiles/up/

Timjag

Ebk,

Thanks for your kind welcome, :-)

Yes, you can always count on spell check to make you look smart when you're trying to impress new people :-)

Ok so, yes I've put 9 volts to Vcc+ and Grnd to -, 4.5 is derived by a standard resistive diver 2x4.7k plus a 47uf cap to stable things up that  which shows up to U3 and U4 on their ground on the + input. The only place ground touches U1&2 is via a pot at the back of U2 which acts as a bandwidth control the more ground that's allowed. With a +\- supply this obviously isn't a problem - it's when you're trying to add a ref voltage, because U1 works as a differential it doesn't *need* a ground reference but the addition of one alters the Q shape.

With the amps not in there seats voltage shows up exactly where it should on all places, with the amps in it seems to get multiplied into 8volts across all outputs.

Timjag

Hi Kipper!

What is the intended use of the SVF?

kind of a funky Wah like a parapedal but with q and boost/cut to do more distruction the schematic is not a million miles away from the parapedal, but I guess the closest would be the parametric filter pedal bob moon designed for Gibson/maestro

What are you plugging it into?

Geetar amp

Do you mean you can't get U1 to power up?

Yup, just hmm

Did you use a dual 100k pot VR2 A and B?

Sure did VR2 is where all the sweepable goodness should happen :-)!!


And if you remove U1
Do you get a bandpass output at U3?

Nope, that's a good one I'll try it


Have you audio probed it?

Nope waiting my my folks to bring my old scope up
Voltages are going to tell you more.

What's your volts?

See above


Thanks for the help :-)


Kipper4

IMO your SVF is over complicated by the U1 comparator

Take a look at this.

http://www.sabrotone.com/?attachment_id=3192

replace the R15 R16 ldrs with a dual 250~500k pot or whatever dual you have in stock.
You have your Q pot in there too. I went with a fixed Q because there comes a point where Hi Q factors mess with the signal and distort it in a bad way.

Forget the envelope detector C5 to D4 you wont need this if you intend to use a treadle to wobble the wah.

"With the op amps in" at 8volts it's slamming at the + rail.

Try removing the U1 and take an output directly from U3
Is that slamming the rails too?

What op amps are you using?

U1 as a comparator is pitting a dry signal fed through R1 against a wet signal from U3.

An audio probe is a different beast to a scope probe.
The audio probe is intended to trace the signal through a circuit to listen to whats happening to it and identify any potential problem areas.

see here
http://www.diystompboxes.com/pedals/debug.html
Ma throats as dry as an overcooked kipper.


Smoke me a Kipper. I'll be back for breakfast.

Grey Paper.
http://www.aronnelson.com/DIYFiles/up/

EBK

Did you add input and output caps since you are now referencing everything to a positive voltage?
  • SUPPORTER
Technical difficulties.  Please stand by.

anotherjim

You do need to AC couple the input and output.

If this is going to be the main part of the pedal, you will need a higher input impedance.
Add an input buffer stage.

Better op-amps?
TL07x can be noisy in this situation where impedances are under 10k. A cheap but good substitute is the NE5532.
For an input buffe,r the high input impedance needed (1M is good) actually makes the TL07x a better choice. From a dual TL072, you can have the input buffer in one amp and use the other to buffer the 4.5v reference - since the R10 path current can be a little high for a plain resistive divider and could wobble under signal conditions, possibly causing feedback.


Timjag

Thanks for the suggestions guys, firstly yes it's decoupled in and out, secondly in addition to that schematic there is another inverting amp driving the input with a 1 meg resistor.

I don't really want the pedal fixed on boost like a Wah I want to be able to control it via vr1 cut and boost just like the mid sweep of a sound console.

What seems to be happening is that the comparitor/differential U1 doesn't work it doesn't cancel. Has anyone used an op amp in this manner with a 9 volt?, I assume they must have!

Anyway thanks again, any and all wisdoms are greatfully received :-)


Tim

Timjag

I've just discovered the icl7660! Gonna make +\- 9v and see how we go from there

anotherjim

Actually, the input of the filter needs to be direct from an op-amp output - no coupling cap between.
Seems to me (could be wrong) that cut/boost action is in VR1 and both ends of that pot need to be from low-impedance sources to work properly.
Otherwise, there's no reason I know of for lower voltage to mess the filter action up, except increased likelihood of clipping with boost+high Q.

Probably a good idea to add -9v for more headroom.

Input buffer is best as non-inverting unity voltage follower (or with gain if needed). Inverted op-amp with 1MHz input resistor puts all of that resistors noise into the audio path (plus that of a 1M in feedback path?) while non-inverting with +input tied to reference by 1M does not.

Timjag

Thanks Jim,

Yes that all makes perfect sense. I think my issues were impedance, but I'm not happy with the ref voltage, - all a bit messy, as I say I'm only just back into this stuff, and it's fascinating to see what you can do with a battery these day!