Topic: Howto: simple Differential Amp + TTL ?

[meandean]

  Glad to help. I'm curious, when you made your measurements, did you try measuring directly across +X & -X as
the scanner would see it (rather than each output to GND)? I would have expected it to give full swing when
measured differentially using the +/-5V power supply.

[Xer0]

didn't checked that and now the old version is overwritten.
maybe. i know the 2nd amp output was stuck constant while the 1st changed
so could be that i swung between them but... thats not whats symmetry opv for

Dean, when i change the top (feedback) Resistor on a opa, i change the gain, right?
And to fine adjust the offset, which solid i have to replace with a poti? the 20k on offset line?

[meandean]

Dean, when i change the top (feedback) Resistor on a opa, i change the gain, right?
And to fine adjust the offset, which solid i have to replace with a poti? the 20k on offset line?

 Yes!

 Make sure to look at the signal between +X & -X when you measure.

[Xer0]

Make sure to look at the signal between +X & -X when you measure.

Makes clean 10V p-p and inverts like it should!

[meandean]

 Great!!

[Xer0]

Ordered all parts today

~30€ electronics, maybe enough to build two Versions of them, SMD and nomal one.

It's gona be Great, yes

[James]

Hey Dean!

Other than the fact that this takes twice as many op-amps, draws twice the quiescent current, and takes up twice as much space, what makes this design better than the one you did for me 6 years ago? 

Just curious.   

James. 

[Xer0]

It's Differential, like the ILDA wants it. Why cheat the offizial specs when you can obey them?

[James]

Simple!

Because that implies that you truly understand what you are doing and not just following someone else's plan.

Converting signals from differential to single ended or singled ended to differential is done all the time in audio electronics.

This is no different.

Differential serves no purpose except if you plan to run very long cables.

Do you have differential inputs on your laser modulation?

The ILDA spec says you do!

James. 

PS the first version of the Pangolin FB3 is single ended.

BTW Dean and I have known each other for about 32 years. We have both been interested in computers and audio electronics since before we ever met.

Although our understanding of things is very different, it was largely developed between the two of us.

We don't work together for shit!

And that turns out to be a good thing.

He programs in VB. I like C/C++.

If he comes up with a schemo, I have to breadboard it and figure out how it works for myself. If I explain how I do something in code, I can not discuss it in C/C++. I have to describe it generically. He has to completely reinvent the whole idea for his way of coding.

If he makes a data file and I try to open it, I see all sorts of things that I would not have seen on my own and vice versa.

I started using sound cards for things other than sound a long time ago. In 1996 I wrote a program that could measure frequency vs. voltage with extreme accuracy over the entire audio spectrum. I still use it! And it doesn't require any additional electronics.

[Xer0]

Do you have differential inputs on your laser modulation?

I have. +2.5 .. -2.5 to be exact. But it's TTL so i screw it and feed the audio directly.
Anyway i choosed a 1:10 gain for X/Y and 1:5 for colors so i could always change it to single ended if i wanted to ...or upgrade to 8ch card by changing the gain to 2:1 on the Vref inverter

[James]

No. I mean what does your laser take as an input signal?

+5VDC and GND.

I realize it can float. That's not the question.

Point is; it's two wires. Not three.

James. 

[Xer0]

Besides, for the colors i use Single End but without opa's first. (>2.5V p-p is on the engineer of Rita said, thats over the SC zeropoint when fed singleended)
I have ordered an TL074 for the case i have to amplify, though.

So i aprreciate your Design too! i just combine the best of all possible worlds, to say so.
I always do that...

[Xer0]

Point is; it's two wires. Not three.

Yes thats right. its maybe even some wires more than SE, cuz you can use common ground on Sing. but have to use - too on Diff.
But i like the up-to-spec distortion-safe design, and i got R2R opas, so if that works with -5V from a ICL7660 how planned i even dont need the big ±12V converter i ordered.

[James]

I would say there are pros and cons of doing this project with differential.

"distortion-safe design"

If you run thin coax like you find in good audio cables to the inputs of your scanners and laser mods, you would have a gounded sheild! You know, the one and only GND; Earth!

In my mind the cons easily outweigh the pros.

cons-----
more complex
more expensive
more consumptive
higher possibility of failure
possibility of componanat miss match

pros----
for X and Y only, cable runs more than 6 to 10 meters


James.  

[meandean]

Hey Dean!

Other than the fact that this takes twice as many op-amps, draws twice the quiescent current, and takes up twice as much space, what makes this design better than the one you did for me 6 years ago? 

 It's real simple- Xero WANTED a differential amp for X & Y, he was also planning to use a +/-5v power supply initially. Yes, SE
is simpler (it's how I do it), but it will not develop enough output for scanners with the lower supply.