Signal Conditioning Problem 1

[lespaul]

I developed a circuit for a module that transforms several types of signals into a digitally compatible signal. I tested this circuit on a breadboard in January and it worked perfectly over the required range of amplitudes and frequencies. I had some PCBs fabricated about a month ago with the signal conditioning circuit and a bunch of digital components. The signal conditioning circuit does not work for a variety of signals now. Low amplitude, low frequency sine waves do not generate a clean signal like I was seeing with the breadboard. So I used the circuit on the breadboard again to make sure everything was the same and that circuit no longer worked! I originally tested the circuit in two locations and it worked perfectly in both locations. I tested it in two locations again and it failed in both locations. I'm completely dumbfounded.

 

[Skogsgurra]

Me too. What do you want us to do? Believe that you tested it thoroughly in January? Or tell you that Ohm's and Kirchoff's or someone else's law have changed since then?

Gunnar Englund

http://www.gke.org

 

[IRstuff]

Or that your PCB is an exact replica of your breadboard?

TTFN

 

[itsmoked]

I had problems like this once. It turned out that my circuitry was very sensitive to RF. Local transmitters would feed in signals that could be rectified by the circuitry and cause unexpected results. Turning the board, moving my hand near it, wrapping it in tinfoil, all caused changes in operation. If you ever look at a real world high freq or high freq susceptible desigs you will find a sizable percentage of the design effort goes into shielding and other protection methods.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[lespaul]

Thanks for the help guys, appreciate the courtesy.

 

[BobM3]

Could the original circuit have been just-on-the-edge of an instability? The slight capacitance or resistance change with the newer circuit is setting if off?

 

[itsmoked]

Very good point BobM3!

Been there too..

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[lespaul]

It's possible the circuit was on the edge of stability.
The original circuit I built was placed into a cabinet and left alone. Not a part on the circuit was touched. When I used the circuit a few months later I had problems. Thats what I don't understand.
The circuit is comprised of a differential amplifier being used as a differentiator and a comparator. There are limiting zener diodes before the differential amplifier to prevent large voltages from destroying the components. The next stage has a comparator with a feedback resistor to provide some hysteresis.
Anybody have any particular problems with these types of circuits?

 

[danw2]

You wouldn't have a flakey power supply? Using the same one on prototype as on board models.

Power supplies never cause problems . . . .

Dan

 

[Skogsgurra]

What do the signals look like on a scope screen?

Gunnar Englund

http://www.gke.org

 

[lespaul]

With a larger amplitude (about 500 mV or higher) the output signal is a fairly clean square wave from 0-5V no matter that the input signal. But with smaller amplitudes, the output is still from 0-5V but there are noise spikes all over the place. When the amplitude is decreased beyond a certain point the signal dies completely. The main problem is with a sine wave input because all of the other signals will be of a much larger amplitude that doesnt change.

 

[Skogsgurra]

That sounds like a design thing, plus noise coming in either from your power supply, input path or oscilloscope ground. You do use the ground clip of the probe, I hope? Not using the power cord ground like bad boys do?

Gunnar Englund

http://www.gke.org

 

[nbucska]

lespaut:
Differentiator has very high gain at very high freq.
so you must limit the bandwidth or else it will be VERY
noisy !!! Make sure the BW is not wider than necessary.


Plesae read FAQ240-1032
My WEB: <

http://geocities.com/nbucska/>

 

[itsmoked]

nbucska has a very good point.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Skogsgurra]

Agree. Didn't see that differentiator word. Yes, of course. Do what nbucska says. Absolutely.

Gunnar Englund

http://www.gke.org

 

[logbook]

Fishy. The key thing here is that the breadboard was working then stopped working later for "no reason". Let's not look at the pcb version as it clouds the issue.

Possibilities:
* The "untouched breadboard" was mounted on a chassis plate or put in a case.
* The "untouched breadboard" had its power supply/input leads removed and reconnected differently.
* The "untouched breadboard" was powered from a different power supply and/or signal generator.
* The power supply, signal generator or sciope developed a fault.
* Someone took the earth lead off the scope or connected local earthing on either the signal generator or the power supply.
* An electrolytic capaitor aged and no longer functions correctly.
* The board was "zapped" by an ESD event due to mishandling.

If none of those are true it is probably something else!

 

[itsmoked]

* One of the million lousy bread board connections oxidizided into a rectifier joint.
* The power leads changed oreintation or length.
* There is a new RF emmitter nearby.


Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Skogsgurra]

Superstition.

It is an unsuitable design.

A design that is sensitive to parameters like those mentioned above is bad. We do not help this guy if we give him excuses not to go over the design and make it better.

One thing that would help is if we could see the schematics. There is an FAQ on how to publish drawings/pictures in faq238-1161 (thanks Itsmoked). Use that to show us your diagram.

Gunnar Englund

http://www.gke.org

 

[lespaul]

Here is the schematic.

 

[Skogsgurra]

A suitable capacitor across the zeners to reduce bandwidth wouldn't hurt. We don't know yor requirements, so you will have to decide value yourself. Also, I wonder how the common mode signal ((V+ - V-)/2) is limited. How is the input signal referenced to your circuit's ground? Even if the amplifier is an instrumnt amplifier, it still has common mode limitations.

Gunnar Englund

http://www.gke.org