Newby Electronics Question....

[Andygator3]

Hello All,

I just joined a few minutes ago and have a question that has been bothering me for years and years....

I recently made 4 opamp circuits. Each circuit was on it's own board. Nothing special other than I used some high resistance resistors: 10,000 Meg range, gold plated parts, die cast aluminum enclosures, etc.

When each device was turned on for the first time - it took a while to settle down. I have seen this before but have no explanation other than the parts are "becomming" what they are?! This only happens during the "birth" or burn-in process. After that, the circuit "knows" what it is.

Anybody have an idea what this is?

Andygator3

 

[shrtrnd]

I don't know if I'm answering your particular question, but I'll give it a shot.
When you first turn-on an electronic device, it requires a little time to reach it's normal
operating temperature. It sounds like you were actually looking at the output signal on
a test and measurement device. I believe that would account for the unusual signal
when the device was first energized. Semiconductors can be particularly sensitive to
outputting unusual waveforms until they reach their operating temperature.
It's not much of a difference to the rest of the circuit probably when you're talking about
an electronic device. But if you're looking at the waveform from an op-amp when you
first energize it in a simple stand-alone test circuit, you'd see some unusual waveforms.

 

[Andygator3]

When you first turn-on an electronic device, it requires a little time to reach it's normal operating temperature..

If that were the case, I would see it on each unit every time I turn it on - but the problem goes away after the first turn-on

Semiconductors can be particularly sensitive to
outputting unusual waveforms until they reach their operating temperature.

All of the parts, equipment, etc. are at the same temperature. (within 1 tenth of a degree or so).

I years past, I would see this same behaviour when I turned on high power amplifiers for the first time. But these little circuits have maybe 10-15 parts. The big 1000 watters had hundreds?!

I put the circuit output into a A to D and then into a computer. I think I was measuring around pico-volts.

 

[CocaCola]

I'm guessing that since you mention amps you are experiencing the caps charging up for the first time, it the caps were discharged due to extended off periods or by another means you would like experience the same initial slow start up...

 

[davenn]

How about showing us your circuit so that everyone can stop guessing as to what is going on

Dave

 

[shrtrnd]

Apparently, I'm over-analyzing your issue.
The operating temperature I'm talking about, is the temperature on the semiconductor wafer,
an extremely small area and relatively quick warming period.
The slowest scope I've got is 500Mhz.
It sounds like your observation, is a new one on me.

 

[BobK]

Nothing special other than I used some high resistance resistors: 10,000 Meg range

10000 Meg?

Bob

 

[(*steve*)]

10000 Meg?

OK, I'm guessing there's a typo there somewhere 10 GigaOhms is *huge*.

Andygator3,

Leakage across the board is going to be an issue, and thus humidity, and even things like how you cleaned your board, and the nature of the environment it's in.

High value resistances like this also mean that any capacitances (even parasitic ones) will generate significant RC delays.

Taking your statements at face value, you may be seeing the effects of the time it takes to charge up input capacitances of your op-amp. I would suggest that input currents are also going to be a problem for you.

But let's see the circuit and confirmation of the resistances. Part numbers for your op-amp would be critical too.

 

[Andygator3]

10000 Meg?

Bob

Yea, (10,000 M)

It is not this circuit as much as it is all circuits I have analysed bringing up the first time. The high voltage/high current amps do the same thing. The average time to "settle down" is about 30-45 seconds.

After the initial startup, the hardware works fine forever. I have seen this on very small opamp circuits to 1000+ watt audio amps / power inverter / arc welder power supplies.

For 30 years or so - I just put it down to the physics of the materials being somehow re-organized on the molecular level during burn in.

I do like the idea of the cap being the culprit. I think I might try taking a circuit built from scratch (all new parts), test for the anomaly, change out the caps, and retry for a duplicate anomaly?! Sounds like a great next step!

Andy

 

[BobK]

I can't even imagine where you got a 10000 Megohm resistor, but if you really are using resistors that high, I can see a lot of problems. How is the circuit constructed? There is likely to be leakage that overwhelms a 10000 M resistor all over the place. I would think you would need special materials and an extremely dry environment to make something work that is relying on that kind of resistance.

Secondly, what is the input bias current of the op amp? If you are using resistors this large, that current could turn out to be a huge voltage drop unless the op amp is really a special one with low input current.


Bob

 

[Andygator3]

I can't even imagine where you got a 10000 Megohm resistor

Bob

Good ole Digikey. I used 2 5000 Meg in series.

I would think you would need special materials and an extremely dry environment to make something work that is relying on that kind of resistance.

Bob

That point brings me to my next question - which prompted my joining....

I am looking for a lower conductivity PCB than FR4. Every time I bring it up with a board house I get blank stares. About 35 years ago, one of the electronic mags had an artical on "IC Bricklaying". This is where you glue everything together and solder point to point. That would do away with the conductive FR4. But what a hassle! Anybody have any ideas?

On the circuit settling issue, it happens on all circuits first time out of the gate. And, it only happens with brand new parts. The issue IS NOT circuit specific. A slight exception here might be processors as they are all burned in.

Thanks!

Andy

 

[(*steve*)]

I'd be very keen to see the circuit. these types of resistances are very exotic.

 

[Andygator3]

I'd be very keen to see the circuit. these types of resistances are very exotic.

I will tell you - but try not to laugh!

I am in the process of building a Quantum Non-Demolition (QND) circuit. Actually, I am building a circuit on steroids. I first read about this project in the April edition (around 1988) of Radio Electronics. It came off as an April Fools Day project. I have not been able to forget the artical for a couple of decades. Finally, I got some time.

I am in the process of building a lab/man-cave where I am going to move all my computers, electronic equipment, parts and beer keg into.

The receiver has to be able to "hear" a bear fart at a thousand miles or more. To do this, every aspect has to be ULTRA thought out. For example, my "ground" has to be an Earth ground that goes PAST the water table. The temperature of the system has to be digitally maintained and controlled with a group of peltiers. I will monitor the humidity of the NEMA-4 enclosure with a humidity sensor hooked up to the "monitor" computer (I would probably have to pipe air through another enclosure for sampling).

Anyway, ultimately I would like to have a Master system in Florida with several other systems hooked up around the country. Those systems can be far smaller. Unfortunatly, the A to D runs about a thousand dollars - not including the computer.

Anyway, you can see why I am soooooo concerned about the conductance of glass epoxy PCB material. I can get temperature stable resistors - but at a hundred dollars each (ouch!!!).

It would have been cheaper for me to just pick up fishing or gambleing - but nooooooo. I have to build a gravity wave receiver?!

Anyway, you can see why I was being a little reticent about saying what I was doing.

If the site wants, I could put a bunch of my notes up as I go along. I wouldn't be able to disclose evrything - but you could get the gist of it.

I welcome any thoughts.

Thanks!

Andy

 

[Andygator3]

As an additional note, solder flux is very conductive. All my contract manufacturers wash the boards. I actually used to use a dishwasher for speed. Of course that was with water soluable flux. Now I use soap, water, and acetone.

 

[(*steve*)]

Are you sure it wasn't an april fool's joke?

Can you post the circuit presented in the original article if not your current one?

Is your background in mathematics, particle physics, etc?

 

[Andygator3]

No, it is not an April Fools Day joke, The Authors continued their work on it for at least several more years and put a peice on the interweb: http://amasci.com/freenrg/grav3.html

That paper along with other spots on the web will keep you busy for a bit. The paper, oddly, says "Reproduced without permission from Radio-Electronics Magazine April 1986 by the Trace - June 1, 1991". I read the original artical and wanted to build the project. The information added over the next several years is great.

My background in electronics go's back to around '72 or so. It runs the entire spectrum from assembly, to purchasing, manaagement, ownership, engineering, sales, etc. etc. I currently am doing all of it at one time, but plan to have a good bit of time to play over the next year. This project is part of the bucket list of things I have always wanted to do but did not have the time.

Take a look at the artical! It is interesting. I specifically wanted to see a waveform resulting from a known mass - like the sun or moon, passing overhead. If I can see that on a single 1st generation system, then I will go to the next generation "all out" system. If that produces better results, then I plan a couple of systems seperated by several hundred miles and hooked up to the internet to show continuity of an event as picked up by different ground stations over time (like the moon passing over head.

The culmination of the project, on the short term, would be the collective mass of the alignment actually showing up in the recordings. From there..... ????

 

[Andygator3]

I cant think of too many fun electronics projects that could be as challanging and rewarding as this. It kinda takes everything to the edge - literally!

It is also like pure science, or basic science. There is no customer, no risk, no money - just fun. I have some other ideas that I would like to incorporate - but the project itself will be a handfull for a while.

Andy

 

[(*steve*)]

Having been in the same room as a liquid nitrogen cooled block of aluminium that weighed about a tonne which was used (unsuccessfully as it happens) to detect gravitational waves, I am pretty certain that some op-amps at room temperature are not going to be quite as sensitive.

I would not be at all surprised if this was an elaborate April fool's hoax that has grown and grown...

Please show us the circuit you're using so we can comment on what else may be happening.

If you can't...

Hahaha, you do realise the link you've provided is a site that collects what I can only describe as "free energy bulldust". And even they describe it as rubbish.

I think you've missed the humour.

 

[KJ6EAD]

Ceramic PCB substrate (Alumina Al₂O₃) has a specific resistivity around 1PΩ/cm. That's 5 orders of magnitude higher than your 10GΩ resistor. FR4 is around 5TΩ/cm.

 

[BobK]

I don't see any 10G resistor in the circuit you pointed to. Is that your own innovation?

Bob