Electronic warmup speedup mod?

[ftransform]

So I am frustrated due to the long warmup time my function generator requires. I'm sure all my other instruments also function better after they warm up but the difference is most apparent in my function generator.

So I had this idea, what if you heat the entire case to like 60C for a minute, and then let it cool off. Would it reach thermal equilibrium faster then slowly heating up?

I was thinking about doing some hairdryer experiments but I figured I would ask first in order to make sure I'm not completely wasting my time.


I have the idea because I usually found that when I am out in the cold I would rather heat myself up too much for a bit and cool down rather then trying to slowly warm up under a blanket. Same result but at least it feels faster to me. Like cold feet in front of a blazing fireplace vs cold feet under a blanket.
**assuming the electronics are alive and have a circulatory system of course!

[ftransform]

bump

[c4757p]

Turn it on ahead of time. I turn all my stuff on when I sit at my workbench even if I don't expect to use it. That way it is warmed up and ready. And just be more patient...

And no, that wouldn't help. Heating it to a constant temperature and keeping it there may, though, hence heated crystals and references. But heating and cooling just wastes time with no benefit - the temperatures still have to equalize when you remove the heat, just in the other direction now.

[Neilm]

Having to wait for electronics to "warm up" is a phenomenon that applies to all electronics, although it is only really noticeable to higher end electronics.  This is the reason that calibration laboratories will leave the equipment turned on 24 hours a day. At work we have a small lab that provides calibration and some of the references in there need to be left turned on for 12-24 hours before they are guaranteed to be in spec.

Neil

[ftransform]

Turn it on ahead of time. I turn all my stuff on when I sit at my workbench even if I don't expect to use it. That way it is warmed up and ready. And just be more patient...

And no, that wouldn't help. Heating it to a constant temperature and keeping it there may, though, hence heated crystals and references. But heating and cooling just wastes time with no benefit - the temperatures still have to equalize when you remove the heat, just in the other direction now.

 I don't think you are correct. Consider this, you are heating a glass of water at 5 degrees C using a lightbulb. You know that eventually the glass is going to reach thermal equilibrium at 30 degrees Celsius (from previous experiments). So you instead pour in water at 35 degrees Celsius, and let it cool down. I believe the water would cool down faster then it would heat up no?


Or, you heat up the water to 28 degrees C using a big heater and let the light bulb do the rest.

Is my understanding of thermodynamics wrong?

One can simply inspect a PCB with a thermal imaging camera after it has been on for several hours, determine what areas are what temperature and litter around preheating lightbulbs or some such in those areas.

If it allows more accurate measurements on seldom used equipment why not? What if you just wanna take a quick measurement thats accurate?

Maybe this is not useful for cal-labs (which are businesses in the business lol) but lower end equipment, like my function generator, which needs to run for 2 hours prior to stabilizing.


WHy not, its no more ridiculous then a soft-on power switch or booting your computer with a solid state drive (to save 30 seconds)
Some of us home users have higher end equipment which we do not want to leave on all the time.

how is it handled in your workplace?

[Neilm]

One can simply inspect a PCB with a thermal imaging camera after it has been on for several hours, determine what areas are what temperature and litter around preheating lightbulbs or some such in those areas.

If it allows more accurate measurements on seldom used equipment why not? What if you just wanna take a quick measurement thats accurate?

Maybe this is not useful for cal-labs (which are businesses in the business lol) but lower end equipment, like my function generator, which needs to run for 2 hours prior to stabilizing.

2 hours for a function generator sounds like you have a problem. I would expect those to be good within seconds. Very different from a volt meter that resolves to 10 nV (from memory).

Neil

[SeanB]

Drift from self heating is very different to drift from external heating, the temperature profile is always wrong. A lot of older precision test equipment has provisions to always power the reference source ( voltage reference, crystal in oven) to be powered in standby, with the rest only being powered when needed. lowered power use, but the precision side was always left with no drift induced in it.

[alm]

2 hours for a function generator sounds like you have a problem. I would expect those to be good within seconds. Very different from a volt meter that resolves to 10 nV (from memory).

Based on what I remember from a previous thread, I believe he expects frequency stability in the order of 10-100 ppm from a basic analog function gen with an RC oscillator. Because of the huge tempco, it might take a long time, if ever, to reach this level of stability.

Drift from self heating is very different to drift from external heating, the temperature profile is always wrong.

I guess you could heat it up in an oven to the mean temperature, then each of the individual components should reach their temperature faster. This would require even heating, not local heating with something like a hot air gun. Not sure if this is worth the effort, though. Just planning ahead and turning it on well before you start working seems like an easier solution to me.