Calorimeter

This is bound to get mixed up with things in Electronics, check both. Physics-specific stuff here, mostly.

Calorimeter

Postby Doug Coulter » Sat Jan 02, 2016 10:59 am

With the recent enormous increase in fusor output (be patient, we have to wait for a patent to clear), I realized we had a need for a sensor that was a lot less sensitive than the super nice ones we already have - even the old B10 tube is far too sensitive for power production levels of fusion.

So, taking a hint from the old sarcastic statement "it's real when you can boil that cup of tea with it" - I'm building a calorimeter to add to the data acquisition setup.
With some searching (Thanks Bill) an good old-type glass-vacuum thermos was found to do this with. Newer ones seem to all be stainless steel and far worse at insulating the contents if one believes the reviews on say, Amazon. And here we want to be able to see tiny changes, and not have a lot of bleed from the room ambient conditions, which are never super stable here.

I decided to use a thermocouple to do the measuring, as it's a method that puts in no power to the measuring device, unlike most others, and is nicely stable.
The trick was finding some really thin thermocouple wire, and setting things up so the leak down the wire isn't dominant (obviously, there will be tests of this later, I'm still making it now). While this interface chip: https://www.adafruit.com/datasheets/MAX31855.pdf gives .25 degrees C resolution, in other uses here I've found we can do better with some signal processing, the inherent noise acting as a source of dither for that. We will, after all, want to be able to see some rather small changes during what we hope will be fairly short remote controlled fusor runs, as we don't want to activate the lab too much. If past experience is a guide, a combination of median smoothing to ditch outliers, and plain old averaging - this is a slow thing by nature - will give us a few more bits worth of resolution. The chip has a cold junction compensator that (miraculously!) seems to agree quite well with other temperature sensors, so changing ambient won't kill us instantly, but of course we'll have to test for that. It should be fairly easy to calibrate for leakage from ambient and time constant if that's required.

So, to get the thermocouple in there - I drilled a hole in the plastic top, which was filled with granulated cork, and pushed it through. But the trick here is to make the sneak path long - so I wound about a foot of the TC wire around a mandrel, and inside the void it's a spiral. The void will be filled with "great stuff" when I'm done.
That will hopefully minimize the leakage down the wire itself.

There's a good chance we will borate the water to improve the sensitivity somewhat, but time will tell. That's why I put some permatex (the good blue gasket maker stuff) over the bare thermocouple end - to prevent contamination of the chromel/alumel. We can also attempt to reflect in some neutrons via moderator in places that wouldn't normally intercept the neutrons that should hit this straight on, but those will be thermal or nearly by then - and only the boron will "see" them significantly. I'm hoping that the ~2.2 megavolt neutrons will heat some water simply by kinetics, mainly. It should be fun to test!

Obviously, the low resolution meter I'm using to check this won't do in the real thing, but it's handy for now. Yes, it's hot in here - cold outside, I just cranked up the fire in the room to get warm.
Calorimeter.JPG
A cheapo calorimeter...hope it's good enough for this.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Calorimeter

Postby Jerry » Mon Jan 04, 2016 7:41 pm

Personally, I would use a 3 or 4 wire RTD and a MAX31865 chip, I designed some temp monitoring stuff last year for the company I was working for and they wanted pretty high resolution.
Jerry
 
Posts: 573
Joined: Sun Jul 18, 2010 12:07 am
Location: Beaverton, OR

Re: Calorimeter

Postby Doug Coulter » Sun Jan 10, 2016 10:49 am

I thought about things like that. I could at least guarantee a positive result if I used such a sensor, since the act of measuring pushes in some power that heats the thing I'm trying to measure. That probably doesn't matter for a lot of things, but here it would require an extra calibration step - rate of rise with no fusion, vs with it.

This may take a few tries, dunno, gotta start someplace. An RTD might make a better cold junction compensation measurement for example. I can always go to more a/d bits in measuring the thermocouple if needed.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA


Return to Metrology

Who is online

Users browsing this forum: No registered users and 7 guests