gauging thermometer question

[daveflintstone]

I look in the regs and didn't see anything about using a digital thermometer when gauging spirits.

Is anyone doing this?

I see this one http://www.novatech-usa.com/Products/Glass-SAMA-Thermometers/4353 and it looks good to me.

What thermometer are you using for gauging?

[Paul Tomaszewski]

I got one of the lab-grade glass thermometers from novatech, then got a cheap digital kitchen thermometer and regularly gauge the digital one off the lab-grade one. That way if I end up dropping or breaking the digital one, it's a quick $10 fix.

[will]

Dave,

Nope. That's not even close. (+- 0.5 degrees guarantees the wrong answer).

This is what you need: Certified 30f-124f Mercury

Once you have that (and it's expensive), you can get several of these (and they're not expensive) to use in the lab and do your own "calibration" by using the Certified thermometer to compare...on a regular basis. Same thing, not Certified

The same folks have great hydrometers too.

First rule: No shortcuts.

Will

[JohninWV]

I use the thermometer Will mentioned and the one you linked to get ballpark figures. HB does have good hydrometers, but they've been backordered on some lately.

[will]

John,

Do you compare the mercury thermometer to the electronic thermometer? If so, what kind of correlation do you get? Does it change as a function of room air temperature? (that's the biggest problem with electronic units)

Will

[JohninWV]

I do compare the electric one to the mercury one...really for my own curiosity. The electric one is about .5 degrees off from the mercury thermometer. It changes a great deal depending on how much of the thermometer is immersed (sp?) into the test sample and how much is out in the air.

[will]

John,

I thought so - proves that those little electric things are "indicators" not "instruments."

I have one on the still head - easy to read, and .5 degrees is not important when you're just watching the bread bake.

It's really hard to beat a precision piece of glass with a pure element inside. I bought a fist-full of them, and was very interested to see that the scales were all different - none of them line with each other, indicating that they produce each one separately. They all read the same...to within .05 F or so.

Will

[Charlie]

Thanks guys. Looks like I will be needing a glass one also.

[daveflintstone]

You are not kidding, these are expensive thermometers.

I bought from ICL Calibration Labs (who I highly recommend as very customer service friendly). The thermometer ended up being $243.

They did offer me a calibrated digital thermometer, but I didn't want to spend the $570 at this time.

It is: Dostmann Model 370 with Pt-100 sensor,

Watertight, Min/Max, -100/300°C, 0.1

Resolution, HACCP tested. Supplied with NIST

traceable, ISO/IEC 17025 accredited report of

calibration at 32, 122 & 212°F and service

case.

I also got my hydrometers from them, $201.50 each.

[Pistol Pete]

You are not kidding, these are expensive thermometers.

I bought from ICL Calibration Labs (who I highly recommend as very customer service friendly). The thermometer ended up being $243.

They did offer me a calibrated digital thermometer, but I didn't want to spend the $570 at this time.

It is: Dostmann Model 370 with Pt-100 sensor,

Watertight, Min/Max, -100/300°C, 0.1

Resolution, HACCP tested. Supplied with NIST

traceable, ISO/IEC 17025 accredited report of

calibration at 32, 122 & 212°F and service

case.

I also got my hydrometers from them, $201.50 each.

Hey guys, so what type of thermometer do you recommend if you are not able to ship mercury to your state?

I've talked with reps from novatech and HB and no one seems to have any idea what they are talking about

[will]

HB has Item # 20528, Enviro-Safe® NIST Traceable Calibrated Thermometer. it's -1 to 60C by 0.1C, so it meets the requirements in the CFR, but it's in C, not F. You can do that correction rather easily.

Good luck,

Will

p.s. the NIST just announced they would no longer be providing calibration services for mercury thermometers.

[Pistol Pete]

Thanks Will- So do you suggest getting the calibrated one and then also some less expensive uncalibrated ones with proper divisions and then using the calibrated to calibrate the uncalibrated? (say that 5 times fast )) Or do you suggest just getting the calibrated and using that? I'm a little bit confused on how to proceed. Thanks again.

[will]

we purchased the calibrated mercury thermometer. we source from HB Instrument, and they've deleted mercury from their catalog, but have some stock remaining. red spirit is the replacement, and it's the same accuracy, but available only in Celsius.

we use the certified thermometer to check the performance of the other thermometers.

we write-down the serial numbers of the thermometer and hydrometer used each time we gauge, just in case there's a problem.

we also have calibrated hydrometers for each product proof range.

occasionally, we also mix reference solutions of 200 proof ethanol with distilled water (by weight) and (1) test our instruments on that, and (2) send it out for testing to an accredited lab to see what they say the proof is.

[Pistol Pete]

OK thanks again Will, I appreciate it. Kind of off topic, but how accurate is your scale? I purchased a 1000 lb .5 lb accuracy floor scale to weigh and mix in 55lb drums and I'm wondering if that will not be accurate enough?

[JohninWV]

I've recently done some proofing using 2 different certified hydrometers and our certified thermometer and a digital thermometer. We then sent them to a friend for some proofing on a Anton Paar. The proofing with the digital thermometers was more accurate than our certified thermometers. The change in hydrometer made little change, but even with the correction, the certified thermometer seems to be off.

Will, are you buying your hydrometers directly from HB or another source? I've tried to buy directly, but no luck.

[will]

it's been years since i was close enough to metrology to do this subject justice. I do remember that 0.5 lb resolution and 0.5 lb accuracy are different, and the uncertainty of reading a digital display is always considered +- 1 least significant digit.

let's do an easy one - you're blending down from 190 to 80 proof, and you receive your raw spirit in 55 gallon drums. if you divide that drum in half, when you're down to 80 proof, you'll have 65 gallons - not good - it won't fit. so, knowing that in advance, you make three batches. 55/3 = 18.33 gallons. you'll need 124.56 lbs of spirit, then you'll add water to a total of 345.1 lbs...43.54 gallons.

let's say you blow the weighment of the alcohol in one direction and the water in the opposite direction. so instead of 124.5 lbs ethanol, you dispense only 124, and your total is 345.6lbs. what's the error in proof? you would be about 0.4 proof low. you can measure that with the hydrometer, and make a quick correction, but it will be difficult and time consuming, and you'll want to make the secondary corrections using a desktop scale to weigh small amounts of water and alcohol.

there are small floor scales that are NTEP rated and have 0.1 lb resolution and will weigh to 600 lb - plenty for a 55lb drum. they're around $600. that would reduce the error to under 0.1 proof (under the same assumptions).

making the assumption that resolution and reading are the dominant factor (we're assuming that the scale is accurate and our use of the scale is the cause of all errors), then when you use your big scale to weigh larger quantities, then the error is a smaller percentage. if you're only weighing 10 lbs, then 0.5 lb is a big deal, but not so much if you're weighing 500 lbs.

good luck,

will

[will]

Will, are you buying your hydrometers directly from HB or another source? I've tried to buy directly, but no luck.

years ago in days of old ... in the darkest depths of Mordor ... i was a temperature instrument engineer.

some of the certified electronic units using RTDs are simply amazing. i believe they behave better over large temperature ranges than thermometers. it seems like it's the regularity of the bore size of the thermometer tube that determines how non-linear a glass thermometer will be. for RTDs, the platinum will respond quite predictably from one temperature to another nearby temperature, but they're going to cost more than $45...which is about what we pay for our working thermometers. i guess they'll break less often too. there are several around here - i should dig one out and have a look at it.

yes, we buy direct from HB - we've been doing that for years. maybe their policies have changed and they've grandfathered us. just received about $1000 worth of toys from them today. i love glass - it's so pretty.

don't forget, Anton Parr machines are instruments too. they're subject to calibration and readability errors. they use a very interesting technology.

[Pistol Pete]

it's been years since i was close enough to metrology to do this subject justice. I do remember that 0.5 lb resolution and 0.5 lb accuracy are different, and the uncertainty of reading a digital display is always considered +- 1 least significant digit.

let's do an easy one - you're blending down from 190 to 80 proof, and you receive your raw spirit in 55 gallon drums. if you divide that drum in half, when you're down to 80 proof, you'll have 65 gallons - not good - it won't fit. so, knowing that in advance, you make three batches. 55/3 = 18.33 gallons. you'll need 124.56 lbs of spirit, then you'll add water to a total of 345.1 lbs...43.54 gallons.

let's say you blow the weighment of the alcohol in one direction and the water in the opposite direction. so instead of 124.5 lbs ethanol, you dispense only 124, and your total is 345.6lbs. what's the error in proof? you would be about 0.4 proof low. you can measure that with the hydrometer, and make a quick correction, but it will be difficult and time consuming, and you'll want to make the secondary corrections using a desktop scale to weigh small amounts of water and alcohol.

there are small floor scales that are NTEP rated and have 0.1 lb resolution and will weigh to 600 lb - plenty for a 55lb drum. they're around $600. that would reduce the error to under 0.1 proof (under the same assumptions).

making the assumption that resolution and reading are the dominant factor (we're assuming that the scale is accurate and our use of the scale is the cause of all errors), then when you use your big scale to weigh larger quantities, then the error is a smaller percentage. if you're only weighing 10 lbs, then 0.5 lb is a big deal, but not so much if you're weighing 500 lbs.

good luck,

will

Thanks Will, you're the man

[daveflintstone]

I've recently done some proofing using 2 different certified hydrometers and our certified thermometer and a digital thermometer. We then sent them to a friend for some proofing on a Anton Paar. The proofing with the digital thermometers was more accurate than our certified thermometers. The change in hydrometer made little change, but even with the correction, the certified thermometer seems to be off.

What digital thermometer are you using?

[JohninWV]

Dave, I'm out for the weekend, but I'll let you know on Monday.

[JohninWV]

Dave,

I use these little suckers for quick gauging:

http://www.novatech-usa.com/Products/Digital-Traceable-Thermometers/4052

[daveflintstone]

Dave,

I use these little suckers for quick gauging:

http://www.novatech-usa.com/Products/Digital-Traceable-Thermometers/4052

Thanks, got one on the way.