Which to Trust!!

[InADream]

I have a mercury thermometer and a digital thermomter in my tank... The digital reads 80.2 and the mercury seems to read 83... I have tried pulling the digital one out and putting it in colder water and then back to the tank and if eventually goes back to 80.2...

Which one would you trust, I was told that mercury has a t/- factor and is more for general than precise...

Dream

 

[Featherdusters]

You could always do the old boiling water trick. Water boils at 100. Whichever one is the closest is the more accurate one.

 

[InADream]

I am guessing water boils at 100 degrees C not farenheit, I was contiplating all my teachings...
The mercury on goes to 100 degrees F...

Dream

 

[Roy G. Biv]

Well... If the mercury one explodes past 100, then you will have only the digital one left to trust.

 

[InADream]

Thanks Pmolan

Dream

 

[Piscator]

Clean them both.

Put them in your mouth.

98.6 F is normal body temp.

Compare them.

 

[Featherdusters]

<a href=showthread.php?s=&postid=11066075#post11066075 target=_blank>Originally posted</a> by Piscator
Clean them both.

Put them in your mouth.

98.6 F is normal body temp.

Compare them.

Jason's idea sounds better xD Unless you are coming down with the flu.

 

[useskaforevil]

or ice water. since body temperature fluctuates. ice water will always be 32

 

[InADream]

Thanks Piscator thats a great idea....

Dream

 

[plyle02]

50/50 ice bath... That is how we calibrate our thermocouples in the rest. business... 32 degrees would be a fine point for calibration. 50% ice to 50% water btw.....

 

[ihopss]

Use three and to go with the majority.

 

[wrott]

Everyone seems to be basing this on a linear curve of accuracy.
To calibrate any kind of thermal measuring devise, you have to take several measurements along the desired temperature range using a NIST standardized devise and 2 others for back-up averaging.
The best way to get a good thermometer is to look at all of them in the store, and buy one that agrees with most of the rest.

 

[Playa-1]

I think a third thermometer is in order to see which ones needs to go .

 

[mfp1016]

I don't think the digital thermometer would stand up well to the boiling water.. .02

I think you should spend the $35 or $40 and get a pinpoint thermometer.

 

[Playa-1]

Maybe you could take them to a LFS and have them test them against their equipment to fing out which one is the pos.

 

[m2434]

<a href=showthread.php?s=&postid=11066099#post11066099 target=_blank>Originally posted</a> by useskaforevil
or ice water. since body temperature fluctuates. ice water will always be 32

Yes, but you would have to use distilled water, because impurites will change this, so this is probably not true for tap water (ice cubes are also tpicaly made out of tap water). It's somewhat irrelevent anyway as wrott suggested. My question is why do you care about .8 deg? Are you doing a scientific experiment or something? I wouldn't trust mercury in my tank, if it broke somehowe... I'd just go with the digital.

 

[useskaforevil]

no, tap is fine. it would take a lot of impurities to change the melting point of water more than a degree. also he's worried about 2.8 degrees, but i'd geuss its a bigger problem that he doesent know what temperature his tank is at because both could be wrong.


"To calibrate any kind of thermal measuring devise, you have to take several measurements along the desired temperature range using a NIST standardized devise and 2 others for back-up averaging."

lol, yes do that.....

 

[Hypo]

trust no one

 

[DrBegalke]

"98.6 F is normal body temp."

There is a ton of variation in body temperature, between people as well as internal variations... Definitely not a method of thermometer calibration.


However, I doubt it matters much. As long the reading is precise (i.e. reproducible), a reading of +/- a couple of degrees from actual temperature isn't going to matter one bit.