Frustrated with Thermometers

The friendliest place on the web for anyone with an interest in aquariums or fish keeping!
If you have answers, please help by responding to the unanswered posts.

neilanh

Sliced Bread
Joined
Dec 25, 2006
Messages
8,482
Location
Northern Virginia
I love the little digital thermometers. Hide a probe behind some plants, digital readout on the side of the tank, etc. They're great. But, are they accurate?

I did an experiment last night. 2 different digital thermometers, and a standard mercury style thermometer, all sensing from the same corner of one of my tanks. My temperatures read:
73.9 (digital 1)
75.3 (digital 2)
~77 (regular thermometer)

Which one is correct?

I fully understand that in the big picture of things, it doesn't matter that much. What matters is that the same thermometer I use to determine tank temp is the same thermometer I use to determine the temp of the PWC water. I get that.

I also get that there's some accuracy tolerances on the digital thermometers as well. But this range is just too extreme IMO. I've always held in my head that the mercury thermometer was gold, and biased my digitals to that. But i don't understand why my spread is so high. A degree I can almost buy in to, + or - (which is really a 2 degree accuracy). One of these digitals was a few bucks, one was 20 (the more expensive one is the lowest temperature, furthest away from the mercury). Obviously, there's nothing proving the mercury thermometer is accurate either, though, but irregardless they should all have been closer I thought.

Strange that the digitals don't come with a calibration on them, however.

All of my tanks in my basement are planted community tanks, so I attempt to keep all of them at the same temperature. This eases the PWC process, using my python I drain all the tanks, then set the water temp once and refill all 3 back to back. In my basement I have a 125g, a 29g, and a 20g, and soon to add a 2.5g (although I obviously won't be using the python for that little guy).

Ok, so this was more of a rant than a question, I realize that. Thanks for entertaining me. It's funny the stupid little things that bug us sometimes, isn't it?
 
I have a digital thermometers and they read the same as other digitals. I have only tested a couple with mercury ones. What brand are you using?

Little things do bug us as any inaccuracies can be catastrophic in some instances. If you were treating for ich, those differences would matter a great deal. I completely understand.
 
I recently got a cheap Coralife one. It seems to agree with my the stick on one and the mercury one that I have.

I doubt that any of them are perfect
 
I have the digital BigTemp on my 125 and it agrees with my $1.29 mercury thermometer. The digital heaters are both off from those. Always test against known quantity. To be sure, I bought a german made, lab grade thermometer (mercury) for $20 and it agrees with my little $1.29 plastic thermometer.
 
Heat rises so I put near the bottom to test. I like to move it around to different spots to see if the water/heat circulation is good.
 
"Strange that the digitals don't come with a calibration on them, however."

Its been a long time since I worked with digital thermometer chips... like mid '80's when I had to do a college course, but the ones on the lfs market are probably LM-117 or a deriviative.
There are 89//91 ohm 1/8W resistors in parallel between pin 2 and 3 which sets it up to read either in C or f. The output is NOT linear between -40 and +40 C. It has a slight curve to it that could be trimmed out with a pot.
Output tolerance is supposed to be +/- 1 degree (C or F) plus +/- 1 on the reading. Can be calibrated to .001 degrees but it takes some extra work and $.
They can be more accurate than a mercury thermometer but the normal ones in the lfs aren't always. Best way to pick one out is have a calibrated one on hand that's accurate and test against the others... the whole shelf if need be. There will be a few that are tolerable.

If you have some electronics background and test equipment adding a potentiometer to the output, maybe a feedback resistor and observing the output over the full temperature range it can be tweaked to be as accurate as you want. (Can also be fed to the input of a differential amplifier (LM 612) to make it even more accurate and resistant to outside forces.)
 
Did you give them all 24 hours to level?
I would trust a digital over the mercury. The murcury could be compremized from a bump or drop.

Then if you have a kitchen therm. then heat some water and use this to find a match.
 
Back
Top Bottom