Note to self on digital thermometers

[Andrew Moxom]

Never take for granted what your digital thermometer is telling you..... The last few film developing sessions have been a little frustrating. No matter what film I tried, and developers used, I thought my negs were a little thinner than usual. This used to happen on the odd occasion, but became more regular. Exposure wise, I know I am bang on accurate. So this weekend I noticed that ALL of my 5 rolls that were developed, were thinner than I really like. They are all printable, but not as robust as I normally like them. To that end, I bought a new digital thermometer and I found my original digital thermometer I use was over stating the temperatures by 2 degrees!! That means I'm developing film at 66 deg F!!

As a test, I put the old and new thermometer probes into the hot water kettle while boiling for a test and the old one is consistently showing higher temps than it should. I placed new batteries into the old unit... No change.

So I plan to test in the next week on my film and belive this will cure the thinner negs.

So note to everyone... Never take anything for granted thermometer wise if you are using a digital thermometer. They can and do drift!!!

 

[Thomas Bertilsson]

I'm glad you were able to figure out what went wrong.

 

[df cardwell]

"ever take for granted what your digital thermometer is telling you"

Never take for granted what ANY thermometer is telling you.

 

[mgb74]

Good reminder. Anything that relies on a sensor can drift. Even with a mercury thermometer the mercury column can separate (but fixable).

 

[Worker 11811]

A container full of ice water is often a better standard to calibrate a thermometer with. The freezing point of water is more constant than the boiling point. The boiling point of water can vary depending on altitude. But for extremes of altitude/pressure or the addition of salt or other substances, the freezing point of water does not vary. At least not by an amount that can be measured without laboratory equipment.

I have a digital thermometer which can be recalibrated. Its instructions call for a container of crushed ice and water to be used as a calibration standard.

Quibbling aside... Yes! I agree. You should not take your thermometer for granted. If your thermometer can be calibrated, by all means do so!

 

[Graham06]

Never have just one thermometer

Probably many folk know this 'cos I read it here and in one of the better darkroom books, but I think the advice to never have just one thermometer is quite good. The idea is that all thermometers are likely off a little, and after a while you get used to your own private temperature scale. Inevitably when it breaks, your new one will show different readings.

What one should do is to have more than one and note how they vary. First it will likely tell you if one has absolute errors, and second it will tell you their relative difference which you can use to adjust your readings when one breaks

 

[DWThomas]

I would vote for the ice/water calibration for another reason. One of the errors in digital (or analog!) stuff is non-linearity. The freezing point is much closer to the normal use temperature than the boiling point and likely to better represent the calibration for temperatures of interest. Alas, it never hurts to do a cross check every now and then

 

[Reinhold]

The coefficient of expansion of mercury is based on the unvarying laws of God.

Those impressive LED numbers on that hi-falootin elektronik gizmo are only a read-out from a collection of parts put together by monkeys.

Get a good gla$$ thermometer and use it to confirm all your temperature dependent instruments and processes. Once everything is confirmed, put it away in a safe place as a standard reference. Do not use it routinely, it can be broken...

Reinhold

www.classicBWphoto.com
if that doesn't work, try: http://mysite.verizon.net/res14rg7y/

 

[fschifano]

The coefficient of expansion of mercury is based on the unvarying laws of God.

Those impressive LED numbers on that hi-falootin elektronik gizmo are only a read-out from a collection of parts put together by monkeys.

Get a good gla$$ thermometer and use it to confirm all your temperature dependent instruments and processes. Once everything is confirmed, put it away in a safe place as a standard reference. Do not use it routinely, it can be broken...

Reinhold

www.classicBWphoto.com
if that doesn't work, try: http://mysite.verizon.net/res14rg7y/

...and that's exactly how I deal with the cheap kitchen thermometers. I have one glass analog thermometer of known accuracy from which I calibrate the el-cheapo ones. The dial thermometers I've collected over the years are horrible and useless. They're inaccurate and drift over the course of a single darkroom session. They look impressive, but they're not at all reliable.

 

[Larry Bullis]

The coefficient of expansion of mercury is based on the unvarying laws of God.

Those impressive LED numbers on that hi-falootin elektronik gizmo are only a read-out from a collection of parts put together by monkeys.

Get a good gla$$ thermometer and use it to confirm all your temperature dependent instruments and processes. Once everything is confirmed, put it away in a safe place as a standard reference. Do not use it routinely, it can be broken...

Reinhold

www.classicBWphoto.com
if that doesn't work, try: http://mysite.verizon.net/res14rg7y/

Right!

A long time ago, I worked in a color lab in San Francisco. The boss was the biggest jerk on the planet (he'd suddenly open the door to your printroom to see whether you were working or ?). If you ever worked there, you know who I mean. However, I did learn some amazing things from him. One day, the UPS truck showed up with a box. "Oh, good, I've been waiting for this!"

He opened the box. It was full of mercury/glass lab thermometers. He took one out, put it in his desk drawer, and handed the box to me with the instruction to put it in the stock room. I asked why he put one in the drawer.

"That's the one that's right." He calibrated his processes with it, and checked the others against it.

How many thermometers are there? Quite a few. How many agree?

I have often used the Wing-Lynch water temp controller at 20°C as my standard. Even so, calibration can slip. The are still problems with the ice water standard, if you are going to be picky. Use distilled water. Finding 0°C isn't the whole picture, either. For a full calibration, you'd need the freezing AND boiling points, since the range can be off as well, and for the boiling point, the barometric pressure would have to be accounted. I suspect this would show up more often with digital, but glass/fluid types are manufactured at different quality standards. What if the column varied in the diameter of it's tube?

Now, for my kitchen thermometer, I will test it with boiling water. So the turkey is two degrees too hot?

 

[nickandre]

Say I want to acquire a thermometer which is "accurate" to .1 degrees, not precise to .1 degree and accurate to 1 degree (useless specificity???) Where might I do that?

 

[Kirk Keyes]

Those impressive LED numbers on that hi-falootin elektronik gizmo are only a read-out from a collection of parts put together by monkeys.

We are not monkeys. We are apes.

Side note:
Q: Are we not men?

A: Are are DeVo!

 

[greybeard]

Say I want to acquire a thermometer which is "accurate" to .1 degrees, not precise to .1 degree and accurate to 1 degree (useless specificity???) Where might I do that?

The closest practical approach is probably to acquire a Kodak Process Thermometer. They were made for both black-and-white and color work (the latter has its scale centered at 75F, as opposed to 68F, if I remember correctly) and are about as reliable as you could ever need. The scale is expanded around the nominal processing temperature, and each thermometer was individually engraved so that the capillary bore diameter variations are unimportant.

That having been said, for photographic work the precision is more important than the absolute accuracy, particularly if an etched-stem thermometer is being used as a check on a more convenient dial or digital thermometer. Column separation (which is easily detected) is about the only thing that can change the calibration of an etched-stem thermometer; the cheap aquarium thermometers and their ilk, which have the tube stuck to a pre-printed card, can have their accuracy destroyed by distortion of the card or the mounting.

Realistically, you'd have to be doing pretty fancy sensitometric work to be concerned over even a half-degree variation in process temperature, given the effects of agitation, water quality, film-lot variation, and initial exposure.

 

[fotch]

I have 3 Kodak Process Thermometers, acquired in different years from different sources and they all agree to within a 1/4 degree F. I use one for checking any other thermometers I use. The dial type are the easiest to read but the least accurate in my experience.

 

[greybeard]

The dial type are the easiest to read but the least accurate in my experience.

Yep. Mine took a two- or three-degree jump a few years ago (it's hard to read to closer than that because the pointer is so wide) which was easily detectable in the highlights of Plus-X developed in HC-110. Once I had figured it out, I just work with the pointer in the "right" place for my personal processing method. But I check it against the Kodak thermometer before doing anything really important.

I bought a dial-type thermometer for my son a few years ago, and found that it was off by more than five degrees at a nominal 68F! Needless to say, it went right back to the store, where for all I know it was later sold to someone else...

 

[Ektagraphic]

We are anlog photographers ...I try to keep it pretty analog. My thomometers seem to be pretty good and they are the alcohol ones.

 

[Worker 11811]

There are still problems with the ice water standard, if you are going to be picky. Use distilled water. Finding 0°C isn't the whole picture, either. For a full calibration, you'd need the freezing AND boiling points, since the range can be off as well, and for the boiling point, the barometric pressure would have to be accounted.

Agreed! Use Distilled water if you really want to be right.
And, yes, if you are measuring the boiling point of water in any capacity you need to account for altitude and barometric pressure. There is an on-line calculator for that if you like:

http://www.csgnetwork.com/h2oboilcalc.html

According to this calculator, at my altitude and pressure, water boils at 210.6 ºF. (I am at 730 ft. above sea level and the barometer reads 29.83 in.)

Therefore, I might suggest to OP to retry his experiment and take the altitude and pressure into account.

 

[wclark5179]

I have the round dial thermometer and the recalibration instructions are as follows:

Fill a tall glass with crushed ice.

Immerse stem a minimum of 3 1/2" in the ice.

Wait a minimum of 30 seconds and adjust nut on top of the stem so pointer reads 32 degrees F.

When I first read this I thought they were talking about me when they mention to adjust the nut. I was thinking when all is done, can't ruin all this ice, lets get to the liquor cabinet! Orange juice!

 

[Bijesh]

Almost all thermometers show the wrong temperature. For B&W, it's enough that the thermometer gives the same reading for the same temperature every time. I've several thermometers and these are the readings at the same temperature:

Mercury - 68
Kodak Color - 67
Dial 1 - 70
Dial 2 - 69
Dial 3 - 58 ?
Dial 4 - 68
Red Alcohol - 68
Blue alcohol 1 - 65
Blue alcohol 2 - 69

The mercury thermometer is the standard and it reads 32 when immersed in crushed ice and I use it only to compare with the others.

 

[mikebarger]

Bill, you should have thown all the ice away except two of the cubes. You just have to refill more often if you have all that dang ice in there.

Mike

 

[hrst]

Mercury/alcohol thermometers that have a separate scale usually have a small calibration line printed in the glass itself. It's usually at 20C. I've seen many meters that have the scale severely misaligned but are very accurate after this tiny line is aligned correctly to 20C at the scale. So, if the scale is separate, you can probably move the scale to calibrate it. If the scale is printed to the meter and is misaligned, you have to find the error and do the math every time.

Digital thermometers can be very accurate but they must be calibrated and tend to drift, so calibration must be redone every now and then. The sensor has to be well insulated against water. If not, it will drift severely.

Mercury/alcohol thermometers don't seem to drift at all or very little. If there is a drift, it may have received a physical shock and the mercury/alcohol is in many pieces. To fix it, just put it in a freezer so that all the mercury/alcohol goes to the "ball" where it comes from.

If the meter does not have a calibration system on it's own, you just have to measure its error and add or subtract every time you read it.

Fever thermometers (digital or mercury/alcohol) are usually very accurate but the range is small, however they are very good for color processing (37.8C/100F for C-41 or E6). Not good for BW, though.

 

[NB23]

Many thermometers are inaccurate. My 100$ Kodak Therm is inaccurate... when I compare it to another thermometer. The secret is to stick to ONE thermometer, even if inaccurate, understand it and work from there, usually by adding a minute or two, depending on how thin the negs are so start with. My therm, even if inaccurate, is consistent thru time and that's all I want from it: consistence.

 

[Martin Aislabie]

The coefficient of expansion of mercury is based on the unvarying laws of God.

Those impressive LED numbers on that hi-falootin elektronik gizmo are only a read-out from a collection of parts put together by monkeys.

Get a good gla$$ thermometer and use it to confirm all your temperature dependent instruments and processes. Once everything is confirmed, put it away in a safe place as a standard reference. Do not use it routinely, it can be broken...

Reinhold

www.classicBWphoto.com
if that doesn't work, try: http://mysite.verizon.net/res14rg7y/

Reinhold is absolutely on the money with this

Consistency in B&W is the key

It doesn't actually matter is your reference thermometer is actually accurately reading 20C - what matters in B&W is that you always use the same indicated 20C consistently

We all adjust our film and print processing time away from the manufacturers’ recommended starting point for time/temperature to give us the results we desire.

So choose a good thermometer and calibrate all your others to it.

Then use the temperature equivalent to 20C on your reference thermometer for all your other work.

Keep your master/reference thermometer safe and don't use it for anything except calibrating work- as accidents sadly do all too frequently happen

Martin

 

[Ronald Moravec]

I use dial ones that are off a little, but I know it and compensate.

They are checked against two Kodak process thermometers that hang only and are never used as working thermometers. One I used for 30 years. One was a hand me down from a year ago.

These were $75 @ when new 30 years ago. Treat with care.

Always remember a man with two watches never knows what time it is.

 

[KenR]

I believe that this is the reason that all developing times are suggestions and that you have to find the correct process time for you. If your negatives look consistantly "thin" with you would increase your developing time or aggitation. I found this out a few years ago when my trusted old mercury thermometer broke and had to be replaced with a "non-toxic" one (I assume alcohol based). It's probably calibrated differently from the old one (not available for comparison) but also takes much longer to equilibrate in use. So I had to adjust my processing times to the new "standard." People report the same happens when they move to a new location and the water is different. That's why they say "your milage may vary".