iCelcius temperature probe?

[Roger Thoms]

I posed this question over on LFPF and thought I'd ask here too. Has anyone tried the iCelcius temperature probe for the iPad/iPhone? Just wondering if they are any good before I plunk down the money for one.

Roger

 

[Lee L]

Accuracy +/- 1°C, so 20°C could be anywhere from 19°C to 21°C. That's a range of 3.6°F. Not what I'd choose to standardize on for darkroom work if that's what you're thinking. Is the probe waterproof and chemical resistant?

But if you want "stunning graphs" and something more "intuitive" than a standard thermometer interface... :confused:

Lee

 

[RH Designs]

You'd be better off with an ordinary digital thermometer, which would be cheaper and at least as accurate. And if you dropped it in your developer tray it would be less of a disaster! Don't confuse accuracy with consistency by the way, most digital thermometers are only accurate to +/- 1 degree C but if calibrated to a master mercury or similar device will maintain that calibration very well, usually to within +/- 0.2 C or better.

 

[RalphLambrecht]

Richard is right. Thermometer accuracy is not as important as its precision. You can live with a thermometer that is within +/- 1°C of the target value (accuracy), but when measuring a specific medium several times, you want those measurements to be as close together as possible (precision).

 

[Lee L]

Yes, I should have mentioned the accuracy-consistency-precision issue. I don't know if the item for sale is calibratable in software. That would be a nice touch. In the end we don't know how the iCelsius sensor is implemented, what sensor (and specs) it uses, or how good the software is. But they don't claim better accuracy than I mentioned in my earlier post. I'd make my own calibrated version for the same or less money.

With many devices, like the LM34/35/36/37, the grading varies, with the more accurate versions selling for over US$8 and the lower accuracy versions for US$2. I doubt that the iCelsius uses a higher spec sensor at its price. However with these devices a simple offset variable will align them correctly to the scale, and then they are all spec'd to be linear within ±0.5°F across their full scales, from -50° to +300°F in the case of the LM34. Again, I'd suspect that's not done very well, if at all, in a hardware/software package at this price point, and given the accuracy claims.

Richard and Ralph are correct on all points, including the digital thermometer being a better value.

Lee

 

[Roger Thoms]

I appreciate all the good advice from everyone, looks like a dedicated digital thermometer is the way to go. Any suggestions on a good one, there seems to be hundreds of different model available. I have no clue as to who make quality verses junk.

Roger

 

[RH Designs]

looks like a dedicated digital thermometer is the way to go. Any suggestions on a good one

I couldn't possibly comment

 

[Richard Lee]

...looks like a dedicated digital thermometer is the way to go. Any suggestions on a good one..?
Roger

Roger,

If lightning-fast delivery isn't a priority, then I would thoroughly recommend these.

At this price, you could (as I did), purchase a spare or two. Cheap as chips and hugely reliable. Supplied with a battery as well.

Richard

 

[Lee L]

I couldn't possibly comment

...because Richard is an advertiser here, and under the rules can't post his own items for sale in the forums. I'm under no such restrictions, so I can tell you that Richard sells this:

http://www.rhdesigns.co.uk/darkroom/html/thermometer.html

And I expect that means he's found it highly suitable. I don't own one of these, but have something similar available locally in the US. I do use his Stop Clock Pro, which leads me to have great respect for his engineering and design expertise.

Lee

 

[RalphLambrecht]

Roger,

If lightning-fast delivery isn't a priority, then I would thoroughly recommend these.

At this price, you could (as I did), purchase a spare or two. Cheap as chips and hugely reliable. Supplied with a battery as well.

Richard

For that price, there must be something wrong with it.

 

[Richard Lee]

For that price, there must be something wrong with it.

For that price, it must be worth a try. After a year of frequent use, it's still proving to be entirely dependable (and I have another two on stand-by).

Richard

 

[Lee L]

For that price, it must be worth a try. After a year of regular use, it's still proving to be entirely dependable (and I have another two on stand-by).

Richard

Just out of curiosity, do you know how closely do the readings match among the three units across the range from say 10°C to 30°C? ...or in ice water, hot tap water, boiling water? If you've checked that, I'd be curious to see the findings.

Lee

 

[RalphLambrecht]

For that price, it must be worth a try. After a year of frequent use, it's still proving to be entirely dependable (and I have another two on stand-by).

Richard

Richard

My concerns with products offered at such prices are not of technical but rather of ethical nature.

 

[MattKing]

The biggest problem I've had in the past with digital thermometers is preventing them from being damaged by moisture.

I've had the best success with the models that have a probe on the end of a cord. Here is an example of something similar:

http://www.digitalthermometers.net/digitalcookingassistantrwt800.aspx

If I were to work in a shared darkroom, just for fun, I'd like to use this one :

http://www.digitalthermometers.net/blacksharkfloatingpoolandspathermometer.aspx

 

[Scott_Sheppard]

Is you want a really fast and spot on unit, This is for you !! - http://www.thermoworks.com/products/thermapen/splashproof_thermapen.html

 

[M.A.Longmore]

.
I've been using the Comark KM28 for the past year.
It Is Digital Thermometer Perfection !

http://www.katom.com/113-KM28P5.html?CID=become

Ron
.

 

[polyglot]

I actually have three of those $1 thermometers. They work fine (they sit neatly in bottles/jugs of developer while warming in a jobo) and are very easy to clean. The display is a bit small though and I don't trust them to be waterproof.

For the larger display and even though it costs twice as much at $2, I prefer: aquarium thermometers. I plan to mount a row of 5 or 6 of them in a piece of tin and rivet that to the back of my jobo behind an acrylic splash-guard. Calibration is by writing an offset under each one with a sharpie. You do need to be careful not to fracture the cable where it goes into the back of the housing though - one of my older ones (used freehand, especially while cooling dev in the fridge when the tap water is 25C) I've had to resolder a few times.

As to Ralph's ethical objections to the price, this is what they all really cost, it's just in this case no western reseller has had the opportunity to slap a big margin on it. You see the exact same items for $2 on eBay and then $8-$15 from AU resellers and then $25+ in retail stores; buying the $25 version doesn't mean the factory is getting any more than $1. Same model, clearly comes out of the same factory. Personally, I have no issue with cutting two or three layers of middlemen out, in fact I view the opportunity with some glee.

In terms of consistency, I've found a difference of up to 1.2C between separate instances of all of these thermometers (I have about 9 all-up) but the difference is consistent to about 0.1-0.2C across 0C-40C. Zero-order calibration (adding an offset) is easy!

Edit: for colour processing, another good choice is the ear thermometer. Maybe not the cheap ones off eBay, but the good ones are designed to be accurate to 0.1C and have a scale centred on 37C for obvious reasons. They generally don't go down to 20C though, so no good for B&W.

 

[RalphLambrecht]

Is you want a really fast and spot on unit, This is for you !! - http://www.thermoworks.com/products/thermapen/splashproof_thermapen.html

That's a pretty steep price for a thermometer!

 

[RalphLambrecht]

Guys, this is way over the top and too clumsy to use.

This is all you need:

http://www.mindfactory.de/product_i...a-Color-Thermometer.html?ia-pmtrack=157757413

 

[RalphLambrecht]

... As to Ralph's ethical objections to the price, this is what they all really cost, it's just in this case no western reseller has had the opportunity to slap a big margin on it. You see the exact same items for $2 on eBay and then $8-$15 from AU resellers and then $25+ in retail stores; buying the $25 version doesn't mean the factory is getting any more than $1. Same model, clearly comes out of the same factory. Personally, I have no issue with cutting two or three layers of middlemen out, ...

Nowhere in the civilized world can you possibly make a thermometer like this for less than $1. I'm afraid when you peel the layers of middlemen back on this one, you'll see something ugly such as people working 12h for peanuts or worse: child and prison layer. Of course, I have no proof of that, but with prices that low, I'll stay on the safe side and go by: If it sound too good to be true, it most-likely is.

I do not shop in dollar stores for the same reason, by the way.

I'm actually quite frugal and compare prices before I buy, but I never quite understood the hunt for the lowest price at all cost.

 

[polyglot]

I agree that there is a good chance of people working 12 hour days under terrible conditions to assemble these. However, my point is that paying more at the retail end for a brand-name item is absolutely no guarantee that that is not happening unless you are careful of the country of origin. And even if it's "Made in USA" or whatever, that just means it's probably assembled there - the LCDs are probably still printed in Korea and the microcontrollers might come from Taiwan. The US has a big semiconductor industry, Intel for example produces much of the world's flash memory, and they do it for very close to the same price as Taiwan and Korea - semiconductors are cheap because of how they're made in insanely large numbers, not because of slave labour.

What goes into one of these thermometers is actually pretty trivial and semiconductors are perhaps much cheaper than you'd thought. There's a junction (just two bits of metal pressed together) to sense the temperature, a microcontroller (less than 1mm^2, costs on the order of 5c since you get about 30,000 of them from an 8" wafer), an LCD (these are printed out on rolls hundreds of square metres at a time like photographic film, likewise it's worth a few cents), two button cells (about 10c) and a plastic casing (fraction of a cent). Then they get assembled, maybe by hand, and that's a few seconds per item for someone earning a couple dollars per hour. You move that production to the west and the only thing that gets more expensive is the assembly (it's now 30s at $17/hour = 14c).

And that is all. Digital thermometers, while containing slightly interesting software (maybe $10k of IP in total), calibration tables and being the product of advanced semiconductor manufacturing, actually have approximately zero incremental cost once they're being stamped out 100,000 at a time. If you look closely, the cost of production is at most 50c with the biggest single cost probably being the batteries.

Edit: what boggles my mind is how they can ship them for the remaining 50c. There's got to be a subsidy of china post or something strange going on there.

 

[RalphLambrecht]

... However, my point is that paying more at the retail end for a brand-name item is absolutely no guarantee that that is not happening unless you are careful of the country of origin. ...

The next best thing to a guarantee is increasing the odds.