Infra-red Thermometer

[Rod]

Over the years there has been much discussion here of the various IR thermometers available - usually related to balancing CH systems and similar.

I now wish to acquire one that is capable of making fairly accurate measurements of skin - at around 37C and slightly over. While absolute accuracy would be good, it is not as important as a) repeatability and b) sensitivity to small changes in temperature. Sure I could go off and find some expensive medical instrument or other but can I get away with a cheapie?

Seems to me at least some of the posters here will have played around with their thermometers and may have the experience to suggest a model?

(In case you are desperate to know why such measurements might be wanted, you could have a look here . You might need to sign up to a free registration process to see this - I do not know.)

[Steve Walker]

I would have thought the medical thermometers available from Boots and other places would be the sort of thing you need - the ones I am thinking of are infra-red type and are meant to be used in the ear, but they could probably be used elsewhere with a simple conical shroud to exclude outside sources.

SteveW

[Gordon Henderson]

I don't think one I have (doesn't have a model name/number on it, but I got it from Maplin), while reading to the nearest 1/10th of a C is good enough for skin.

Or maybe it is, but my skin always appears cooler than 36/37C. (which it probably is)

If I stick it under my armpit, then it reads 36.1, but the palm of my hand is currently 34.4 - that's with it pressed right to my palm. Back of my hand - 32.8, inside wrist 34.0. Top of foot (I've been barefoot most of the day) 32.2, sole 29.9 ...

Gordon

[Rod]

Thanks Steve. Should have said, we have one of those. But it is very limiting in that you can only make a few consecutive measurements. After that it needs to 'rest' for a while before it will measure again. There seems to be no way round that - it is its built-in way of doing things. Also, if it thinks the temperature is below a certain point, it will simply not make a reading.

[Steve Walker]

Ah. The only option I can think of there is having a handfull of them and cycling through them so that each one gets a rest, but I don't know if that'd be practical as I've no idea how long they have to rest for. I wonder if any different models would be better - I noticed that some versions are significantly faster at taking readings, but I've no idea what else would vary in their specs.

SteveW

[Steve Walker]

IIRC the norm with good old fashioned mercury clinical thermometers is to take oral readings as read, but to add half a degree centigrade for an underarm reading. Rectal readings are the best, giving core temperature.

SteveW

[Gordon Henderson]

I think what I was saying was that measuring the skin temperature might not be a good thing - it's going to be highly variable - unless that's actually what the OP wants...

Gordon

[Escom]

I hospital we use skin temperature sensors which have self-adhesive pads so that the semiconductor diode is in contact with the skin resulting in changes in the voltage due to temperature changes in the skin. The sensors are disposable and therefore quite cheap but the monitor that they plug into costs an arm and a leg. Normal clinical thermometers, including infrared tympanic thermometers, are designed to measure core body temperature in the hypothermic, normal and pyrexial range, and are not suitable for measuring skin temperature which is much lower.

Normal core body temperature is around 37 degrees Celsius whereas normal skin temperature is around 32 degrees Celsius, but it depends where the sensor is placed on the skin - readings on the skin of the head (which has an excellent supply of arterial blood) are usually around 32 degrees but are much lower in, for example, the lower legs and feet, because the blood supply is less. Obviously, skin which is covered with clothing is warmer than skin that is not covered, irrespective of where the skin is.

[Harry Bloomfield]

Rod wrote on 01/08/2009 :

I have bought two different models from Maplin and I really cannot fault either of them for accuracy, speed of reading or anything else really. Both were put thoroughly through their paces.

Probably the most unexpected thing it could do was measure the sky and cloud temperatures :D

The first much more expensive one, I dropped, wrecking the display. The second one I paid £19.99 for, but they are now £29.99 (orange and grey).

So far as measure skin temperature goes - as no doubt you will know, skin temperature is not the same as body core temperature. Depending upon losses, the skin can be much cooler. The nearest I could get to a core temperature was armpit temperature, about a degree or so low of core temperature.

[Owain]

However that assumes they are all calibrated extremely close to each other if the measurements are to be repeatable across different patients, at different times, on different thermometers.

One thermometer per patient is, I thought, what happened with the old mercury thermometers in a little holder at the bedhead, to ensure that the trend was recorded accurately over time.

Owain

[Steve Walker]

It's easy enough to do comparisons to permit corrections between thermometers though.

SteveW

[Steve Walker]

I was only saying that you were correct, you will read different temperatures at different parts of the body and if you are using this to read body temperature, you must correct for it.

However, the OP's link is to a dermatological reference and suggests that variation of skin temperature at affected sites is what they are looking at, not actual body temperature. As it happens I have a skin problem (originally diagnosed as Urticaria, but then as possibly linked to Sarcoidosis, but I still don't know) and one of my symptoms is hot, itchy patches that form temporary weals with the lightest of scratches - but it's unrelated to the OP really.

SteveW

[Rod]

Time to jump back in.

This is for a very specific problem where the local skin temperature can be anomalously high. That is, higher than the deep internal temperature in the area. The cause is not known but it does strongly suggest that there is something going on other than a weird blood flow issue (which has often been claimed to be the underlying problem).

Although measurement will not help directly, it seems to me that we can at least try to see how things are progressing.

Thanks so far to all. Any comments on the cheap Order Code: N19FR at Maplins? Not just because it is less expensive, but because it is neater.

[Rod]

You are on the ball, Steve! Not knowing isn't much fun - we have had three years of that! BTW A cold kitchen floor is very useful for this.

Got a friend with sarcoidosis - he was diagnosed quite a long time ago and the medics seemed all to be interested because it was new to them.

But the thing I always want to know about is reflected in my sig. And that can directly or indirectly cause numerous dermatological conditions.

[Peter Parry]

The accuracy of these is plus or minus a few degrees. The resolution is more a function of the display chosen than the devices ability to measure to anything like that degree of fineness. I doubt if you will get consistent or reliable readings if you are hoping for fraction of a degree measurements.

[Steve Walker]

Yes, I was checked for that, but later I actually reached the stage of a Consultant telling me that I had Lymphoma and that they would need to get a camera in and to take samples purely to determine stage and grade, then I arrested when they were trying to bring me round, only to be finally told that it wasn't Lymphoma at all, but Sarcoidosis. All this over Christmas

2007 and while my wife was 8 months pregnant!

SteveW

[Harry Bloomfield]

Rod laid this down on his screen :

I cannot seem to find that one, my current one is the N93FX. Much like cheap multi-meters, the stated accuracy is quite poor, but I have yet to find a cheap multi-meter with more than the basic digital + or - count inaccuracy when compared to my calibrated standards. The biggest source of inaccuracy with these IR temperature instruments, is the type of surface you are measuring. A matt black surface providing the best accuracy.

Ebay is a good source for these. Most of them are produced in China and via Ebay you can import one direct from China, but watch out for the postal charges and if paying more than £18, possibly some duty to be paid.

[Rod]

She ain't going to paint her feet black... :-)

That model is on - looks like it is also at CPC (at a higher price).

Will trawl ebay and elsewhere to see what else is out there. Thanks.

[Dave Liquorice]

I looked through last night and see that the cross beam laser IR thermometer (N85FR) has a "Basic accuracy: +/- 1.0% of reading" and a "Distance to spot size ratio is 12:1". That might be at the measuring distance of 6" though.

The N85FR is on offer at =A349.99 the same as the grey/grey N28BJ (Optical resolution: 8:1 Distance to spot size ratio, Basic accuracy:

+/- 2% of reading or +/- 2=B0C / +/- 4=B0F.).

The orange/grey N93FX is =A329.99 but that is "Optical resolution: 8:1 distance to spot size ratio, Basic accuracy: +/- 2.5% of reading or

2.5=B0C / 4.5=B0F"

The pocket N19FR is =A316.99 "Infrared accuracy: +/- 2% of reading or

+/- 2=B0C/4=B0F, -10=B0C to -35=B0C: +/- 4=B0C, 14=B0F to -13=B0F: +/- 7= =B0F" "Field of view: 1:1 optics ratio".

[Andrew Gabriel]

They are mostly calibrated for an emissivity of 0.95. Human skin has an emissivity of about 0.98 (verses a black body emitter of 1.0) so skin should read more accurately than a perfect black body emitter.