Which type of thermocouples should I use for measuring laser diodes temperatures?

[kitatit]

Hello laser friends

It's been a while since I've been here. I've been lurking recently and it's gotten me vibed to get back into experimenting mode.

Planters Youtube videos have gotten me all excited to get in and do a bit of science to, 1 get a much better understanding of optics and 2, better understand heat transfer and dissipation from diodes through to heat sinks.

His success with using PC CPU liquid cooling prompted me to buy a water cooling system myself. I bought this one http://www.coolermaster.com/cooling/...r/nepton-240m/

I haven't spent much time trying to understand the maths of how to calculate what size heat sink for a given heat load because I'm still a bit perplexed by much of the theory. I need some hands on experimenting and recording of results to see things in action. Ive got some pretty modest kit for recording temperatures but I think it will prove useful. It's a Picotech TC-08 8 channel thermocouple data logger which you can log temps via your PC. The science involved in choosing the heat sink for my projector involved asking peoples opinion and choosing a very big heat sink for overkill. It has done the job but jeesh it is a bit heavy.

For a couple of projects I'm working on, It is much more important that the cooling system be more closely optimised to reduce the weight. So for my own understanding and to hopefully add
some knowledge to the pool, I want to do some temperature logging at several points on a laser system, from the diode housing(as close to the diode as possible), the base plate near the diode housing on the laser side, the base plate on the bottom side close the where the water-cooling block is and ambient temp.

Here is a useful article about testing LED temps. http://www.osram-os.com/Graphics/XPi...rmocouples.pdf

Now the questions I've got about thermocouples are;

-Should I use K type or T type?
it seems T Type are the go as they are more accurate but still cover the temp range for diodes.

Below are the ones I'm looking at.
http://www.tcdirect.net.au/deptprod.asp?deptid=180/21

http://www.tcdirect.net.au/deptprod.asp?deptid=190/1



Would it be ok to use the expose junction type in the laser diode housing? I'm planning on drilling a 1mm hole into one of Dave's brass and copper housings and glueing it in with Arctic silver or Arctic Alumina Adhesive. They are my first choice as the are cheaper than the metal clad mineral insulated type and have a quick response too. I'm probably over thinking it but I didn't want to buy a bunch of them and find out they aren't suitable or "electrical incompatible being so close to the diode.

Will it matter which type I use?

Cheers
Kit

[mixedgas]

T is a pita to wok with, not so flexible, and hard to weld. Cheap Modules and Chips are readily available for J and K.
Says the guy who has to string 220 feet of T on monday to 21 differernt locations then calibrate a data logger for it.
Simply because some guy here 30 years ago had an extra spool of T, it became the department standard. Which is painful because everbody else here uses J and K.

The slight improvment of T for low temperatures will not offset the higher cost of T hardware and controllers.

Minute currents flowing thru an open junction will mess you up. But I'd rather have to try to isolate that junction then use a slow jacketed TC, unless my controlle masses were much larger then the mass of the jacketed junction.

Mr. T-Gas...

[kitatit]

Hi Steve,

Thanks for your thoughts. Ive used jacketed TCs and did notice a big difference in response time going from a 1.5mm down to a 0.5mm. I see now that this company I bought TCs from before has jacketed ones down to 0.25mm. If they weren't so pricey I would have gone for that. I've mostly used the compression fittings to mount the jacketed ones quiet a few years ago doing automotive R&D. as Dave's diode housings are so small there isn't much space to screw the compression fitting in so I could save the jacketed expensivish TC for another test later. Seems whatever I use, it will be glued in via Arctic Alumina and not removable.

OK, say I go for an open junction K type with the twisted insulation. When you say it will mess me up, what issues will I have and how could I overcome them? I'm picturing the TC junction sitting at the bottom of the 1mm hole in contact with the brass/copper housing which would be in electrical contact with the base of the diodes "flange/case". Could I just hook up my multimeter to one side of the TC lead and the other to Dave's diode housing and set it to continuity. Then apply the Arctic Alumina adhesive and push it down to the bottom of the drilled hole until it contacts, then back off just a hair so there is no electrical connection and make sure it doesn't contact while the epoxy is curing? Is that how you would isolate the junction?

[VDX]

... I was developing TLD-sensors some years ago and used PT100 resistors and 24Bit-ADC's to measure and control the 'selfcalibrating' of the sensors at 60 degC with 0.002 degC accuracy and at 140 degC with 'only' 0.01 degC.

There are several types available with flat 'open' ceramic chips or as 'sealed' 2mm thick ceramic rods ...

Viktor

[catalanjo]

@ "Picotech TC-08 8 channel thermocouple data logger which you can log temps via your PC."
Just between Kit&Cat where dja get dat from? They cost £250 in UK
Cheers

[kitatit]

Hi Cat

I bought it about 3years ago from www.picotech.com.

Have a look here on the ozi RS site
http://au.rs-online.com/mobile/p/dat...38207069636F26

It is $371 AUD without tax. Could be worth keeping an eye on as the Aussi dollar has been dropping quiet a bit.

If RS Aus won't ship to you, let me know I can get it for you and send it. It would just be 10% extra for ouR GST (VAT).

Good luck.
Kitty Kat