Transitor as heater

Heating, yes; sensing, no (unless you get one of those special metal and glass package ones that's miliwatts and kilodollars..).

I have here a single-AA flashlight that I made, fitted with a high power LED, which runs about 1W at 67% electrical efficiency. The emitter itself is going to be around 10%, or overall, about 60mW light output, within the

350-700nm band. Holding it point blank to my lip and turning it on and off, it feels about like warm breath, except I'm holding my breath, and it's switch operated. Absorption is absorption.

The same goes for radiation below the blackbody peak, of course; induction heating doesn't have to resonate with the work to make it glow. (Which is, in effect, a very efficient, extraordinarily noisy, wideband upconverter. :) )

Tim

--
Seven Transistor Labs 
Electrical Engineering Consultation 
Website: http://seventransistorlabs.com

'Tain't quite as bad as that. Blacken one side of a piece of copper foil with a torch. Heat it from the back with a fibre-coupled 850-nm laser, and put the sample on the front (shiny) side.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

My problem with an optical power measurement, is that beyond the ~10% error level it get's hard/expensive. How good is the blackened bit of copper?

George H.

that,

pulse

at

hoping to be just able to wait ~5 time constants or so. I must admit I thought it might be fun to watch the heat leak out of the transistor.

Ya know George, the longer i think about this the less i like trying to get a specific power for a specific duration with a transistor, and the more i like SMD resistors, perhaps mounted "upside down".

?-)

r.

ially)

of controlling the current

nsistor as a temp sensor with the c-b shorted. (diode connected transistor .) Could I just push a bunch more current through it for a heater.

from heater to temp sensor. Do I use relays or analog switches?

I needed to use a transistor as a log/antilog element but ambient temperatu re is part of the equation. I used a matched quad transistor array from Ana log Devices. One transistor is the temp sensor, 2 are used as heaters and t he fourth transistor actually does the 'work'.

The control opamp was from a note by Jim Williams and it works very well. I had to use 15 ohm emitter resistors on the heaters to limit the current to a safe value when cold. Watching the emitter current of the heaters and th e temp sense output while placing your finger on the 14 pin DIP case shows heater current going up while you touch and falling back down when you let go. The temp output shows a constant 462mV which corresponds to 69C interna lly.

It can be very good indeed, if e.g. you fold it back on itself and just have the fibre come through a small hole. Multiple bounces off black surfaces are the ticket.

But for your use, probably a bit of anodized aluminum, with sample on one side and a Kapton flex on the back. I'd try putting the component side next to the aluminum, with the resistors and RTDs in good contact with the back. A blob of indium solder on the top of their pads will squash down very nicely on the Al, and will survive temperature cycling (as long as you don't melt it on the high-T excursion). The anodizing will give you nice durable insulation with low thermal resistance.

Thin copper traces on flex are the lowest thermal conductance solution that I know of. You can make them serpentine if you like, to gain about a factor of 3.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

I've used that trick quite often, since seeing it in an app note back around 1990, when the MAT-04 first came out. Works better in the ceramic DIP package than the SMT ones, but they don't make those anymore of course.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

How do you get pick-n-place to place them upside down? What sort of manufacturing process hoops do you have to jump through to do that?

Might have to re-reel them, but you don't need to. Indium-solder them to the anodized Al, top down.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

Won't the solder short the resistors?

...but you have given me another idea. How expensive is indium solder?

To get the heat out, just mount them normally, soldered to biggish copper pours and then vias to copper on the opposite side of the board. Most of the heat flows out of the end caps.

Or use some nice copperclad aluminum nitride.

Inverted resistor mounting is useful at picosecond speeds. The reduction in inductance can be dramatic.

--

John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    

Precision electronic instrumentation

Not unless you let it bridge between the pads.

Expensive.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

On 2/15/2014 12:21 PM, John Larkin wrote:

Tends to be slow, though. I'm a big fan of maximizing bandwidth in thermal systems.

Interesting idea, thanks.

Cheers

Phil Hobbs

-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203 Briarcliff Manor NY 10510

hobbs at electrooptical dot net

On 2/15/2014 12:21 PM, John Larkin wrote: > On Sat, 15 Feb 2014 11:33:13 -0500, Phil Hobbs > wrote: > >> On 2/15/2014 11:30 AM, snipped-for-privacy@attt.bizz wrote: >>> On Fri, 14 Feb 2014 21:01:06 -0800, josephkk >>> wrote: >>> >>>> On Thu, 13 Feb 2014 18:45:32 -0800 (PST), George Herold >>>> wrote: >>>> >>>>> On Thursday, February 13, 2014 7:43:58 PM UTC-5, Clifford Heath wrote: >>>>>> On 14/02/14 02:51, Tim Wescott wrote: >>>>>> >>>>>>> On Thu, 13 Feb 2014 07:26:01 -0800, George Herold wrote: >>>>>>>> Transistor as heater/ temp sensor. >>>>>>> Using it as a temperature sensor would be complicated by the fact that, >>>>>>> due to its recent service as a heater, it would be warmer than its >>>>>>> surroundings. >>>>>> >>>>>> But you have time - measure the temperature curve after a heating pulse >>>>>> and correlate that with a thermal model. Using a thermal model is the >>>>>> only way of calculating how much of your heat has reached the target at >>>>>> a given point in time, so you have to do it anyway. >>>>>> >>>>>> Clifford Heath. >>>>> Yeah, Well I wasn't thinking as far ahead as curve fitting. I was hoping to be just able to wait ~5 time constants or so. I must admit I thought it might be fun to watch the heat leak out of the transistor. >>>>> >>>>> George H. >>>> >>>> Ya know George, the longer i think about this the less i like trying to >>>> get a specific power for a specific duration with a transistor, and the >>>> more i like SMD resistors, perhaps mounted "upside down". >>>> >>> How do you get pick-n-place to place them upside down? What sort of >>> manufacturing process hoops do you have to jump through to do that? >>> >>> >> Might have to re-reel them, but you don't need to. Indium-solder them >> to the anodized Al, top down. >> >> Cheers >> >> Phil Hobbs > > To get the heat out, just mount them normally, soldered to biggish copper pours > and then vias to copper on the opposite side of the board. Most of the heat > flows out of the end caps. > > Or use some nice copperclad aluminum nitride. > > Inverted resistor mounting is useful at picosecond speeds. The reduction in > inductance can be dramatic. > >

-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203 Briarcliff Manor NY 10510

hobbs at electrooptical dot net

Yeah, $100 for 3' x .030" wire. It's a bit much for production but still interesting as a lab toy.

I don't think flipping surface-mount resistors will reduce tau much, unless you put some thermal filler between the resistor and the PCB. The air gap there will have a much higher thermal resistance than the path through the alumina.

For a huge power-to-mass ratio, use AlN or BeO RF terminator resistors. Maybe George could use a flange-mount terminator as his heater.

Here are some waveforms, 0805 inserted in a gap in a CPW trace:

The bump at cm 3.5 is the resistor's inductance.

--

John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    

Precision electronic instrumentation

I'm talking about indium-soldering the tops of the caps to the back of the anodized Al. The thermal diffusivity of plastic is roughly 1000 times slower than aluminum.

Looks like about a factor of two improvement. Nice.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

MAT14 is SOIC and available.

Sure, I know about that, but (a) they gutted the datasheet, meaning that they gutted the testing, and (b) they don't come in CERDIP.

CERDIP packages have Kovar leads, which have much lower thermal conductance. (They also have worse thermocouple offsets, so you have to be careful there, or at least you did when you could still get them.)

Cheers

Phil Hobbs

(*) I wrote a paper long ago that included this trick:

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

at,

e

t

ng to be just able to wait ~5 time constants or so. I must admit I thought it might be fun to watch the heat leak out of the transistor.

Hmm OK why is that Joseph? I think I'm going to have to run four wires dow n to do either ther resistor or transistor as heater. (measure the voltage and current.) I guess there is some beta current error.. I've ordered a b unch of pnp medium power transistors. Beta's in the ~200 range but that dr ops at low temperature. But the SOT89 packs look great... they weight less than 100 mg! I guess I should look at some FETS...

George H.