# Platinum heater

Discussion in 'Electronic Components' started by Cornelius, Aug 27, 2003.

1. ### CorneliusGuest

Hello All

I am a student attempting to manufacture thin film heaters by depositing
platinum on a ceramic base. My heaters keep burning open when I apply
even low voltages. Have any of you ever tried this. Can anyone give me
any suggestions on how to make this work or how to predict the failure
of such a heater or a resistor in general.

Thank a bunch

Cornelius

2. ### Dana RaymondGuest

This is a stab in the dark... Do your heaters need to be heating something
so as to not burn out?
The thermal load will lower the thermal resistance, thereby keeping local
temperatures below the melting point of platinum.

Make sense?

Dana Frank Raymond

3. ### TheoGuest

Place the heater in some fluid, like water.
The heater will not burn out and the water will warm up
as a result.
QED

4. ### John LarkinGuest

What's the size? Film thickness? Typical resistance? Is there a
serpentine pattern, or just an overall film layer? How are you
terminating to it? How much voltage are you applying?

John

5. ### Steven BastienGuest

Cornelius,

Your question is out of my expertise, but this story could be interesting to
you.

I once used steel wire to make a high wattage resistor as a temporary test
load for a driver circuit. On the first attempt, I used a guitar string
(0.009 in diameter) and long enough to give 5 ohm resistance. With 15 V
applied, this quickly heated in the air (because of the small surface area
and poor thermal conductance to air). In addition, the resistance of the
wire dropped significantly as it heated, creating a type of thermal runaway
condition. The wire glowed red hot and then burned out quickly. On the
second attempt I used thicker stainless steel wire and immersed it in water,
which worked well enough for temporary testing of my circuit. A week later,
on telling the story to my neighbor, I found out he had a box of old 8 ohm
100 watt resistors in his garage and two in parallel would have worked
perfectly. (what are the odds?).

Anyway, the point is to check thermal conductivity of the platinum to the
ceramic base, and the temperature sensitivity of platinum resistance. If
you find that platinum resistance drops as temperature increases, it might
be better to use a current source rather than a voltage source. This would
create a thermal feedback since P = I^2 * R and avoids thermal runaway from
P= V^2/R.

Steve B

6. ### Phil HobbsGuest

If the ceramic is too rough, and the metal too thin, the film may not be continuous.
This will lead to local hot spots. Make sure that the metal is several times thicker
than the rms surface roughness, and work down from there. Also, you haven't told us
where the heaters burn out--is it somewhere in the middle or right at the leads?
What do the failed units look like? Are there hunks of ceramic missing, are there
cracks, or just a toasted-looking region in the metal?

Cheers,

Phil Hobbs

7. ### CorneliusGuest

Thanks to all who offered advice on my problem.
I am using uncoated platinum metal. I am not sure it would be safe to immerse it
in water.
I measured the resistance of the films with a multi-meter. The are on the order
of 2000-3000 Ohms, and yes they are in a serpentine pattern. The failure site
varies on different runs but most occur at a transition pint in the film width;
ie at a point were I from a width of several millimeters to a width of a
fraction of a millimeter.
I do suspect local overheating but I don't know how to confirm that suspicion or
prevent it if that is the case. I will try and locate a current source as was
suggested, but failing that I am still open to suggestions.

Again thank you for all the positive feedback.

Cornelius

8. ### CorneliusGuest

Thanks to all who offered advice on my problem.
I am using uncoated platinum metal. I am not sure it would be safe to immerse it
in water.
I measured the resistance of the films with a multi-meter. The are on the order
of 2000-3000 Ohms, and yes they are in a serpentine pattern. The failure site
varies on different runs but most occur at a transition pint in the film width;
ie at a point were I from a width of several millimeters to a width of a
fraction of a millimeter.
I do suspect local overheating but I don't know how to confirm that suspicion or
prevent it if that is the case. I will try and locate a current source as was
suggested, but failing that I am still open to suggestions.

Again thank you for all the positive feedback.

Cornelius

9. ### CorneliusGuest

Thanks to all who offered advice on my problem.
I am using uncoated platinum metal. I am not sure it would be safe to immerse it
in water.
I measured the resistance of the films with a multi-meter. The are on the order
of 2000-3000 Ohms, and yes they are in a serpentine pattern. The failure site
varies on different runs but most occur at a transition pint in the film width;
ie at a point were I from a width of several millimeters to a width of a
fraction of a millimeter.
I do suspect local overheating but I don't know how to confirm that suspicion or
prevent it if that is the case. I will try and locate a current source as was
suggested, but failing that I am still open to suggestions.

Again thank you for all the positive feedback.

Cornelius

10. ### CorneliusGuest

Thanks to all who offered advice on my problem.
I am using uncoated platinum metal. I am not sure it would be safe to immerse it
in water.
I measured the resistance of the films with a multi-meter. The are on the order
of 2000-3000 Ohms, and yes they are in a serpentine pattern. The failure site
varies on different runs but most occur at a transition pint in the film width;
ie at a point were I from a width of several millimeters to a width of a
fraction of a millimeter.
I do suspect local overheating but I don't know how to confirm that suspicion or
prevent it if that is the case. I will try and locate a current source as was
suggested, but failing that I am still open to suggestions.

Again thank you for all the positive feedback.

Cornelius

11. ### CorneliusGuest

Thanks to all who offered advice on my problem.
I am using uncoated platinum metal. I am not sure it would be safe to immerse it
in water.
I measured the resistance of the films with a multi-meter. The are on the order
of 2000-3000 Ohms, and yes they are in a serpentine pattern. The failure site
varies on different runs but most occur at a transition pint in the film width;
ie at a point were I from a width of several millimeters to a width of a
fraction of a millimeter.
I do suspect local overheating but I don't know how to confirm that suspicion or
prevent it if that is the case. I will try and locate a current source as was
suggested, but failing that I am still open to suggestions.

Again thank you for all the positive feedback.

Cornelius

12. ### CorneliusGuest

Thanks to all who offered advice on my problem.
I am using uncoated platinum metal. I am not sure it would be safe to immerse it
in water.
I measured the resistance of the films with a multi-meter. The are on the order
of 2000-3000 Ohms, and yes they are in a serpentine pattern. The failure site
varies on different runs but most occur at a transition pint in the film width;
ie at a point were I from a width of several millimeters to a width of a
fraction of a millimeter.
I do suspect local overheating but I don't know how to confirm that suspicion or
prevent it if that is the case. I will try and locate a current source as was
suggested, but failing that I am still open to suggestions.

Again thank you for all the positive feedback.

Cornelius

13. ### Anthony Q. BachlerGuest

heating of any substance creates free electrons, which enhance conduction.
The more you heat it the more free electrons, when a substance starts
visibly glowing that is because you are tearing electrons from the inner
shells and they are emitting photons as they fall back to the low energy
states.

14. ### Phil HobbsGuest

Spinach. The number of free electrons in a metal is a very, very weak function of
temperature--every metal atom is already contributing at least one, so where are the
others going to come from?

Cheers,

Phil Hobbs

15. ### ddwyerGuest

My knowledge also limited but I suggest you do not have a uniform film.
If evaporating initially deposit nichrome to give adhesion and then
whilst under vacuum platinum.
I once powered a transistor as a quartz crystal heater in a vacuum
envelope.
My high thermal resistance connections were through fine platinum wires
which glowed red heater with 100mA.
Interesting to note that platinum wire is a temperature sensor under
other circumstances so its Tc is well known.