Small Heater Advice

[dgminala]

I'm starting a project in which I need a small heated container (sort of a small ovenized enclosure) heated to around 40C.
I'm looking at various ways of heating the enclosure. I've looked at adhesive-backed silicone-based heating elements and polyimide heaters and even explored the possibility of using nichrome wire and kapton tape to fashion a heater.

The heater power is small, about 5W at first estimation. The heater will need to wrap around a smalll aluminum enclosure, and the adhesive backed heaters look well suited for that.

I'm asking for recommendations as to a prefered type of heater, based on previous experiences. Has anyone here used any of these types of heating elements, and what are the pos & cons of each? For instance, does the adhesive deteriorate over time? Is one type more prone to burnout than the other? In the event of failure, are the adhesive-backed heaters difficult to remove?

There doesn't seem to be a major difference in the cost of the different types of heaters, so that's not a big issue. I'm only making a few of these, so not looking at volume pricing.

Any suggestions? Thanks for any you might have.
DaveM
QuoteReport

[thm_w]

At 40C with quality 3M adhesive it won't deteriorate. But you'll have insulation on the outside, which helps hold it in place right.

Is there a reason you can't just use a aluminum resistor inside or outside of the enclosure, does it need even heating and has to be thin?
What is the power supply voltage used for heating?

[Warpspeed]

Dave, how about a metal clad resistor ?
These are available from a few watts, to several hundred watts in any required resistance.

Bolt one of those to your aluminum enclosure, either inside or outside, and the heat will spread wonderfully well.
Use the largest physical size that is convenient, even though only five watts are required.
Cheap, simple, and they look good.

[NiHaoMike]

40C is not that hot at all, a power transistor in a current sink configuration will allow easy control of the heat dissipation.

[Siwastaja]

If you have a control PCB there anyway, and have some free routing area on either side (or mid layer), you could use a long zig-zag trace. Power output won't be accurate but does that matter if the target is just to regulate 40degC in an insulated case, probably anything between 2-10 W does fine.

Cost would be a single switching transistor on a microcontroller IO pin. The trace could run off from 5V supply.

[dgminala]

At 40C with quality 3M adhesive it won't deteriorate. But you'll have insulation on the outside, which helps hold it in place right.

Is there a reason you can't just use a aluminum resistor inside or outside of the enclosure, does it need even heating and has to be thin?
What is the power supply voltage used for heating?

Yes to both issues.  There is no room inside the enclosure for a big resistor, and the temperature does need to be maintained rather accurately throughout the inside of the enclosure.  That's why I settled on a heater that can be formed closely around the enclosure, such as an adhesive backed flexible heating element can provide.  I thought about having a resistor or transistor provide the heat, but the enclosure doesn't have the mass necessary to distribute the heat and maintain the temperature.

I've been inside a number of OCXOs such as the HP 10811 and 10544 models, which have a massive machined chunk of aluminum that can do the job well.  I don't have access to the machinery and expertise to run it to manufacture such a contraption.  Would be nice if I did, and having a machine shop make the parts is just too expensive.

And in answer to the question about heater voltage, everything must run from a 12-15v supply, which has already been determined.

Thanks for the suggestion,
DaveM

[Warpspeed]

How about a commercial diecast aluminium box ?
Bolt the resistor(s) on the outside, then embed the whole box in suitable insulation.

I once did that using a Peltier cell for both heating/cooling, and it worked rather well.

For most accurate temperature control, have you though of a box within a box structure ?
The outer box corrects for really large ambient changes fairly quickly, it can have a less damped response.
The inner box then has less of a battle, and can have a much slower more measured response requiring less power and lower temperature gradients.

[daniel444]

how about  silicone coated carbon fibre heating wire
try getting the smallest diameter you can

[DavidKo]

Take the aluminum box or metal tube, wrap it in glass fabric and for the heating use kanthal heating wire (create the "coil" on the glass textile). Additionally you can isolate it with glass fabric. Homogenous heating you will achieve by homogenous coil. Such a system can work also for higher temperatures. For lower temperatures you can use for insulation from something less heat resistant.