Thermistor for oscillator

MrThermister inscribed thus:

A thermistor with similar characteristics is commonly found in computer power supplies that have variable speed fans. The old HP Vectra (486 Vintage) used one to good effect. As do some Compaq models.

Good Luck !

Is it for a repair ?

Graham

No it isn't.

The RA53 was a low mass (bead) thermistor in a glass tube. The thermal time constant and indeed the heating power which is critical for oscillator stabilisation would be wildly different for one thing. I'm also doubtful that the resistance is going to be the same too.

Graham

Use a very small lamp instead,like the ones used in internal lighting of switches. The positive temp co. of these can stabilize a wien bridge nicely. You can even see the resistance change with the small current applied by an ohm-meter.

You might consider adapting the glass-encapsulated ptc's in DO7 or DO35 body sizes. Thermally isolating the parts might give usable performance, particularly with DO7, where the thermistor is not physically coupled to the glass - hence the relatively long thermal time constant in measurement applications - 55seconds.

Removing the leads and replacing them with lighter gauge wire would increase this further.

Murata makes PTC thermistors in SMD 0402 packages, which is pretty small, however they don't begin to see 4K7 impedance till their bodies hit 65degC or more, which is hardly likely to fit into a low-power scenario.

I recall that the RA53 had to be protected from stray light, if repeatable performance was to be achieved at 20degC.

RL

I'm sort of curious why the ptc ended up being used - I assume it was because of the greater beta. An NTC of sufficient sensitivity could ideally be used to stabilize amplitude if it's position were swapped with the reference R.

RL

Eeyore inscribed thus:

Hi Graham, I agree that the RA53 is a low mass device! However I did say similar! I had a look at a couple today, that I salvaged. At room temperature they both measure 4.65K. In terms of size they are about the same as the glass headed map pins (2.5mm) diameter and 1.5mm thick.

FWIW I also measured a salvaged thermister from a computer mainboard (the one under the cpu) which is physically similar in bead size (about 0.5mm, I couldn't measure the diameter) to the RA53... That measured a whopping 15K at room temp.

I see. That's different to what I thought you meant in that case. I haven't seen one of those.

The ones I've seen of that type have a much larger bead than the RA53 and critically the leads are much thicker and will conduct heat away, which means its self-heating won't be any use for stabilising an oscillator.

Graham

Eeyore inscribed thus:

If the ones from under the cpu could/can be used, the one I measured today was like a narrow strip of very thin flexible pcb, about an inch long. The bead looked like a black dot at one end connected to flat conductors that came out at the other. When a cpu is in place this would be pressed up against the bottom simply by the springiness of the lead/s. Crude but effective !