Small PID regulator for temperature control

[janne102]

I am trying to put together a compact temperature control instrument to be placed inside a small chamber. For this I would like to buy a PID regulator, preferably with a sensor (Pt100) input and a relay output to switch a battery on/off. The difficult part is that I only have a volume of about one cubic inch (15 cubic centimeters) available. I would prefer to buy something as complete as possible so I can get going as fast as possible.
Does anyone know where I can buy compact PID regulators for temperature control?

 

[felixc]

If you are willing to try making your own PID algorithm (not such a big thing), you may want to take a look at the "basic stamp" preprogrammed PICs from Parallax. Add an SPI A/D converter to read the PT100 and a regulator and you have something that might fit within a cubic inch.

Felix

 

[nbucska]

The 1 cubic" is more than enough for the controller but
it may not be enough for the power stage.

Can you give some specs?

<nbucska@pcperipherals DOT com> subj: eng-tips
read FAQ240-1032

 

[jsolar]

You mentioned PID and relay in the same request. PID is for continuous control and is difficult, especially the D part, in a bang bang. You will be better served with a bang bang controller, since it looks like you do not have proportional control. Otherwise, stick wit PI. You would want to get our PI working before turning on the D part of the algorithm. I would suggest not using the D unless absolutely demanded by the performance of your application (unlikely).
Best Regards,
John Solar

 

[buzzp]

Omega Engineering has a wealth of information and many types of temperature controllers for sale, amongst other companies. Of course, if the 1 cubic inch is true, then your likely out of luck trying to fit a relay in 1 cubic inch, and the electronics for a PID controller.
Can't you just put the thermocouple in the controlled environment - "chamber"? What are you controlling? As mentioned, PID might not be for you.

 

[OperaHouse]

Be curious if anyone has tried this. I was playing around PTTC thermal fuses and these have a very sharp resistance change around 100C. One of these and a resistor would make a nice po boy oven for a crystal.

 

[itsmoked]

ummm, don't you need to de-power them for them to reset?

 

[OperaHouse]

Not exactly. Anything that drops their temperature will start to reset them. Just touching your finger to them will drop the resistance. The ones I experimented with need about 3-4W in ambient temperature to maintain their "off" conditions. This was a 4A unit and the "off" current was about 1/4A @12V. Raise the voltage to 24V and it was 1/8A. Whatever it took to keep the temp at the transition point.

 

[itsmoked]

I don't think it would work...

You have a box.. A heater run thru a PTC that is residing in the box..

You throw the switch.

The box heats up.

The "PTC fuse" trips eventually at say 90C. It stays tripped down to 20C??

Kinda too much hysteresis maybe?

Not that special PTC couldn't do the job. Raychem makes heater ribbon that is PTC material in a matrix. The ribbon heats up to it's design temp and no more since the matrix expands and the resistance climbs.

You could probably put this stuff in an insulated inclosure that would then stay at some fixed delta from the outside environment.

 

[OperaHouse]

You should play with one or at least look at the curves for a RXE before you comment further. PTTC does not "trip", drop the temp of it a couple degrees and the current increases dramatically.

 

[OperaHouse]

Look at the chart in this article

http://www.eetasia.com/ARTICLES/

2005MAR/B/2005MAR01_CTRLD_TA01.pdf

 

[OperaHouse]

Well, that didn't seem to work. Suggest you google "polyswitch" and find the eetasia article on the firat page.

 

[itsmoked]

OperaH;

I have used PTCs many times... They're in at least four products I've designed. Generally they are a major pain in the butt to use. They are so sensitive to ambient that half the time your product will fry b4 they trip and at low temps the PTC will trip prematurely!!


I hunted down the specific pdf you mentioned which catagoriclly stated that "the power must be removed" to reset them.

I have included the relavent passages below;

"When the fault is cleared {(and power is removed)}, the system begins normal operation with no service or part replacement necessary."


"{Once the power is removed} and the fault is cleared, the
temperature of the PolySwitch device drops below its trip
temperature, allowing it to reset to its low resistance state which enables the circuit to operate normally."


But all this a-side, these suckers aren't temperature specified! They are CURRENT specified at some ambient. This means you'd have to jam specific amounts of current thru the PTC to get it to trip *near* some specific ambient.


Furthermore PTCs drift *A LOT* 5% in 1000Hrs. So your temp setting would change over time. Rather quickly!

If you don't really care about the temperature as in water pipe anti-freezing heat tracing, where the pipe must remain above 32F, then poly type temp control will work fine as no one cares about drifting from 40F down to 38F.

This all a mute point anyway as this poster was an obvious *PAR* {Post And Run}

 

[OperaHouse]

A polyswitch doesn't know current from Shinola. It is purely a temperature device with a little resistance.