Temperature stabilised enclosures Heating and or Cooling

[beanflying]

Starting to look at some options for just heating enclosures for some Voltage References but also potentially cooling for other applications as my shack gets to well over 30 degrees.

These little 30-40W 12V cartridge heaters from evilbay are a bit over $1 each so hooked up in series 4 of them would give circa 10W driven from 12V using a linear reg to keep the noise at bay. The jackets of the elements are isolated from the electrics too btw. Stuff the rear of them with kevlar thread or similar or make some teflon plugs to keep the heat in?

Anyone had a go at Peltier devices strapped direct to cases or maybe better running them on water cooled loops to remove the electrics/noise away from the case?

Had a good search around and apart from the recent posts in the KX thread didn't find anything.

External monitoring/control of a 3 or 4 wired RTD to PIC or Arduino solution would be the most accurate option for control.

[TiN]

I think 10W of heat flux would be waaaay to much. I usually see around 1.5-3.5W numbers peak with my TEC setup (simple noname TEC strapped to bottom of the box) when there is ~30C differential).
Benefit of the TEC setup is that you can have heatflux in both directions, not only heating.

[Andreas]

Hello,

I fear that the cartrige heaters will give a unequal temperature distribution between the corners and the center of the enclosure.
Thats why I use always (mirror) heater foils.
For cooling I use a car cooling box. (with fan to distribute the "cold").

And do not forget to mount a over temperature switch if you leave your Arduino/PIC unattended.

You can also use a NTC for controlling the temperature.
It can be connected directly to the ADC input of a microcontroller.
A 10 Bit ADC will give you around 0.1K temperature resolution.
With a 12 Bit ADC (e.g. PIC24FV32KA302)  resolution is 0.025 K.
I use 33K NTCs with 27K pull-up resistor which is a compromise
of self heating of the NTC and the max source resistance for a unbuffered ADC input.

I am using e.g. 100Hz PWM for the heater which is suppressed by the multimeter
because it is a multiple of the (50 Hz) line frequency.

setup is usually similar to here:
https://www.eevblog.com/forum/metrology/t-c-measurements-on-precision-resistors/msg462298/#msg462298

with best regards

Andreas

[timb]

You could always use two power op-amps in a bridge configuration for linear drive of a TEC as that’s pretty low noise. (Though I wouldn’t do that for more than a 5W TEC unless you have room for a large heatsink for the amps.)

That said, TI has a nice inexpensive H-Bridge that can be used for PWM based TEC drive up to 15A, with a couple of added inductors and filter capacitors. (So long as you keep the ripple below 10% you’re OK to drive a TEC with PWM in the hundreds of kHz range).

A third option would be to use “time proportional” control, coupled with a pair of relays, SSRs or any off the shelf H-Bridge. This is basically really, really slow PWM. Essentially, you set a time window (say 10 seconds) and turn the TEC fully on for a portion of that window (50% would be 5 sec on 5 sec off).

I’ve currently got a small TEC based cooler (designed to hold 6 cans of soda) that I converted to a temperature control chamber for testing voltage references. It uses the last option mentioned above. It’s got a range of 5c to 65c (at 25c ambient) and can maintain the setpoint +-0.25c, which is pretty good considering it uses nothing but parts from my junkbox.

[Andreas]

Hello,

you could also let the Peltier running at full load and heat against the (slow reacting) cooler to get quick to the wanted temperature.
(a Peltier ages faster when there are frequent temperature changes).
I also use several temperature sensors (minimum one at the heater and one at the DUT) to get a faster control loop.

With best regards

Andreas

[Echo88]

I suggest MTD415-module + high power bridge IC like OPA2544 for bigger Peltiers.

https://www.eevblog.com/forum/metrology/mtd415-a-low-cost-rs232-controllable-bipolar-1-5a-tec-controller-module/
http://www.ti.com/lit/ds/sbos037/sbos037.pdf

[tszaboo]

Jim Williams suggested this circuit.
It is stable to 3.5mK with a 20K change, so probably quite good. Also, it is linear and Linear.

http://cds.linear.com/docs/en/lt-journal/Thermoelectric_Williams_Mar02_Mag.pdf

[beanflying]

Thanks for the ideas so far. I have a couple of Peltier modules as well to play with as well.

I have a few (16  )of these enclosures so some experimenting is in order to see how they heat or cool.

[Kleinstein]

The efficiency of peltier coolers is best if the current is reduced proportional to the temperature difference. Thus for running them for a 10-20 K temperature difference, one would ideally not use the  full rated current but more like 1/3 to 1/2.
Still the power density is quite high and thus a good thermal contact (and maybe a fan) is need at the outer side, especially for cooling.

For temperature control it is a good idea to not only measure the temperature of the inside, but also the temperature of the other side of the peltier element, as this will be a mayor path for external perturbations.

[timb]

Jim Williams suggested this circuit.
It is stable to 3.5mK with a 20K change, so probably quite good. Also, it is linear and Linear.

http://cds.linear.com/docs/en/lt-journal/Thermoelectric_Williams_Mar02_Mag.pdf

Just FYI, that circuit is designed for laser diode control and isn’t suitable for control for what the OP wants without a lot of modifications.

The TECs integrated into these lasers are relatively small and run at lower voltages and currents (5V@2A or less), which is OK since they are cooling a relatively small thermal mass. (The laser, TEC and thermistor generally come as a complete, integrated package.)

When used for scientific, medical and imaging applications these lasers require very, very, very high temperature stability (0.01c or better) to stabilize their output.

The enclosure the OP is using is much larger than the type laser diode assembly that app note is targeting. Therefor the thermal time constants will be much greater, making it much more difficult to stabilize the loop gain bandwidth of the error amplifier. You’d also need dual supplies for that circuit (+-7V for a 5V TEC or +-15V for a 12V TEC) which is a pain.

You’d be better off getting two power op-amps (TI makes a nice one capable of over 2A and up to 36V) and driving them with a DAC from a micro in a bridge configuration.





For the temperature sensor, if you don’t need better than 1% absolute accuracy or 0.125c stability, you can’t go wrong with using the venerable DS18B20. A lot easier than fiddling with thermistors and it’s relatively low speed, so you don’t have to worry about noise. Also, it only requires two wires (data and gnd) so it’s easy to wire up.

Then you’d simply implement a PID algorithm on the MCU and away you go. (There are plenty of existing PID controller libraries, written for every platform and every language, some of them pretty good, too!)

[beanflying]

In my stash of bits I have a Willhi Peltier controller but a more dedicated one can get made at some stage. I haven't used this model before but their standard 240V fridge controllers work well. I will rig a couple of RTD sensors across the enclosure and do some playing on the bench. Bigger Peltier modules most likely needed?


http://willhi.com/product_detail.php?product_id=115&lang=en

[tszaboo]

Jim Williams suggested this circuit.
It is stable to 3.5mK with a 20K change, so probably quite good. Also, it is linear and Linear.

http://cds.linear.com/docs/en/lt-journal/Thermoelectric_Williams_Mar02_Mag.pdf

Just FYI, that circuit is designed for laser diode control and isn’t suitable for control for what the OP wants without a lot of modifications.

You obviously haven't read it. loop compensation leads to AN89, where it revealed that the laser used has a temperature time constant measured in minutes. Also, the analog method is 2-3 orders of magnitude better at the temperature control. I have no idea, why would you need a micro with all the extra work, instead of changing an RC value, and looking at it with a scope. For worse results.

[ManateeMafia]

In a pinch you can try the ILX Lightwave LDT-5910(B). I have one that works well and uses standard thermistors like the 44006/44007. It can be controlled via GPIB and is a good match with the low power peltier modules. Don't expect Keithley 2510 performance but it can be bought in the $100 range.

[Kleinstein]

Jim Williams suggested this circuit.
It is stable to 3.5mK with a 20K change, so probably quite good. Also, it is linear and Linear.

http://cds.linear.com/docs/en/lt-journal/Thermoelectric_Williams_Mar02_Mag.pdf

Just FYI, that circuit is designed for laser diode control and isn’t suitable for control for what the OP wants without a lot of modifications.

You obviously haven't read it. loop compensation leads to AN89, where it revealed that the laser used has a temperature time constant measured in minutes. Also, the analog method is 2-3 orders of magnitude better at the temperature control. I have no idea, why would you need a micro with all the extra work, instead of changing an RC value, and looking at it with a scope. For worse results.

Wether it makes sense to use a digital control with an µC instead of analog control depends on the time constants involved. The still rather fast Laser with a few minutes time constant might still work better with analog control, but for a larger and thus slower system it gets more an more attractive to use digital control, as the long time constants are difficult to to implement in analog hardware. Also with digital control anti windup  is usually standard  because it is easily implemented, while analog anti windup is tricky, especially if low drift is needed.  Anti_Windup can be important if the limits of the heater  / cooler are reached. Even for time constants in the minutes range, it takes quite some extra effort for the analog implementation - µCs and ADCs got better since AN86, so things have changes since than.

p.s.:
Another reason to use digital control is that peltier elements are nonlinear. Digital correction for this is easy, analog nonlinear functions tend to be temperature sensitive and thus drift can be a problem.

[timb]

Jim Williams suggested this circuit.
It is stable to 3.5mK with a 20K change, so probably quite good. Also, it is linear and Linear.

http://cds.linear.com/docs/en/lt-journal/Thermoelectric_Williams_Mar02_Mag.pdf

Just FYI, that circuit is designed for laser diode control and isn’t suitable for control for what the OP wants without a lot of modifications.

You obviously haven't read it. loop compensation leads to AN89, where it revealed that the laser used has a temperature time constant measured in minutes. Also, the analog method is 2-3 orders of magnitude better at the temperature control. I have no idea, why would you need a micro with all the extra work, instead of changing an RC value, and looking at it with a scope. For worse results.

Wether it makes sense to use a digital control with an µC instead of analog control depends on the time constants involved. The still rather fast Laser with a few minutes time constant might still work better with analog control, but for a larger and thus slower system it gets more an more attractive to use digital control, as the long time constants are difficult to to implement in analog hardware. Also with digital control anti windup  is usually standard  because it is easily implemented, while analog anti windup is tricky, especially if low drift is needed.  Anti_Windup can be important if the limits of the heater  / cooler are reached. Even for time constants in the minutes range, it takes quite some extra effort for the analog implementation - µCs and ADCs got better since AN86, so things have changes since than.

Thanks! That sums up part of what my reply was going to be.

I have read AN86, but like Kleinstein pointed out, things have gotten better since that app note came out. Essentially, it takes a lot of trial and error to stabilize an analog control loop over long time constants. It’s much easier to modify the Kp, Ki and Kd values in firmware on a uC. Not to mention the fact that you can get PID libraries with auto tune functionality that will, at the very least, give you sane starting values to hand tune.

If I needed precise 0.005c control of the temperature of a laser in a $500,000 piece of equipment I was building, sure, I’d go with a fully analog solution (both the control loop and the TEC driver) and use a DAC to control the setpoint.

Or, if I only needed 0.01c control, I might use an analog control loop and filtered PWM drive of the TEC. (In fact, Linear makes a part that essentially takes an analog voltage on the input and drives four MOSFET’s in an H-Bridge configuration with high speed PWM on the output. It’s designed specifically for TEC control.)

However, you can get easily get 0.1c stability with a MCU and off the shelf H-Bridge using filtered PWM drive of the TEc.  0.05c or better is achievable if you throw in a 12 or 14-bit DAC and go with a two power amps in a bridge configuration.

Keep in mind, analog drive isn’t really feasible for TECs over 5V@2A. That’s OK for laser diodes, since most TECs coupled to them fall into that range. However, the OP may want larger TECs for his enclosures, in which case analog drive becomes harder. (You really need big heatsinks and forced air cooling for the power amplifiers or discrete transistors.)

Also, for what the OP is doing 0.1c stability should be more than enough. So yeah, I’d much rather use a digital solution, as I could implement it a lot quicker. It would also scale to different sized enclosures a lot better than a purely analog solution would.

[timb]

In my stash of bits I have a Willhi Peltier controller but a more dedicated one can get made at some stage. I haven't used this model before but their standard 240V fridge controllers work well. I will rig a couple of RTD sensors across the enclosure and do some playing on the bench. Bigger Peltier modules most likely needed?


http://willhi.com/product_detail.php?product_id=115&lang=en

Keep in mind that’s not a temperature controller but a thermostat. It uses a relay and Bang On/Bang Off type control. Taking into account the built in hysteresis and overall accuracy, it’ll keep the temperature +-2c of the set point, absolute best case.

[beanflying]

Thanks TimB, I got it at the time more for interest as I have a couple of Peltier based fridges I use for my Coffee 'day job'  but I have gone down a Arduino/RTD path for those

Hysteresis is adjustable down to 1C and an Offset to the NTC is built in to the firmware. But as you say commercial fridge applications only.

The driver Chip is badged WH7016K and from memory the front board is the same as their other controllers so simple plug in to their stock controller/display board and a different output board to allow for the invert switching relays.

[timb]

Thanks TimB, I got it at the time more for interest as I have a couple of Peltier based fridges I use for my Coffee 'day job'  but I have gone down a Arduino/RTD path for those

Hysteresis is adjustable down to 1C and an Offset to the NTC is built in to the firmware. But as you say commercial fridge applications only.

The driver Chip is badged WH7016K and from memory the front board is the same as their other controllers so simple plug in to their stock controller/display board and a different output board to allow for the invert switching relays.

Yeah, they’re great for basic temperature control, they’re just not meant for precision. And you’re right, that one does specify 1c hysteresis (though you still have to add thermal lag between the TEC and sensor causing overshoot into the stability figure). I must have been thinking of the specs of a similar looking unit from eBay that I’ve worked with in the past.

[doktor pyta]

For heating-only application You may consider using self adhesive heater which may result in better temperature homogenity:
https://www.tme.eu/gb/details/fg12v_35w/heating-elements/

or similar

[beanflying]

Depending on whats in the box low power heating pads internally make good sense I think. To my way of thinking the benefit of a stable temperature internally somewhere above a 'normal' ambient is a more simple option than needing to heat and cool. Also means no need for fans or heatsinks for Peltier modules.

In my case around 30-35 C fixed would do nicely, any hotter and it's time for a beer

[cdev]

Whatever you do to increase insulation and thermal mass around your application (you could use foam to create a cooler like box, the outer of perhaps two thermally insulation layers)  might be an easy thing thats not likely to create any new problems. So much preferable to heating and cooling its worth increasing insulation and thermal mass to a substantial amount before adding minimal amounts of heating, IMHO.
 
Thats just my thought, not an expert in this stuff.

[beanflying]

Plastic Pelican style case with the Aluminium heated boxes inside surrounded by pick and pluck foam might do the insulation job or even pluck some more and run blue foam outers for the aluminium?

I have quite a few 5&6S 5000 LiPo's (beyond R/C use) around to run the box too.

[timb]

Plastic Pelican style case with the Aluminium heated boxes inside surrounded by pick and pluck foam might do the insulation job or even pluck some more and run blue foam outers for the aluminium?

I have quite a few 5&6S 5000 LiPo's (beyond R/C use) around to run the box too.

That pick and pluck foam isn’t a great thermal insulator. I’d look for small styrofoam coolers (designed for shipping stuff with the addition of ice packs) which you can find cheap on Amazon in a myriad of sizes.

You can also buy styrofoam sheets of various sizes that could be cut and glued together to make a form fitting insulated case for your metal boxes. (You really need a hot wire cutter to cut them properly, however you can rig one up with some PVC piping, metal fittings and nichrome wire; I made one up for $10 in plumbing supplies! Just hook it to a bench power supply and you’re good to go.)

[beanflying]

I have a hot wire cutter and blue foam from building R/C wings so a custom foam box is easy. Depron in 2,3 and 6mm is worth a look for circuit board insulation as it cuts clean with a sharp knife and is a more closed cell than white styrofoam.

I have seen EPP or similar being used on some of the threads here which isn't a good insulator but it is certainly the most abuse proof and mechanically shock proof to keeps boards away from mechanical shock.

[Kleinstein]

When using a peltier element for control, there is no need to do a lot of insulation. The TEC will give quite some thermal coupling from the outside to the inside. It still makes sense to have the rest of the case with a thermal resistance that is lower than the TEC internal one, but no need to go very much below that. Some insulation is good, but no need to overdo it.

As much of the heat-flow in will also be through the TEC, it really makes sense to measure the temperature on both sides of the TEC and not just the inside.