Interesting low-cost temperature controller we tested (~1 mK stability)

[7k7k]

Hi everyone,


We’ve been doing some temperature stabilization experiments recently for laser diode and optics setups.


As many of you know, temperature drift can easily affect wavelength stability and optical performance.


During some testing we came across a surprisingly capable temperature control setup based on TEC control and high-resolution sensing.


Under stable lab conditions we were able to get temperature stability close to ~1 mK (around ±0.001 °C), which was honestly better than we expected for a relatively simple and low-cost controller.


It made me curious about what people here typically use for temperature stabilization in laser or photonics systems.


For example:


• laser diode stabilization
• nonlinear crystal ovens
• spectroscopy experiments
• detector cooling


Are most people still working around ~0.01 °C stability, or are millikelvin systems becoming more common now?


Also curious what sensors people prefer for this level of control (thermistors, RTDs, etc).


Would love to hear about your setups and experiences.

[7k7k]

While looking into TEC control solutions I actually came across a controller from SenseFuture that claims something like this:


1. Temperature measurement standard deviation around 0.2 mK, with <1 mK drift over 24 hours.
2. Temperature control stability roughly ±0.001 to ±0.01 °C depending on the setup.


The surprising part is that the price is only around $326.


That made me a bit skeptical honestly — achieving millikelvin stability with a relatively low-cost controller sounds almost too good to be true.


Has anyone here actually tested controllers in this price range that can reach ~1 mK stability in real optical setups?


This is the page where I saw the specs:
https://www.sensefuture.com/

[TheHermit]

These ColorDrive & TECDrive are the best solutions that I'm aware of (w/0 Kvant experience caveat). Purchased a complete setup (including RGB Sparrows, EMS8000 galvos, optics & mounts) last year for a DIY projector build that's still sitting on the back burner.
LiveLasersystems online customer service chat is well monitored, human, very knowledgeable and helpful. CEO Phillip personally tuned my Colordrive One w/TEC drivers to the specs of the diodes I had selected.
However upon arrival in the USA, Fedex customs asked, "Do they emit radiation?" Then, blocked my parcel until I submitted a FDA-2877 form, then added $200 fee on top of my $2.8k order. That was while tariffs were in limbo.
Hope this helps.
Roj

[mixedgas]

Take a look at:

A microcontroller-based driver to stabilize the temperature of an optical stage to within 1 mK in the range 4–38 C, using a Peltier heat pump and a thermistor sensor

Instrumentation and Measurement Science

https://doi.org/10.1088/0957-0233/7/11/015

By Bill Sloman

Kind Regards,

Steve

[7k7k]

That’s very interesting—I’ve encountered similar TEC power consumption issues myself, particularly when using larger modules.


Our testing revealed that exercising more precise current control—rather than simply relying on a current-limiting resistor—and optimizing the PID loop helps reduce both power consumption and temperature overshoot.


I actually came across a compact TEC controller that excels at regulating current and maintaining stability, thereby simplifying the management of external power supplies and thermal dissipation.


Just out of curiosity—are you looking to stabilize the temperature of a laser diode, or perhaps another type of optical component?

[VDX]

... some +20 years ago I was co-developing high accuracy temperature sensing and controlling modules too -- the electronics was based on the aduc847 - https://www.analog.com/media/en/tech...45_847_848.pdf

It's a 24Bit ADC, where we used 22Bit resolution to measure/regulate against PT100, PT1000 or NI200 sensors.

We could sense down to 0.1 DegC and regulate our modules to 0.3 degC around 60 DegC for the housing and 120 DegC for a special membrane to get consistent readings for gas - thermal conductivity sensors.


Here the company-link and to the high resolution thermal conductivity modules -- https://messkonzept.de/en/

[mixedgas]

OK, It was on Academia.edu

See Attached

[7k7k]

modulate temperature via PWM + MOSFET

That’s a really interesting approach.


If you're open to alternatives — a compact TEC controller might save you quite a bit of effort compared to PWM + MOSFET.


I’ve tried one from https://www.sensefuture.com/ and it handled current control and stability quite well, especially for small TEC setups.
b59d1be886eaf65e5dd431aa49c4979c_origin.png

[7k7k]

... some +20 years ago I was co-developing high accuracy temperature sensing and controlling modules too -- the electronics was based on the aduc847 - https://www.analog.com/media/en/tech...45_847_848.pdf

It's a 24Bit ADC, where we used 22Bit resolution to measure/regulate against PT100, PT1000 or NI200 sensors.

We could sense down to 0.1 DegC and regulate our modules to 0.3 degC around 60 DegC for the housing and 120 DegC for a special membrane to get consistent readings for gas - thermal conductivity sensors.


Here the company-link and to the high resolution thermal conductivity modules -- https://messkonzept.de/en/

This is actually quite similar to what I came across — but implemented in a much more compact and integrated way.


I’ve been using a TEC controller from https://www.sensefuture.com/ that achieves similar (~mK-level) stability, but without the need for complex PWM + MOSFET + tuning setup.

[7k7k]

Sincere thanks to all for the helpful suggestions.

The fully assembled product's PCB + connectors + aluminum case + cooling fan(s) costs less than the $168 SenseFuture High-Precision Digital Temperature Controller.

Roj

That makes total sense — your approach is definitely more cost-effective, especially if you're comfortable building and tuning everything yourself.


From what I’ve seen, achieving ~1 mK stability is possible with a well-tuned DIY setup like yours, but it usually takes quite a bit of effort (PID tuning, dealing with thermal lag, noise, etc.).


For me personally, I was more focused on the optical side of the project, so I didn’t want to spend too much time optimizing the temperature control loop.


That’s where a dedicated TEC controller made sense — it’s definitely more expensive, but the stability and ease of use were noticeably better out of the box.


So I guess it really comes down to whether you want to optimize for cost or for time / convenience.