Author Options:

Limiting current? Answered

Good day,

I am interested in an effective way of limiting current for a TEC application. This application involves using the TEC's (12V @ 1A) to reduce the incoming air temperature in a noise isolating enclosure from ambient temp to about  5-10 deg C cooler. In doing some research on implementing the TEC's I have found that these units operate at their greatest efficiency around 1/3 of max amp input and lower when you do not have a large heat load to "pump" away. So all of this being said I would like to operate a number of TEC's at  this low amperage range, and to be able to cap their maximum current draw between .1 and .3 amp. 

Any input would be greatly appreciated, thanks



2 years ago

Thank you all, for the helpful information. I have just joined instructables and appreciate the speed and quality of information shared.

So... At this point my brand new variable voltage/amperage DC lab power supply has shown up and I am definitely looking forward to experimenting with it and various setups with the TECs I already have on hand.

With respect to the current regulation, I have also just received my order of logic level mosfets which I had been considering using for current regulation in conjunction with the TECs. The information provided on "constant current source and active current source circuits" was fantastic, definitely provided a more detailed understanding of how to properly use the mosfets than I previously had, and has joined my bookmark collection.

My remaining question is, for a long term application, not just an at home project, what would be the best and most appropriate way to regulate current for such an application. I have a feeling that iceng is sitting on a golden nugget there, although I can imagine much more involved...

What has me looking for a different(or more involved) solution is that from my understanding so far, although it sound like transistors or specifically mosfets for this application are an effective way to control current on and off, and throttle to a certain degree, they do not perform as well when in a constantly throttled state. Also from what I understand, they're operating at an inefficient range when throttled, and hence produce much more heat. Additionally I was also under the impression that there would be a voltage drop too when achieving amperage regulation in this way.

So please correct me if it sounds like I have come to any misunderstandings in anything I have described, I definitely don't want to overlook a simple solution that would be appropriate. That being said I have had a look at some TEC controllers online, and although I cant really tell much from the pictures I've seen I have a feeling it would be a fair bit more than even the active current source curcuit provides.

Just to share the info, I figured I'd pass on the site I had found that is very useful for properly calculating TEC specs to thermal loads, and goes over some of the peek efficiency info I had briefly touched on in my original question.


Thanks again for all the helpful info.


2 years ago

Might be badly old school but since you will waste the energy into heat anyway:
Option 1: Take some car break lights in parallel to reduce the available power.
Halogen bulbs work too.

Option 2: Salvage a heating element from an old toaster - they are great when you only need a few Ohms.

Option 3: Wind your resistor onto some ceramic insulator using Nichrome wire.
With a clamp you can even create an adjustable resistor if your tube is long enough.

Option 4: Use a Buck Converter that can be adjusted for both voltage and max current.

Jack A Lopez

2 years ago

I suggest you use a bench type power supply, the kind with some knobs on the front for adjusting the voltage and current limits, plus little indicator lights labled "CC" and "CV" for to tell you which limit it's bumping against. CC and CV stand for "constant current" and "constant voltage". Also this type of power supply comes with some kind of numerical display for to tell you both how much voltage (in volts) is across your load, and how much current (in amperes) is flowing through it.

The way this kind of power supply works can be understood graphically. I mean, you draw a plot of I-V curve,


for your load, and then draw a horizontal line (V = Vlimit) and a vertical line (I=Ilimit), and the actual kind of regulation you get (either constant voltage or contstant current) depends on which line intersects the load's I-V curve.

I was looking for a page that explains this mojo, and this one,


gives a decent explanation of how this works.

The cost of this kind of bench power supply tends to scale with the amount of physical power (in watts), equivalent to its maxium voltage multiplied by its maximum current, it can supply.

A possibly cheaper version exists, in the form of "adjustable CC/CV" power converter boards, like those sold on eBay. Typically those are DC-to-DC, meaning they want input power as DC. Actually there might also exist versions of these, intended to be powered by AC mains power.

The reason these power converter boards are cheap, is because its kind of a "bare bones" kind of deal. That is to say there is typically no numeric display, to tell you how much current, or voltage, is going to the load. Also instead of knobs, you get tiny little pots on the board itself, made to be turned by a screwdriver.

Either a bench type CC/CV power supply, or the cheaper single board equivalent of the same thing, I think both of those would be less hassle than the work involved in actually building an adjustable power supply from scratch.

Final note: If you have several of these thermoelectric modules, and you want them to all share the same current, then the easiest way to do that, is to wire them all in series. You know, because circuit elements wired in series share the same current. The voltages across each element add together.

For example, if one module has a voltage of 8 V across it, at 0.1 A, then the voltage required to push 0.1 A through five of them wired in series, is 5*8 = 40 V.

If you can find in documents, or discover via experiment, the I-V curve for your thermoelectric modules, this will help your design efforts. I mean, if you can know, prior to building the thing, approximately how much voltage and current it is wanting.


Answer 2 years ago

+1 on the simple ckt you found, for a newbie who wouldn't understand a 317

or best answer.