Maker Pro
Maker Pro

Driving Peltier Device through L298N to control Temperature

sachin2611

Sep 10, 2013
9
Joined
Sep 10, 2013
Messages
9
Hi,

I want to drive peltier element to control temperature of one module. Peltier specifications are:
Imax = 9A and Vmax = 3.75V.

Such 4 peltiers are connected in series to control to module temperature.

I am planning to use L298N for driving it. As shown if I connect L298N two pins in series then I will be able to drive current upto 4A. But, to control my module temperature, Peltier is sinking average 4A current.
So, It I can't use single L298N IC as it will heat too much and also it will damage.

So, I want to know whether it is possible to use two L298N ICs in parallel to give higher current.
If so then please advice me for input and output pin connections.

Thanks & Regards,
Sachin Panchal
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
There are several reasons why your idea won't work. Mainly, because you can't drive Peltier devices using a pulse-width-modulated signal because they are not ohmic in any sense; their voltage vs. current graph is nothing like a resistor.

I haven't investigated whether you can parallel L298N devices up, but I doubt it. It is not important in this case because you can't use PWM drive anyway.

To drive four 3.75V 9A Peltier devices in series, you need a 15V DC power supply that can supply 9A, i.e. a 135W power supply. Linear power supplies at this level are very bulky and wasteful of power, so a switching supply is the obvious choice. You will also probably want to use current regulation instead of voltage regulation.

At this level of power you should run your switching supply directly from the AC mains. So you need a line-powered switching supply that can provide 15V DC at 9A (135W output) with current regulation.

Personally I would modify a PC power supply for this application. This involves rewinding the secondary side of the transformer and modifying the feedback circuitry. There are articles describing how this can be done - Google those keywords (modify PC power supply rewind transformer secondary).

They may not cover converting voltage regulation to current regulation. To do this you need to sense the output current across a low-value shunt resistor and use the voltage as the source of the feedback signal. This would be easier with a power supply that has the control circuitry on the secondary side with isolated drive to the switching device(s).

If you want to design something from scratch, have a look at line-powered switching regulator ICs. Download the data sheets; they usually have several suggested application circuits. You will need to wind your own transformer.
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
There are several reasons why your idea won't work. Mainly, because you can't drive Peltier devices using a pulse-width-modulated signal because they are not ohmic in any sense; their voltage vs. current graph is nothing like a resistor.
Kris,

I am curious as to why not? Certainly at a low frequency (sub 1 Hz) you could use PWM to control the temp, no?

Bob
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
You can turn the current source ON and OFF using PWM but you can't use PWM in the usual way; for example if the Peltier devices need 15V DC (there are four of them connected in series, and each requires 3.75V DC), you can't use a 20V DC supply and PWM it at 75% to get an equivalent heating or cooling effect because the Peltier devices aren't ohmic.
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
Sure, but people control the "brightness" of LEDs using PWM as well, which is the same thing.

Bob
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Yes, once the LED is converted into a voltage-controlled component using a current limiting resistor or a linear constant current driver, neither of which is practical at 9A!

A high-power current-regulating power supply is required simply to drive a Peltier device. That power supply's output current could be set to the maximum allowable Peltier device current then controlled with a PWM Enable signal, or the current could be varied to set the amount of heating/cooling. The latter is a better option IMO. But an L298 is not part of either method.
 

Fish4Fun

So long, and Thanks for all the Fish!
Aug 27, 2013
481
Joined
Aug 27, 2013
Messages
481
I have played with TECs a fair amount, amazing devices; however, they are lossy beasts! They are at their best in cooling applications where the target temperature is greater than ambient, for instance CPU cooling. They are at their worst when attempting to cool something below ambient. If you look carefully @ the specifications you will note that maximum efficiency occurs as the temperature difference between the hot/cold sides approaches zero. As the temperature difference between the sides increases, the efficiency declines pretty rapidly.

As far as PWM, my experience is as Kris suggests: use a PC power supply. Additionally, I would suggest you do not operate the TECs at maximum; you will get much longer life out of them operating in the 70%-80% range. That said, 4 * 3.75 = 15V 15V * 80% = 12V. I would connect all 4 in series and connect them directly to the 12V rail of a PC supply.

Good Luck!

Fish
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Peltier devices are current-driven, and running them at 80% of their rated VOLTAGE will not necessarily cause 80% of the rated CURRENT to flow, because of something called the Seebeck effect, that I know nothing about. The heating or cooling effect is specified based on the current flow, not the voltage applied. So the supply to the device should be controlled for current, not voltage.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Another issue is that apparently there are issues with using them with PWM.

I understand that you need to keep away from a particular frequency band (i.e. be higher or lower than) to avoid mechanical stresses.

It seems you have quite some latitude:

41. Can I use pulse-width modulation to control my Peltier device if I keep the voltage at VMax or below?

Yes, and this is one of the most electrically-efficient ways to control voltage to your device—although you must observe some precautions. As long as you keep the voltage at VMax or below, you will effectively pump heat whenever the duty cycle applies voltage to your system; when the power is turned off, the heat pumping will stop. By pulse-width modulating a suitable voltage, you can easily control the extent of heat pumping by simply varying the duty cycle of the pulses. The great thing about this approach, is that it allows you to minimize power dissipation in your control circuit—especially if you use power MOSFET's for switching (a subject which goes beyond this particular question).

Significant precautions must be employed with PWM, however. First of all, the PWM should be at a high enough frequency to minimize thermal stresses to the TE devices. While we like to keep the frequency in the low killihertz (Hz) range, in many applications these days we must compromise at around 120 Hz for the sake of electromagnetic compatibility. Another important issue is the potential for generating electro-magnetic interference (EMI) in the wiring to the TE device. If you are using PWM, you may need to shield your power wiring or keep it away from any sensitive electrical signals. A stitch in time to confront these issue early, can save a lot of corrective work deeper in the design cycle.
From here.
 

sachin2611

Sep 10, 2013
9
Joined
Sep 10, 2013
Messages
9
There are several reasons why your idea won't work. Mainly, because you can't drive Peltier devices using a pulse-width-modulated signal because they are not ohmic in any sense; their voltage vs. current graph is nothing like a resistor.

I haven't investigated whether you can parallel L298N devices up, but I doubt it. It is not important in this case because you can't use PWM drive anyway.

To drive four 3.75V 9A Peltier devices in series, you need a 15V DC power supply that can supply 9A, i.e. a 135W power supply. Linear power supplies at this level are very bulky and wasteful of power, so a switching supply is the obvious choice. You will also probably want to use current regulation instead of voltage regulation.

At this level of power you should run your switching supply directly from the AC mains. So you need a line-powered switching supply that can provide 15V DC at 9A (135W output) with current regulation.

Personally I would modify a PC power supply for this application. This involves rewinding the secondary side of the transformer and modifying the feedback circuitry. There are articles describing how this can be done - Google those keywords (modify PC power supply rewind transformer secondary).

They may not cover converting voltage regulation to current regulation. To do this you need to sense the output current across a low-value shunt resistor and use the voltage as the source of the feedback signal. This would be easier with a power supply that has the control circuitry on the secondary side with isolated drive to the switching device(s).

If you want to design something from scratch, have a look at line-powered switching regulator ICs. Download the data sheets; they usually have several suggested application circuits. You will need to wind your own transformer.


Hi KrisBlueNZ,

Thanks for your response.
I already had driven single peltier device with PWM using L298N. But, As I mentioned I need more current I have to do something else.
And also 3.75V and 9A are maximum rating of petlier means be of safer side I must take precaution that voltage drop should not exceed 3V and 6A for single peltier.

Also, I want module not depedent on PC power supply so, advice me for another option.

Thanks,
Sachin
 

sachin2611

Sep 10, 2013
9
Joined
Sep 10, 2013
Messages
9
I have played with TECs a fair amount, amazing devices; however, they are lossy beasts! They are at their best in cooling applications where the target temperature is greater than ambient, for instance CPU cooling. They are at their worst when attempting to cool something below ambient. If you look carefully @ the specifications you will note that maximum efficiency occurs as the temperature difference between the hot/cold sides approaches zero. As the temperature difference between the sides increases, the efficiency declines pretty rapidly.

As far as PWM, my experience is as Kris suggests: use a PC power supply. Additionally, I would suggest you do not operate the TECs at maximum; you will get much longer life out of them operating in the 70%-80% range. That said, 4 * 3.75 = 15V 15V * 80% = 12V. I would connect all 4 in series and connect them directly to the 12V rail of a PC supply.

Good Luck!

Fish

Hi Fish,

Thanks for response.
As you mentioned I am planning to drive 4 series peltier with maximum 6A of current and max. single peltier voltage drop 3V.

I am planning to drive peltiers with micro controller. So, most probabilily I will go with PWM.

Also, Is there any other option than PC supply. As I want to use whole module independent of PC.
 

sachin2611

Sep 10, 2013
9
Joined
Sep 10, 2013
Messages
9
Another issue is that apparently there are issues with using them with PWM.

I understand that you need to keep away from a particular frequency band (i.e. be higher or lower than) to avoid mechanical stresses.

It seems you have quite some latitude:


From here.

Hi Steve,

Thanks for response,
Yes, I am planning to drive Peltiers with 250Hz PWM.
But, I want that drive 4 peltier in series such that individual voltage drop does not exceed 3V and series current does not exceed 6A.

Also, As module for which I am controlling temperature is big. So, to get required temperature I want to give average of 5A current so that It will cool module within 15-20 minutes.

Thanks,
Sachin
 

sachin2611

Sep 10, 2013
9
Joined
Sep 10, 2013
Messages
9
Update: Driving Peltier Device to control Temperature

Hi,

One more requirement is added to my project and that is to make temperature controller such that instead of 4 series peltier if I connect one, two or three then also I should work.

So, Old idea to drive it with direct 12V such that 4 peltiers have drop of 3V will not work.

Means, My new requirement is to fix voltage below 3V for any number of peltier and current up to 6A.

Please suggest me.

Thanks & Regards,
Sachin
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
It has been mentioned earlier that you should consider a variable current source rather than a variable voltage source.

If done correctly this will work for 1, or 4 devices.

One option is to use a PWM circuit to drive the devices via an inductor and some other components... If you limit the ON time to that required for the current to rise to a particular set value then you essentially have a variable constant current source.

Depending on the frequency and value of the inductor, it will operate in either continuous or intermittent mode (either is probably fine for this)
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
I think I will stand by my recommendations from post #2 in this thread. That is a power supply that delivers a constant regulated current to the Peltier devices.

If you don't want to modify a PC power supply, you may need to build your own switching power supply - either line-powered, or powered from another supply, if you have one - that can deliver a variable current up to 9A with an appropriate voltage range - at least 3~12V.

You still haven't told us what kind of power source you intend to use as the DC supply for your Peltier devices.

You can use PWM from your MCU to generate a control voltage to set the current, and run a steady current through the Peltier device. This can be done as part of the feedback loop for a line-powered power supply.

Perhaps you have a specific reason why you want the Peltier current to pulse ON and OFF. I don't think that would be a good idea, but if you really want to do it that way, please explain why. And read the reference that Steve linked in post #9 on this thread.

Steve's latest suggestion is good too. Use a DC supply of, say, 20V and a switching MOSFET, catch diode, and inductor in a standard buck regulator configuration with the Peltier device(s) across the output. You can then drive the MOSFET (via a MOSFET driver) from the PWM output of your MCU (depending on the PWM frequency, you may need quite a large inductor). You need to monitor the Peltier current with the MCU to control the PWM properly. This would require some control loop design in your firmware and would give you the same effect as I described, but using the inductor as the current limiter. Good idea Steve!
 

sachin2611

Sep 10, 2013
9
Joined
Sep 10, 2013
Messages
9
Hi Kris and Steve,

Thank you very much for your valuable support.

I am planning to use DC power supply of 12V, 15V or 20V with current capability above 6A.

Actually, I am fresher and I was aware about only one method to drive peltier with variable bidirectional current is that use PWM from MCU with feedback and drive H bridge.

But, what I was worried that how I will drop supply 12V DC voltage to constant 3V across peltier.

I will look at standard buck regulator method as mentioned by you.
Please suggest me for some references.

Thanks,
Sachin
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
I assume that by bi-directional current, you mean you want to be able to vary the current from (say) +9A to -5A, not provide it with AC.

The former allows variable heating and cooling, the latter is just silly.

Also you need to be careful that near the switching point (assuming you have some sort of thermostat) that you don't switch from heating to cooling to heating to cooling... That would also be bad.

You can separate the (variable) constant current source from the H Bridge so you can control direction and magnitude.

For various reasons, you'd probably want to turn off the constant current source for some time (maybe a few milliseconds) before you use the H Bridge to change the direction of the current. In practice, I'm not sure you'd want to change from heating to/from cooling in a particular hurry anyway.
 

sachin2611

Sep 10, 2013
9
Joined
Sep 10, 2013
Messages
9
Hi steve,

Its too good.
Yes, I must have to turn off the constant current source for some time before changing the direction of the current.

Thanks,
Sachin
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Sure, you can use an H-bridge to switch the polarity of the current. I would not use an L298N for this; use some big grunty N-channel MOSFETs with an H-bridge driver IC. As you said, you should power down the circuit while the polarity is being reversed.

The H-bridge would be a separate stage that connects between the power supply and the Peltier device. You still need a power source that can deliver a voltage over a range of at least 3~12V and current up to 9A.

As I've said twice before, I think the best solution would be a mains-powered ("off-line") power supply with the feedback loop used to regulate the output current. The next alternative would be a DC power supply of around 30~35V rated for 5A output current, running through a buck converter used as a variable current source.

The output current could be controlled using a PWM or DAC output from the microcontroller, and the H-bridge would also be controlled with a microcontroller output.

This would give a steady current through the Peltier device, in either direction. The amount of current would be set by the PWM or DAC output from the microcontroller, and current regulation would be done by the buck converter.
 

sachin2611

Sep 10, 2013
9
Joined
Sep 10, 2013
Messages
9
Hi Kris,

I think now idea is getting clear to me. Now, I will go for schematic and then will ask you for reviews and suggestions.

Thanks,
Sachin
 
Top