Building a PWM filter for a Peltier

[arronar]

Currently, I managed to drive a Peltier element using the Syren 10A Driver using PWM from a raspberry pi. But because I want to set up a PID control later for its temperature, and the Syren 10A driver uses PWM to drive the Peltier, I read online that it could be good first to make a filter for the PWM.

The Peltier element that I use is either the ET-190-1010-1212-RS or the TES1-12704. By searching on the internet I found this picture describing a PWM filter:



The difference here is that in this schematic they use a MOSFET to drive the Peltier and the resistance of the Peltier itself.

So I was wondering if I can use this filter for my Peltiers or I should change some of the resistor/capacitor values and how one can calculate these.

Any help or idea will be appreciated.

[RoGeorge]

...I read online that it could be good first to make a filter for the PWM...

Do you have a link about that, please?

I see no reason why the unfiltered PWM wouldn't work.  The thermal inertia of the Peltier element will act as a filter anyway.

[arronar]

Here is a post from this same forum.

[Zero999]

Is it really necessary to filter the PWM? As long as it's above the thermal time constant of the Peltier element then I don't see why it would need to be filtered.

[arronar]

Is it really necessary to filter the PWM? As long as it's above the thermal time constant of the Peltier element then I don't see why it would need to be filtered.

If you are asking me, I don't know...

[mikerj]

Is it really necessary to filter the PWM? As long as it's above the thermal time constant of the Peltier element then I don't see why it would need to be filtered.

It hurts the performance of the TEC quite a bit since you are effectively driving it hard during the on period, so the I2R losses (which have to be pumped) are higher.  If the TEC is over sized for the job it may not matter, if you need to squeeze out the best performance it definitely will.

EDIT:  Look at the example given here for an idea of the efficiency loss with PWM vs DC, it can be huge.

[wraper]

Is it really necessary to filter the PWM? As long as it's above the thermal time constant of the Peltier element then I don't see why it would need to be filtered.

Because they drastically loose efficiency with increased power applied.

[mvs]

By searching on the internet I found this picture describing a PWM filter:


The difference here is that in this schematic they use a MOSFET to drive the Peltier and the resistance of the Peltier itself.

So I was wondering if I can use this filter for my Peltiers or I should change some of the resistor/capacitor values and how one can calculate these.

It is not a filter, it is just a basic buck voltage converter. You can use it, but you need calculate inductor value.

You can use this formula:

L=Vin*(1-D)*D/(2*dI*Fsw)

Vin - supply voltage
D - PWM duty cycle. 50% is worst case scenario.
dI - inductor ripple current. Typicaly 20-30% of load current is a good point to start. You will need low ESR capacitor with ripple current rating better then dI.
Fsw - PWM switching frequency.

For schematic in your first post i would calculate:
dI=20%*6V/1.5R=0.8A
L=12V*(1-50%)*50%/(2*0.8A*20000Hz)=93.75µH

[rhb]

Just put a largish capacitor on the PWM output, then monitor the capacitor voltage for the PID.  Do a search on using PWM for a DAC.

I'm working on the same thing, but it's dormant at the moment as I'm trying to root my Instek 2000E.

[Zero999]

Is it really necessary to filter the PWM? As long as it's above the thermal time constant of the Peltier element then I don't see why it would need to be filtered.

It hurts the performance of the TEC quite a bit since you are effectively driving it hard during the on period, so the I2R losses (which have to be pumped) are higher.  If the TEC is over sized for the job it may not matter, if you need to squeeze out the best performance it definitely will.

EDIT:  Look at the example given here for an idea of the efficiency loss with PWM vs DC, it can be huge.

Yes, that makes sense. Driving at 100% current, with 50% duty cycle will result in double the resistive losses of reducing the current by 50%. I forgot, it's not like a simple resistive heating device.

[arronar]

Hey. What if I don't use the inductor because I couldn't find one with 100uH/10A or 94uH/10A in the local store? Is it going to work?

[rs20]

Hey. What if I don't use the inductor because I couldn't find one with 100uH/10A or 94uH/10A in the local store? Is it going to work?

What do you mean "don't use the inductor"? If you just leave it out, the peltier can't be powered. If you replace it with a solid wire link, then it's just unfiltered PWM at which point you should get rid of the capacitor too (because it, and the MOSFET, are going to stress each other out a lot for no good reason.)

[arronar]

Yes, you are right. Excuse me for my noobish questions. Its the first time I'm trying to play with these ideas.
 I found a 10uH in my storage. What about using it?
Or else I should buy these

[wraper]

Yes, you are right. Excuse me for my noobish questions. Its the first time I'm trying to play with these ideas.
 I found a 10uH in my storage. What about using it?

Too small inductance unless PWM frequency will be at least a few hundred kHz

[arronar]

Too small inductance unless PWM frequency will be at least a few hundred kHz

Because I'm using the R/C mode (page 11) of the driver I suppose that the PWM frequency is at 50Hz.

Is there any other way to filter it? Another schematic probably.

[mikerj]

Too small inductance unless PWM frequency will be at least a few hundred kHz

Because I'm using the R/C mode (page 11) of the driver I suppose that the PWM frequency is at 50Hz.

Is there any other way to filter it? Another schematic probably.

From the second page in the manual "SyRen 10 and Syren 25 feature a PWM frequency of 32kHz".  In R/C mode the PWM duty cycle can only be changed at a 50Hz rate, but the PWM frequency itself is still high.

[chemelec]

Peltier devices work best with Pure DC, Not a high frequency PWM.
Most important to try and keep the Hot side, Not Hot.

Pulsing it On and Off at a Slow Rate can work quite good.
Example: 5 to 10 seconds on, 5 to 10 seconds off, ETC.

[rs20]

Peltier devices work best with Pure DC, Not a high frequency PWM.
Most important to try and keep the Hot side, Not Hot.

Pulsing it On and Off at a Slow Rate can work quite good.
Example: 5 to 10 seconds on, 5 to 10 seconds off, ETC.

So your advice is, "Don't do PWM, do PWM instead?"?

It's already been explained why any form of pulsing/PWM, regardless of the rate, is a bad idea. The heat transfer is proportional to the current but the resistive losses a proportional to the current squared (I^2*R), so feeding 2A at 50% duty cycle has the same heat transfer as DC 1A, but double the resistive losses. Given how significant the I^2*R losses are to the thermals of a peltier device, this is massively significant.

The OP is doing "filtered PWM", which is just a slightly odd way of wording "a buck converter". Which generates (practically) DC as its output. So please don't be derailing him/her with this incorrect advice.

[Zero999]

Peltier devices work best with Pure DC, Not a high frequency PWM.
Most important to try and keep the Hot side, Not Hot.

Pulsing it On and Off at a Slow Rate can work quite good.
Example: 5 to 10 seconds on, 5 to 10 seconds off, ETC.

So your advice is, "Don't do PWM, do PWM instead?"?

It's already been explained why any form of pulsing/PWM, regardless of the rate, is a bad idea. The heat transfer is proportional to the current but the resistive losses a proportional to the current squared (I^2*R), so feeding 2A at 50% duty cycle has the same heat transfer as DC 1A, but double the resistive losses. Given how significant the I^2*R losses are to the thermals of a peltier device, this is massively significant.

The OP is doing "filtered PWM", which is just a slightly odd way of wording "a buck converter". Which generates (practically) DC as its output. So please don't be derailing him/her with this incorrect advice.

Yes, there should be no reason why low frequency PWM is any better than high frequency PWM, apart from lower skin effect losses, which won't account for much unless yuou're talking about stupidly high frequencies.

Let's put some more figures into the equation.

Suppose the resistance, i.e. the part of the load which doesn't do any useful work, is 3Ohm and the full power current consumption at 12V is 2A.

Full power consumption:
P = I²R = 2²×8 = 12W

50% power consumption, using PWM.
P = 12*0.5 = 6W

50% power consumption, using half the DC current.
P = I²R = 1²×8 = 3W

Conclusion: reducing the current by half uses a half of the power of PWM at the full current.

[chemelec]

I have built many Peltier coolers and I find the best option is to monitor both the Hot and Cold side temperatures, using a comparator circuit to turn the peltier on and off. Set correctly you can get better efficiency.

[wraper]

I have built many Peltier coolers and I find the best option is to monitor both the Hot and Cold side temperatures, using a comparator circuit to turn the peltier on and off. Set correctly you can get better efficiency.

It's better to not keep it on/off at all but running constantly at minimum current required.

Set correctly you can get better efficiency.

It's a way to greatly compromise efficiency to begin with. And it does not make much sense financially too. You need to use larger heatsink, higher power PSU, control additional temperature sensor.

[Kleinstein]

....
Let's put some more figures into the equation.

Suppose the resistance, i.e. the part of the load which doesn't do any useful work, is 3Ohm and the full power current consumption at 12V is 2A.

Full power consumption:
P = I²R = 2²×8 = 12W

50% power consumption, using PWM.
P = 12*0.5 = 6W

50% power consumption, using half the DC current.
P = I²R = 1²×8 = 3W

Conclusion: reducing the current by half uses a half of the power of PWM at the full current.

PWM is even worse than that: the additional 3 W are dissipated in the Peltier element and this way effectively to half the amount at the cold side. So 50% PWM is not only using twice the power, but also has less cooling effect. It is especially bad if the current is high, near the maximum current.  As an extra point the added power makes the hot side a little warmer and this way reduces the a bit more.

It is allready more efficient to use a simple linear regulator or series resistor to reduce the current than to use PWM. It takes the same power, but at least that extra power is lost in the regulator or resistor and not in the Peltier element.

Slow PWM can be bad for the reliability as it causes a superimposed temperature cycling. Faster PWM at least reduces the temperature swing. So it is slightly better, but not much.