Maker Pro
Maker Pro

Need help with design issues - resistor questions

IdahoCowboy

Jul 7, 2015
4
Joined
Jul 7, 2015
Messages
4
Hi all, thank you for your help up front - i have been beating my head on my workbench for a couple weeks now. I am new to the electronics world, but I have tried to do as much research as I could.

I have a couple questions;
1: I have a 48 V power supply (12.5 A). Its feeding a number of thermoelectric Peltier plates (see image of schematic), each plate draws 14.4 Vmax and 6.4 Imax. I need the top layer of plates to be colder then the bottom layer, thus why i have two pairs wired in series, then a group of 4 wired in series. By my math using Ohm's law, the resistors I need for the first two pairs is : (14.4 * 2) = 28.8. 48V -28.8V = 19.2V 19.2V / 6.4a = 3Ω ....? right ???

so my understanding is that the first two pairs need a 3Ω resistor. Now, with 48v power supply feeding 12.5amps I connected a 25Ω potentiometer to test this after adjusting it down to around 4 ohms resistance...and I fried my pot. So that made me question if at 1/4 or 1/2 watt resistor will do the job. I tried looking this up in the forums here but could not find a definite answer. Will a 1/2 watt resistor do the job for the current i am putting through it?

Question 2: should I be wiring the first two pairs of plates in parallel instead of in series? I have enough current to supply two disks in parallel, but am not sure if this is an advantage for me or not.

Thanks again for any feedback / suggestions
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,739
Joined
Jun 21, 2012
Messages
4,739
... By my math using Ohm's law, the resistors I need for the first two pairs is : (14.4 * 2) = 28.8. 48V -28.8V = 19.2V 19.2V / 6.4a = 3Ω ....? right ??? Right. So far, so good.

so my understanding is that the first two pairs need a 3Ω resistor. Now, with 48v power supply feeding 12.5amps I connected a 25Ω potentiometer to test this after adjusting it down to around 4 ohms resistance...and I fried my pot. So that made me question if at 1/4 or 1/2 watt resistor will do the job. Good question! Did you multiply the square of the current by the resistance to find out how much power is dissipated in the 3 Ω resistor? Or, alternatively, you calculated 19.2 V at 6.4 A across your 3 Ω resistor... that's about 123 watts.

I tried looking this up in the forums here but could not find a definite answer. Will a 1/2 watt resistor do the job for the current i am putting through it? No.

Question 2: should I be wiring the first two pairs of plates in parallel instead of in series? If you do this, then 48 V DC is applied to the module which is way beyond its maximum rating. You could wire four in series as shown in the bottom half of your schematic. I have enough current to supply two disks in parallel, but am not sure if this is an advantage for me or not. Your cold plates will work over a range supply voltage, up to 14 V DC maximum. The current will be proportional to the supply voltage. Wiring four in series, assuming equal voltage drop across each one, means your 48 V DC supply will provide each module with 12 V DC without the need for any resistors.

Thanks again for any feedback / suggestions
Peltier coolers are wonderful toys. The model you are using can be safely connected directly to the 12 V DC electrical system of an automobile without the need for any "current limiting" resistors. However, since are using a 48 V DC electrical system, you must be careful that each Peltier device has less than its maximum rated voltage applied when connected to that power source. The easiest way to do this is to wire several of them in series. Assuming all devices are identical, in series each will drop the same voltage. So, four in series will each drop 12 V DC from your 48 V DC supply.

Wiring only two in series means you need some way to limit the current through each device. A series resistor will work, but it wastes a huge amount of power. A better solution is to use a pulse-width-modulated power source. This can be your 48 V DC supply applied through power MOSFET transistors to apply a variable duty cycle to the Peltier devices, thus limiting their average current to the maximum amount the datasheet allows. It would be better to connect three devices in series, rather than three or four in parallel, to a PWM 48 V DC supply.

I have attached the datasheet for the TEC1-12706. Study this to learn the limitations of the device.
 

Attachments

  • TEC1-12706.pdf
    312.8 KB · Views: 115

IdahoCowboy

Jul 7, 2015
4
Joined
Jul 7, 2015
Messages
4
Peltier coolers are wonderful toys. The model you are using can be safely connected directly to the 12 V DC electrical system of an automobile without the need for any "current limiting" resistors. However, since are using a 48 V DC electrical system, you must be careful that each Peltier device has less than its maximum rated voltage applied when connected to that power source. The easiest way to do this is to wire several of them in series. Assuming all devices are identical, in series each will drop the same voltage. So, four in series will each drop 12 V DC from your 48 V DC supply.

Wiring only two in series means you need some way to limit the current through each device. A series resistor will work, but it wastes a huge amount of power. A better solution is to use a pulse-width-modulated power source. This can be your 48 V DC supply applied through power MOSFET transistors to apply a variable duty cycle to the Peltier devices, thus limiting their average current to the maximum amount the datasheet allows. It would be better to connect three devices in series, rather than three or four in parallel, to a PWM 48 V DC supply.

I have attached the datasheet for the TEC1-12706. Study this to learn the limitations of the device.

Great info! thank you so much - I will post an update when i have it working.
 

Colin Mitchell

Aug 31, 2014
1,416
Joined
Aug 31, 2014
Messages
1,416
There is going to be an enormous difference in current taken by the panels when supplying 12v as opposed to 14v.
This is the FIRST thing you have to consider.
 

IdahoCowboy

Jul 7, 2015
4
Joined
Jul 7, 2015
Messages
4
Great info! thank you so much - I will post an update when i have it working.

Thanks Colin, that is an issue for this project as I need to try and Max out the current for the top array of peltier plates so they get as cold as possible.

Hevans1944 - I have done some research on the Pulse Width Modulation using a MOSFET transistor. This is a handy bit of skills that I can use on other projects - thank you again. I have some questions however. My understanding (summarized ) is that the PWM takes the avg from 0 to some maximum voltage - in my case 0 to 48 V to create a duty cycle (based on the frequency of this cycle) that gives my device a voltage of whatever I need. So if I need 14 volts, i would calculate the duty cycle to give me roughly 30% of the max voltage thus supplying the 14 that I need.

My concern is that if my peltier plate has a Vmax of 14.4 volts, is this cycling of 0-48V at a given frequency going to cause any damage to my device? I understand that it ends up getting an avg between the two - but I don't want to fry the peltier plate either.

My thoughts is to use an Arduino that has PWM management running through a MOSFET transistor to control each of the 4 top layer Peltier plates, then just connect the bottom array of plates in series (like in the schematic) because they dont need to be maxed out like the top array. Does this sound like I'm on the right path?

Thanks,
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,739
Joined
Jun 21, 2012
Messages
4,739
...

My concern is that if my peltier plate has a Vmax of 14.4 volts, is this cycling of 0-48V at a given frequency going to cause any damage to my device? I understand that it ends up getting an avg between the two - but I don't want to fry the peltier plate either. ...
That is a valid concern and I don't have an answer. Best bet would be not to exceed the maximum voltage that can be safely applied, whether PWM or not. Just to be safe. Or try it out on just one peltier plate at low duty cycle and 48 V to see if it "smokes" it.

... My thoughts is to use an Arduino that has PWM management running through a MOSFET transistor to control each of the 4 top layer Peltier plates, then just connect the bottom array of plates in series (like in the schematic) because they dont need to be maxed out like the top array. Does this sound like I'm on the right path?

Thanks,
That sounds like a good plan. You should also consider placing a large electrolytic capacitor across the peltier devices to integrate the PWM and avoid exceeding the 14.4 V DC maximum voltage specification. This will slow down the thermal response time, but that usually isn't a problem for cooling applications.
 
Top