TEC cooled Diode testing station, suggestions?

[Uranium]

BB would This work?

Try these, i personally love them. They can handle so much power.
http://laserpointerforums.com/f60/check-out-these-mosfets-51381.html

 

[Meatball]

I like your PCB design...

For my testing station, I'm just going to include a current regulated and monitored power supply on board, and I'm just going to monitor the heatsink temp near the diode using a thermometer with a display. I'll then use an external trim pot to adjust peltier current by hand as needed.

This is something you might consider as well, I'm all about keeping things controllable, and simple.

DDL for the diode.
DDL for the peltier.
Power for the fan, and displays as needed.
Nothing fancy here!

 

[amkdeath]

I like your PCB design...

For my testing station, I'm just going to include a current regulated and monitored power supply on board, and I'm just going to monitor the heatsink temp near the diode using a thermometer with a display. I'll then use an external trim pot to adjust peltier current by hand as needed.

This is something you might consider as well, I'm all about keeping things controllable, and simple.

DDL for the diode.
DDL for the peltier.
Power for the fan, and displays as needed.
Nothing fancy here!

DDL for the peltier won't work, LM317 only does 1.5 Amps. Need 5Amps!

 

[amkdeath]

I'm finishing up the mark 3 board which includes the MOSFET regulator for the peltier.


EDIT: Getting 2 of the MOSFETs Uranium linked and 2 of the IRF540As. I'm getting 6 boards altogether, so I can test out different combos.

EDIT: here it is.


So this one incorporates the Super Simple Power MOSFET linear regulator as well as the LM317T for the Laser Diode, the 7805's for the Fan and the voltage regulators for the LCD's. Also has output for LCD Ammeter to show Current running through the LD, LCD Volt meter to show V across the LD. Has the MOSFET regulator for the peltier, complete with the PWM control. Also includes has through holes for switches for each and every device.

Now, what to do with the PWM? I understand it's designed to control Voltage, but a peltier can also be controlled temperature wise using variable voltage instead of current, right? And I am assuming something along the lines of an arduino will be needed to control the input of the PWM.

I got no input regarding the 0.58VDC Dilemma, so I just assumed 0.58VDC and calculated that two 1/4 watt 0.3Ohm resistors and a 3 watt 0.47Ohm resistor in parallel ( 0.114 Ohms) would yield a good 5.1Amps of Current, and with a 1.5V drop I can power the peltier off some ~14VDC Lithiums.

The Knimrod LPM would use two 9V batteries, the Peltier Heatsink fan uses 1 9V battery, the LD will use some lithium packs (around ~7-8VDC), the Ammeter LCD will use a 14VDC Lithium pack, the Voltmeter LCD will use a 9V and the Peltier will use ~14VDC lithiums.

The board measures ~ 2inches by 1 Inch

 

[Meatball]

DDL for the peltier won't work, LM317 only does 1.5 Amps. Need 5Amps!

Ha! Forgot to mention that my peltier is considerably smaller... It gets HOT at 800mA!

About the peltier control, I believe that peltiers are strictly current controlled. The PWM may be what you're using to control duty cycle. By controlling how long the thing is on/off, you can effectively determine how hot/cold the thing gets.

Or am I totally missing what you're trying to do? I'm trying to follow this thread, but I had to skip past all that arduino stuff..

 

[amkdeath]

Try these, i personally love them. They can handle so much power.
http://laserpointerforums.com/f60/check-out-these-mosfets-51381.html

Ha! Forgot to mention that my peltier is considerably smaller... It gets HOT at 800mA!

About the peltier control, I believe that peltiers are strictly current controlled. The PWM may be what you're using to control duty cycle. By controlling how long the thing is on/off, you can effectively determine how hot/cold the thing gets.

Or am I totally missing what you're trying to do? I'm trying to follow this thread, but I had to skip past all that arduino stuff..

Hmm, sounds viable, the original idea was to have a temperature sensor input to the arduino which would see the temp. of the LD and try to keep it at a set temperature. As Temp. went up, the arduino would send signals to whatever's controlling the Peltier to increase the current, and vice versa. But I guess with the PWM I can have it like home cooling systems, turn it on, once LD reaches like 10 degrees Celcius, turn off, then turn on again at like 15 degrees celcius.

 

[Meatball]

Hey go to the TEMPERATURE CONTROL section of this website, I found a VERY simple and easy way to maintain a set temp.

Put a couple of these circuits in, with rectifier diodes on the outputs of each, and you can have one put +current through the peltier at a set temp, and the other can be set to put -current through the peltier at a slightly different temp.

The Heatsink Guide - Peltier cooler information

 

[jbtm]

I have a 3x3 inch peltier...I never got it to frost over yet and I did 12V 10A to it...Heh...Even using a large 10X10X1" heatsink

Nice project though, hope it goes good

 

[amkdeath]

Hey go to the TEMPERATURE CONTROL section of this website, I found a VERY simple and easy way to maintain a set temp.

Put a couple of these circuits in, with rectifier diodes on the outputs of each, and you can have one put +current through the peltier at a set temp, and the other can be set to put -current through the peltier at a slightly different temp.

The Heatsink Guide - Peltier cooler information

Sounds awesome, but I figured if I am going to get an arduino to handle graphing and logging, why not just use it to control the peltier?

 

[Meatball]

Heh, if you can, go for it! Like I said, I prefer simplicity. I'm not arduino friendly, and would rather let the thing sit there and regulate itself.

 

[amkdeath]

Ok, here is mark 4



I added the 1 Ohm resistor for output to the arduino for logging V and A across/through the LD.

I also made it smaller, it is now 1.825" x 1.100", which will allow for easier spacing and cheaper manufacturing.

 

[jbtm]

Is true...I figured when I did water cooling systems using a peltier, I used a microcontroller that monitored the temperature. If it went 3 above, it would set the peltier to do a higher duty cycle, If it was 3 below the set, it did less duty cycle. Because I had a lot, 500W mosfet was used. Simple way I guess to control temp

 

[Bionic-Badger]

Sorry, I out for some time and didn't notice this thread. The 0.56-0.58V to 0.7V represents the linear voltage region for those BJT transistors. BJTs allow current to flow from the emitter to the collector based on if there's current going into the base of the BJT. They also need 0.7V from the emitter to base to turn fully on. That's what he's talking about with the 0.56-0.58V, and the resulting resistor needed at R2.

So say you need 5A regulation:

R2 = 0.58VDC / 5 = 0.116 ohm

Yes, it will need to dissipate up to 3W of power.

The 0.58V value will depend on the curves for your NPN transistor, but is probably right.

Another thing to try is to simply treat the TEC as a motor (as they're current-driven devices), and use, or make, an H-bridge (another example for an actual TEC) to control them. Here's a 9A pre-made one from Sparkfun, but they're not too hard to make with four power mosfets. For example, these are pretty cheap.

Since you already have/will have the Arduino, you can use it to do all the PWMing for the H-bridge.

The main source of heat on your mosfets used for an H-bridge will be from the switching. If they're in the full-on or full-off state they use virtually no power, but it's the transition period that they use the power. I think you'll be PWMing at a few Khz for the longevity of the peltier.

 

[amkdeath]

Sorry, I out for some time and didn't notice this thread. The 0.56-0.58V to 0.7V represents the linear voltage region for those BJT transistors. BJTs allow current to flow from the emitter to the collector based on if there's current going into the base of the BJT. They also need 0.7V from the emitter to base to turn fully on. That's what he's talking about with the 0.56-0.58V, and the resulting resistor needed at R2.

So say you need 5A regulation:

R2 = 0.58VDC / 5 = 0.116 ohm

Yes, it will need to dissipate up to 3W of power.

The 0.58V value will depend on the curves for your NPN transistor, but is probably right.

Another thing to try is to simply treat the TEC as a motor (as they're current-driven devices), and use, or make, an H-bridge (another example for an actual TEC) to control them. Here's a 9A pre-made one from Sparkfun, but they're not too hard to make with four power mosfets. For example, these are pretty cheap.

Since you already have/will have the Arduino, you can use it to do all the PWMing for the H-bridge.

The main source of heat on your mosfets used for an H-bridge will be from the switching. If they're in the full-on or full-off state they use virtually no power, but it's the transition period that they use the power. I think you'll be PWMing at a few Khz for the longevity of the peltier.

So I'm a bit confused, the H-Bridge is used to choose the direction of flow of electricity? And would I have this in addition to the MOSFET DRiver or what?

Thanks,

amk

 

[Bionic-Badger]

No, the H-bridge is built from the MOSFETs. In general, it's there to allow the motor to move either direction. You could probably get away with just one MOSFET that is controlled in the same manner as the four would be (with PWM, and the controllers it uses). However, you can use pre-made H-bridges to do this work instead of buying MOSFETs and rigging it up yourself. Another advantage is that you can even use this to "heat" systems with the peltier by reversing the current, instead of relying on the laser diode to be the heat source, allowing for more heat control (say you want to see how temperature affects wavelength in a laser diode). Lasers like greens also need to be slightly warm to be optimal.

 

[amkdeath]

Ok, I see.

Also, what is an effective method for cutting ABS plastic enclosures?