Quick "proof reading" needed :)

Hey guys...

I'm building an Arduino controlled reptile egg incubator (my turtle is digging, so I can expect eggs any moment). That means I'm quite limited with time and have to get it done with what I have at hand. The incubator itself is completed, now comes the electronics. I successfully made a tiny PCB with a DS18B20 temperature sensor and managed to get a "thermostat" code working (for now, if the temp inside is too low, a blue led turns on, if it's too high, a red one turns on and if it's in the "ideal" zone, both turn off), anticipating the use of a peltier element I have at hand as a heater/cooler.

Now, I have to drive it somehow, and even be able to switch polarity. Fortunately, I don't need PWM, so I think I could use this controller, based on a DPDT relay:

http://www.me.umn.edu/courses/me2011/arduino/technotes/dcmotors/bidirectional/bidirMotor.html

I managed to find a nice 5V DPDT relay (two, actually) on an old modem, however, I have no TIP120s at hand, the closest I have is an IRFZ44N. I've re-drawn the schematic in Eagle and I'd need someone to check if it looks good before I etch my last copper clad board to make it. Also, I left the 2N3904 when I re-did the schematics, but since I don't have one, I guess it's not a big problem if I change it with another small NPN, right?

Here it is:

Quick, constructive responses are more than welcome :) Thank you in advance, on behalf of future baby turtles as well ;)

Cheers,

Sebastian

The IRFZ44N is not a logic-level MOSFET (won't turn on at 5V) so not a good choice there.

You should have a diode across the relay coil to protect the 2n3904 from a big turn-off spike.

Otherwise it looks pretty OK....

-- The Gadget Shield: accelerometer, RGB LED, IR transmit/receive, speaker, microphone, light sensor, potentiometer, pushbuttons

Your circuit design looks good. However you will have a problem with the IRFZ44N N-channel power mosfet, because it’s not a LOGIC LEVEL mosfet, the arduino’s +5vdc output voltage from a digital output pin will not switch the mosfet into full saturation.

You didn’t mention how much current your peltier element draws. So you either have to find a NPN power transistor with collector current rating much higher then your peltier current draw, say X3, or use a logic level mosfet also rated well above the peltier current ratings. The advantage of a suitable rated logic level mosfet is that you maybe able to get by without a heatsink attached to the transistor, which I suspect you will need for any NPN transistor you use. Again that’s just a guess until you state the current requirement for the peltier element.

I like this logic level mosfet from SparkFun: http://www.sparkfun.com/products/10213

Lefty

Hey guys, thanks for the reply. I hope I’ll be able to find another MOSFET to use… I didn’t even think about the logic-level thing, since all I was thinking about is that my peltier draws a lot of current (relatively, at least).

It’s a model TEC1-12706. Here’s the data:

Qmax (W): 50
Vmax (V): 14.4 (I’ll be driving it at a slightly lower voltage, 12V)
Imax (A) 6.4
Module resistance (Ohms): 1.98

Perhaps if I tear apart an old ATX power supply I’ll be lucky enough to find a decent MOSFET… The thing is, I really don’t have much time, and there are no electronics stores here, waiting for mail to arrive is not an option, unfortunately :stuck_out_tongue: So, basically, what I’m looking for now is ANY N-channel MOSFET that can handle at least 7amps and switch at logic level… right? :smiley:

Perhaps if I tear apart an old ATX power supply I’ll be lucky enough to find a decent MOSFET…

Not likely, and what you find most likey won’t have a useable device part number, so you won’t know it’s ratings.

The thing is, I really don’t have much time, and there are no electronics stores here, waiting for mail to arrive is not an option, unfortunately So, basically, what I’m looking for now is ANY N-channel MOSFET that can handle at least 7amps and switch at logic level… right?

Right. So no time, no local stores. You make it hard to help.
However as you said you don’t have time to fool around, you have future children to think about. So as time is money SparkFun will be happy to ship to you by overnight shippment:

Domestic

This is for shipping addresses within the United States. We offer 9 options:

FedEx - (All FedEx shipping methods are tracked and fully insured)

•FedEx First Overnight - next business day by 8:30AM, later for rural
•FedEx Priority Overnight - next business day by 10:30AM, later for rural
•FedEx Standard Overnight - next business day by 3PM, later for rural
•FedEx 2-day Air
•FedEx Express Saver - 3 Day
•FedEx Ground/Home - not day guaranteed, but averages 3-5 business days

Think of the children!

Lefty

Unless you live in Slovenia, then it takes 2 weeks :)

I have a friend who is an electronic engineer and used to build those huge scrolling LED displays, he might have some useful parts. I'll call him tomorrow.

OR... I was thinking... does anyone think the same thing (switching off and on and being able to reverse polarity of the TEC) could be accomplished somehow using TWO DPDT relays and a bunch of "regular" transistors? :) I have those ;)

S.

could be accomplished somehow using TWO DPDT relays and a bunch of "regular" transistors? I have those

Yes it can be done with just two relays, one with DPDT contacts to handle the polarity reversal and one just SPST relay in series with +12vdc power source to control on/off commands. That would require two digital output pins (heat or cool and power on and off) and two transistors to handle the two relay coils. However relay contact ratings have to be at or above the heating element current rating.

Lefty

Got it. I guess it doesn't matter if I use a DPDT (actually, half of it) as the SPDT, right? Since all I have are DPDTs :P And yeah, they are surprisingly powerful ones (I don't have the datasheet, but I've seen a few projects using the exact same type involving serious loads)... I'll try to figure out a schematic and post it here so you'll tell me if I'm doing it right... in the meantime, thank you a lot :)

Edit: P.S.: I just figured out I can also use the "free" half of the second DPDT to turn on/off the fan, so I'm now actually using an overkill part :D

Ok, it's 3am here, I'm sleepy as hell and not completely in possesion of all my brain capacities... but I managed to sketch something based on your feedback.

I'm pretty sure it's messed up, so please, bear with me, and if you're feeling nice, offer me corrections :)

Here we ho:

Does it seem like a working thing? BTW, would the BC109s be good for the job? Also, are the values of the resistors important or can they be "just something around there"? :)

Thanks again in advance,

Cheers, Seb

So....the fan is always on, the Peltier element is on by default UNLESS PIN0 (ON/OFF) is high. Is that the intended behavior?

I still recommend diodes across the relay coils :)

-- Beat707: MIDI drum machine / sequencer / groove-box for Arduino

Sort of. The symbol I found in Eagle assumes it's a normally closed relay, but mine is normally open, so that should mean the Peltier is OFF unless PIN0 is HIGH, right?

Is the current/direction reversal ok, or did I miss anything?

Forgot the diodes. Any general purpose should work, right? :)

Remove the jumper wire from P1-S1 on K2. It's shorting out the 12vdc battery when deenergized.

Lefty

Thanks, I see now it's wrong. Otherwise it seems ok, then? :) Wow, I'm getting good at it :P