Connect with us

Heat Pads for Animal Hospital

Discussion in 'General Electronics Discussion' started by Raven Luni, Nov 25, 2012.

Scroll to continue with content
  1. Raven Luni

    Raven Luni

    798
    8
    Oct 15, 2011
    Greetings,

    This is my prototype design for a heat pad system for the wildlife rescue centre where I volunteer.

    heater1.jpg

    The pads are made by sparkfun and their performance seems to differ quite a bit from what they said in the video. I have the 5cm x10cm ones and they draw 0.2A @ 5V (the guy said it was 0.8A). That gives them a 25 ohm resistance.

    The power supply isnt determined yet - will probably be old PC power supply with a battery backup. I've made the voltage adjustable mainly for performance reasons (at 5V the pads are slow to heat and quick to lose heat, and if the temperature is regulated, then an extra kick shouldnt harm them). The 4W resistors are all I have at the moment that will handle the ~0.6W dissipated there. I'll get something more suitable later.

    Feedback / criticism etc. (bracing myself :p )
     
    Last edited: Nov 26, 2012
  2. Raven Luni

    Raven Luni

    798
    8
    Oct 15, 2011
    D'OH! Just realised I typo'd the title of this thread. It should be HEAT pads not head pads :p Can someone change it please?
     
  3. CocaCola

    CocaCola

    3,635
    5
    Apr 7, 2012
    If you go into advanced edit you can change the topic...

    I can't comment all that much on this because you know what you need best...

    But, when I was dealing exotic animals for a living I used a lot of over the counter 'pig blankets' for this, that or under the floor heating fim for houses... Both running off the mains with common light dimmers to adjust the temp...

    I can see your design is more compact and isolated heat, maybe better for your application?
     
    Last edited: Nov 25, 2012
  4. Raven Luni

    Raven Luni

    798
    8
    Oct 15, 2011

    I think my main concern is probably the switching transistors and the fact that the power used is comparable to that for the heating element itself (I was originally going to use a 33 ohm then realised I didnt have any of those with a high enough power rating). Is there a more efficient way to do this?
     
  5. BobK

    BobK

    7,682
    1,688
    Jan 5, 2010
    Why use a 6A transistor to switch 200ma?

    I would use a MOSFET rated at about 500ma and switch the + and - on the comparator so that another inverter is uncessary.

    Bob
     
  6. Raven Luni

    Raven Luni

    798
    8
    Oct 15, 2011
    Because I have lots of them and because the current requirements are likely to change and mosfets are evil :p
     
  7. Raven Luni

    Raven Luni

    798
    8
    Oct 15, 2011
    To be clearer - the current relationship is not linear. At 6V it jumps to 400mA (just checked that). I also might want to put more than one on the same controller (they are pretty small after all).

    And mosfets are evil :p (did I say that)
     
  8. BobK

    BobK

    7,682
    1,688
    Jan 5, 2010
    MOSFETs are not evil. And if they are, it is only when you have high switching frequencies, which is not the case here.

    Bob
     
  9. Raven Luni

    Raven Luni

    798
    8
    Oct 15, 2011
    Well I've built teh power supply. That transistor gets HOT (heat sink is too hot to touch after several minutes at 600mA).

    Am I going to need an insanely big heatsink or am I doing something wrong?
     
  10. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,497
    2,838
    Jan 21, 2010
    The heat pad should be the only thing getting hot!

    Which transistor is getting hot? (or is it both of them)

    The output stage you have there is a little evil.

    Connect the transistors up as a Darlington and I think that problem will go away.
     
  11. Raven Luni

    Raven Luni

    798
    8
    Oct 15, 2011
    The pass transistor for the power supply
     
  12. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,497
    2,838
    Jan 21, 2010
    What's the input voltage, output voltage, and current. That transistor could be dissipating more than is being dissipated in thee heating pad.

    Is a regulated supply actually needed?

    And urk, using a fixed regulator as a variable regulator.
     
  13. Raven Luni

    Raven Luni

    798
    8
    Oct 15, 2011
    Been using 12V for input, Regulated supply is needed else it cant work as a calibrated thermostat. As for variable supply, the 5V in the spec for those things gives a pretty pathetic performance so I wanted to option to boost it.

    In that particular test it was set at about 7.5V and was drawing about 600mA if I remember.
     
  14. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,497
    2,838
    Jan 21, 2010
    Since your temperature detection is done by comparing a pair of potential dividers, they should be relatively insensitive to variations in voltage.

    Perhaps you should invest in a small switchmode regulator. Your linear regulator is dissipating more heat than your heat pad.

    As another alternative, you could place the regulator in the heat pad as well.
     
  15. Raven Luni

    Raven Luni

    798
    8
    Oct 15, 2011
    Hmm - cant quite get my head round the temperature bit. If the voltage fluctuates, both comparator inputs are going to change with it as well as voltage to the heat pad - and that has a nonlinear voltage / current relationship (5V = 200mA, 6V = 400mA, I'll need to plot it properly at some point). But I'll check all that anyway and take detailed measurements. You could be right.

    As for the regulator, I had thought about making use of the heat. Also if its running off an old PC supply (which I plan to do) I could take it off the 5V rail but as I mentioned, the performance is crap at 5V so only the 12V would do.
     
  16. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,497
    2,838
    Jan 21, 2010
    The only thing that will really change as the voltage rises and falls is that the amount of heat produced by the heat pad will change. So the rate at which you can heat up the pad will differ.

    The only problem is that the pad may not be able to safely handle the higher voltage.

    The best thing is to try it out. Run it at 5V, let the temperature stabilise and measure the temperature. Then run it at 12V and see if the temperature changes significantly.
     
  17. Raven Luni

    Raven Luni

    798
    8
    Oct 15, 2011
    Oh the endless debugging. 7:24 in the morning and I've been up all night tying to figure out why both transistors seem to be permanently turned on. I've got an LED in place of the heat pad and a 10k resistor in place of the thermistor for testing. The comparator works fine, several volts develop across the 2.2k resistor, I've replaced the 2n3904 twice....
     
  18. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,497
    2,838
    Jan 21, 2010
    Did you take my advice to wire up those transistors as a darlington pair?
     
  19. BobK

    BobK

    7,682
    1,688
    Jan 5, 2010
    Have you considered a microcontroller and PWM? This would solve your regulator problem.

    Bob
     
  20. Raven Luni

    Raven Luni

    798
    8
    Oct 15, 2011
    Ok - finally got some time to work on this again. Thanks for the suggestions all. I got it working by taking out the 2n3904 alltogether (and moving the thermistor to the positive side). That also eliminates those power resistors. I had been concerned with overloading the op amp and / or not having enough current to drive the tip41a which is why I wanted a 'stepping stone' in between but it works fine, doesnt even heat up (well the heat pad does of course).

    Steve's point about maintaining constant temperature regardless of voltage seems to held up, but there are still alot of parameters that need to be tweakable.

    Increasing the voltage allows the pad to heat up faster and reach its desired temperature. This also results in greater power supply efficiency and less heat in the pass transistor, but it draws alot more current (over an amp just for 1 pad). For these reasons I want to keep it variable otherwise I'd do away with the regulator.

    I'll have to redesign that regulator. Microcontrollers are out of the question for the time being. There was a suggestion or 2 for driving those mosfets I have so I might give that a try (some boost for my buck :p )

    I think it might be worth having a failsafe as well. The worst that can happen is the thermistor fails and the animal gets cooked due to uncontrolled heating. Maybe a second thermistor / comparator set slightly higher that will latch the whole thing off and raise an alarm (dont really want an audible alarm though as it might drive the animals mental - same with bright flashing)....

    Any more thoughts?
     
    Last edited: Dec 5, 2012
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-