Could I make a simple tiny heater with 1/2 watt resistors?

[Jay112]

I apologize about the low milliamps thing. Most likely I was just reading my (new) multimeter wrong. I had given my old multimeter to a friend just a few days ago, and if I still had that one I would've been able to see the discrepancy.

Now I'm getting 330mA, tested with 2 different meters. That makes sense for a camera and the new heater I installed, right? I'm too tired to go back up on the roof today, otherwise I would test it with and without the heater. Maybe I'll do that another day.

[Gyro]

Simple answer first, the same resistors at 24V will dissipate four times the power... 4W rather than 1W, you'll be over-running your 1/2W resistors.

I suppose it's possible that a small CMOS camera might consume so little current, I would have expected a bit more considering it's driving 75R video. Maybe not.

I'd still be tempted to put the resistors in the airflow so that they're acting directly on the dome rather than losing heat into the lower casing.

P.S. Just seen your note on the current, yes that seems more reasonable. Sounds like time to take a break for the day then 
 

[Jay112]

Thanks for the info, Gyro!

About the fan: When it's 0 degrees outside and the fan is blowing, do you think maybe the end result will be that it dissipates most of the heat from the PCB, whereas if the fan wasn't on there might be a warmer area where the PCB is generating heat?

About the wattage: if my heater is currently drawing 1 watt (I think it's actually 1.1), and I have 6 resistors, that would mean that each resistor is carrying 1/6 of a watt, right? So then with 24VAC, since it will dissipate 4 times the power, that would mean that each resistor is carrying 4/6 (or 2/3) of a watt, which is over the 1/2 watt rating, right? So then if I simply added 3 of the same resistors to the series, then each resistor should be carrying 4/9 of a watt, which should be safe? Is that correct?

BTW I would definitely consider moving the heater in front of the fan like you suggested, but if I'm going to be switching between 12 and 24 volts depending on the outside temperature, then I can't be having the fan plugged in because I won't be able to go up there on the cold days to unplug it.

[Jay112]

Ugh, now I'm certain I did the math wrong, since by adding resistors the resistance and the wattage go up too.

[Paul Moir]

R = 9 * 22 ohms = 198 ohms
V = 24 (Volts AC RMS, so 24 volts DC equivalent power)
P = V^2 / R, so P = 24*24 / 198 =  2.91W

2.91W / 9 resistors = 0.323W per resistor. 

[Jay112]

R = 9 * 22 ohms = 198 ohms
V = 24 (Volts AC RMS, so 24 volts DC equivalent power)
P = V^2 / R, so P = 24*24 / 198 =  2.91W

2.91W / 9 resistors = 0.323W per resistor.

Thanks Paul!

I'm thinking maybe 8 of these resistors would be better, because then my wattage is higher when I'm running it on 12v DC. And it looks like if I used 7 it would be too much wattage when switching to 24v.

R = 8 * 22 ohms = 176 ohms
V = 24 (Volts AC RMS, so 24 volts DC equivalent power)
P = V^2 / R, so P = 24*24 / 176 =  3.27W

3.27W / 8 resistors = 0.41W per resistor.
----------------
R = 7 * 22 ohms = 154 ohms
V = 24 (Volts AC RMS, so 24 volts DC equivalent power)
P = V^2 / R, so P = 24*24 / 154 =  3.74W

3.74W / 7 resistors = 0.53W per resistor.

[Paul Moir]

Hey, no problem at all.

But now you have to put your engineer hat on and tolerance & derate.  Those are 22 ohm resistors but they're +- 5% tolerance so they could all be 20.9 ohms.  And that 24VAC is probably +- 10% or more.  I've seen 28VAC on 24VAC doorbell transformers.  So with 8 resistors:
8 * 20.9 = 167.2 ohms
V = 28
P = 28*28/167 = 4.7W

4.7 / 8 is too much.  And going with the somewhat lazy but easy derate of 80% of maximum, you're not going to want to use the resistors at more than .5W *  80% = 0.4W each.

(This is just a little pedantic, since you could just measure things and make sure they're OK.  I'm just illustrating the point.)

[Jay112]

I want to thank everyone for the help! This forum is amazing!

I went on the roof (for the FINAL time today!) and installed the last 2 resistors, so now the heating element has 8 resistors total.

The best part of it is that I can easily switch between .82 watts and 3.27 watts, simply by switching the (indoors) power supply from 12vdc to 24vac. So the solution was more robust than I would've ever thought.

Edit (now that I see Paul's new post): I can't believe I forgot to test the resistance over the heating element while I was up there! It would've been so easy! One thing I did test was the current drawn, and it was 57mA at 12v. I think that's a little lower than I would've expected.

I'm curious though: If I'm only using 24v on the coldest days (definitely less than 20 degrees F), would that coldness help the resistors be able to handle more power without burning out?

[Jay112]

Sorry for my beginner stuff, but according to Ohm's Law, if I was measuring 57mA at the heating element, that would mean the resistance should be 210 Ohms at 12v, right? But I used 8 x 22 Ohm resistors, which should've technically equaled 176 Ohms. With something like that, is it safe to assume that the resistors were just off? I think that would be more than the 5% tolerance limit.

I should've double-checked the voltage at the heating element while I was up there too, since obviously that might've thrown off the results.

[Gyro]

Yes I would think that the resistors are just a little off - they are carbon film ones so their temperature coefficient is pretty high.

Edit: Your supply at the end of the cable is bound to have dropped a little with the additional load, making the resistance 'appear' higher (plus the voltage drop across the current range of your meter).

The low temperature will certainly help their dissipation limit, this is based on maximum temperature before the element and case material (lacquer) degrade, so they de-rate with higher ambient temperature. Forced airflow (your fan) improves this even further against natural convection.

P.S. Sorry, I keep forgetting that you're turning the fan off at 24V

[Jay112]

Yes I would think that the resistors are just a little off - they are carbon film ones so their temperature coefficient is pretty high.

Edit: Your supply at the end of the cable is bound to have dropped a little with the additional load, making the resistance 'appear' higher (plus the voltage drop across the current range of your meter).

The low temperature will certainly help their dissipation limit, this is based on maximum temperature before the element and case material (lacquer) degrade, so they de-rate with higher ambient temperature. Forced airflow (your fan) improves this even further against natural convection.

P.S. Sorry, I keep forgetting that you're turning the fan off at 24V

Thanks for the info!

If you can figure out a way that I can have a 12v DC fan work with either 12v DV or 24v AC, then I'd be able to keep the fan running. It would be interesting to hear if there's an easy solution to this.

[Gyro]

I guess either a small 3 terminal 12V regulator or a series resistor and parallel 12V zener diode across the fan. I doubt that small fan pulls much current and the series resistor and zener would dump a bit more useful heat at 24V. Again you could put these in front of the fan to cool them, with no significant dissipation at 12V.

Come to think, if you used a higher wattage zener and lower value series resistor, you could use them as the actual 'heating element' which only dissipates power at 24V but runs the fan summer and winter! Does that count as lateral thinking? 

[Jay112]

I guess either a small 3 terminal 12V regulator or a series resistor and parallel 12V zener diode across the fan. I doubt that small fan pulls much current and the series resistor and zener would dump a bit more useful heat at 24V. Again you could put these in front of the fan to cool them, with no significant dissipation at 12V.

Come to think, if you used a higher wattage zener and lower value series resistor, you could use them as the actual 'heating element' which only dissipates power at 24V but runs the fan summer and winter! Does that count as lateral thinking? 

That's an awesome idea! I'm glad to have learned how it could be possible.

I don't think I have any zeners on hand, only regular diodes. I wonder though if some of the PCBs I save from broken electronics might have a zener of the correct wattage. I don't think I have any 12v regulators either.

Would a standard 12v DC wall wart contain either of those 2 components? I have a few broken wall warts that I saved in case I can use the parts in the future.

Otherwise I'll probably add some parts like that to my next ebay order, since they'll probably be useful for future projects anyway.

Your username reminds me of Greek food.

[Gyro]

I was going to say... you'll probably have to buy the bits (zener anyway). No it's unlikely that a wall wart will have one in, sorry.

Zeners commonly come in 400mW, 1.5W and 5W ratings, the 5W would give the largest package, so the greatest surface area. At 100mA it would dissipate 1.2W, 200mA would double that to 2.4W.I would think that the fan pulls less than 50mA so the resistor could be sized somewhere in the region of 56R 3W would give you 200mA plus a bit for the fan. That would give you a maximum heat output (at 24V) of about 5W. Varying between 12 and 24V would let you tune this.

You'd want to put both the Zener and the Resistor directly in the airflow with long leads to maximize heat transfer.

Your username reminds me of Greek food.

Haha, thanks. I used to have a bit of a passion for gyroscopes.

Chris

[Jay112]

Thanks bro! I appreciate your help!

What do you think of these diodes here? Is this a good one for a project like this?
http://www.ebay.com/itm/50pcs-1N5349B-IN5349B-Zener-diode-12V-5W-DO-15-IN5349-/291496699039

And then I was thinking it'd probably be useful for me to have 5v ones on hand for future projects, since I frequently work with 5v. Does it seem logical to stick with 5W ones, even at 5V where I never work with such high wattages?
http://www.ebay.com/itm/20pcs-1N5339B-1N5339-5W-5-6V-DIP-Zener-Diode-/371384223742

And I've never purchased such powerful resistors before. Do these 3W ones look decent?
http://www.ebay.com/itm/250pcs-50-value-1ohm-1M-3W-3-Watt-Metal-Film-Resistor-Assortment-Kit-3699-/301234991533

[Gyro]

Yes, they look fine. You'd probably get a better deal from a local electronics distributor though. Digikey or Mouser?

[Jay112]

Really? Whenever I looked up prices, I always found Digikey and Mouser to be more expensive. I've only purchased once from Digikey, for an Xbee module.

Like for the 12v 1N5349B diodes:
50 pieces from eBay + shipping = 0.156 cents each
From Mouser (http://www.mouser.com/ProductDetail/ON-Semiconductor/1N5349BG/) I would have to purchase 1,000 in order to beat that price (and then the cheapest shipping method is an extra $5 on top of that).

Digikey required a minumum order of 4,000 for most of them, but for the ones that had smaller minimums they were frequently 50 cents a piece (http://www.digikey.com/product-search/en?keywords=1N5349B).

And then for the 5v 1N5339B diodes:
From eBay they were .275 cents each.
On Mouser I would have to purchase 100 in order to beat that price, not including the extra shipping costs.

Am I missing something? I'm very much a beginner with all of this stuff, so it's very possible I'm overlooking something.

[Maxlor]

Am I missing something? I'm very much a beginner with all of this stuff, so it's very possible I'm overlooking something.

Reliability. With Digikey, Mouser and the other brand names you can be sure they're not trying to screw you over (except on price, but at least they're open about that...), you get non-counterfeit parts. With ebay, there's the luck of the draw. There are many excellent sellers in that marketplace, but also black sheep that'll sell factory rejects, slightly mislabelled products (B-grade sold as A-grade, or no grade specified at all) or outright fakes.

[Marco]

I estimate this will only raise temperature a couple degrees (5mm of acrylic, 0.02 thermal conductivity, 100 cm2 surface, 4 watt). If you need more you could try putting the resistors directly next to the sensor and gunk it up with hotglue for insulation, you'd want a thermostat though at that point.

[Jay112]

Reliability. With Digikey, Mouser and the other brand names you can be sure they're not trying to screw you over (except on price, but at least they're open about that...), you get non-counterfeit parts. With ebay, there's the luck of the draw. There are many excellent sellers in that marketplace, but also black sheep that'll sell factory rejects, slightly mislabelled products (B-grade sold as A-grade, or no grade specified at all) or outright fakes.

That makes a lot of sense! I'll definitely keep that in mind if I ever start working on more serious projects.

I'm at the stage where 90% of my failures/explosions are due to stupid beginner mistakes, so currently I would prefer to save money and blow out the cheap components, instead of having to worry about blowing out more expensive parts.

[Jay112]

I estimate this will only raise temperature a couple degrees (5mm of acrylic, 0.02 thermal conductivity, 100 cm2 surface, 4 watt). If you need more you could try putting the resistors directly next to the sensor and gunk it up with hotglue for insulation, you'd want a thermostat though at that point.

Thanks Marco! I'll definitely keep that in mind if I don't notice much improvement on the cold days.

I've been brainstorming a little with this idea: I could fairly easily bring a cable from another power source into the dome. Then if I get some 3W resistors I could make a more powerful heater. Then I could use a simple thermostat (I already have a spare on hand) that only turns on that 2nd power supply (which is connected to the heater) when the temperature drops below freezing. Then I could use a powerful heater when it's cold out, and I won't have to worry about overheating or wasting energy when the temperature rises.

[Jay112]

Last night the temperature was down to 15 degrees, and the camera image was crystal clear the entire time. I've never seen that happen before I installed the new heater. It looks like it's working! Thanks guys!

[Gyro]

Thanks for reporting back, a nice payback for all that time climbing on the roof. Well done! 

At least you have a better handle on the dissipation required - let's see how it does at 14 degrees and below (climate permitting). 

[Jay112]

Hey peeps - I hope everyone is doing well! I just wanted to give a quick update about the heater you guys helped me make, now that the winter is over.

The project was a huge success. Running on 12v mode the camera was able to get down to 0° F without having any problems, but the image would start to stutter when the temperature dropped below zero. That's a big improvement to how it was before, when the image would start stuttering below 20°.

But then on 24VAC mode, the camera was able to function well even below -10°, which basically means it functioned well all winter. Pretty soon I'll unplug the heater and plug back in the cooling fan, and then I'll be set until next winter.

[Brumby]

Good to hear.  Thanks for making the effort to give us an update.

Maybe we should put together a label to put on such projects...



Hmmmm.....  I have some ideas