Power resistor strange rising in temp

[RoGeorge]

I was probing a 15\$\Omega\$ resistor at 5Vdc (0.336A measured, Pd = 1.68W).  It's been sitting on the bench for an hour now, in free air, with nothing but two alligators from the power supply at 5V, and its temp keep rising... 

It measures 64*C already (147*F), and still increasing at about 0.5...1 degree per minute.

The resistor is from a former 8'' floppy disk.  It has a body of aluminium 3.2x1.5x1.5 mm (in inch 1.2x0.6x0.6), with mounting ears.  Marked as REMIX-R6192 15RK TN, most probably 8W nominal power by this page:  https://vikiwat.com/en/resistor-68ohm-8w-5-remix-r6192.html



I don't understand how such a big resistor can get so hot in open air?  And how that after an hour of constant power and constant conditions, sitting on the bench in open air, the temperature didn't settle yet? 

[james_s]

Why is that surprising? It's dissipating power which is constantly adding thermal energy to it. It's losing thermal energy to the room via conduction, convection and radiation. At some point the temperature will rise to the point of equilibrium where the heat losses equal the electrical energy going into the resistor but you are not there yet. With almost 2 watts being dissipated I would expect it will reach 80-100C before that happens.

The 8W nominal power is probably assuming it is bolted to some kind of heatsink. It may also be specified at an operating temperature of 125C or more, if you can find a datasheet that will tell you.

[Siwastaja]

Sounds unsurprising as you have not mounted it to any heatsink, and it can't radiate due to the bare aluminum finish. That only leaves convection cooling with not much surface area.

If you measure constant power in, then surely it can't heat up more than that. Or maybe you are soon being nominated for Nobel prize on going overunity.

[RoGeorge]

Why is that surprising?

I guess it's my limited experience with power electronics, then.  I was expecting an Al block the size of a thumb to dissipate 1.7W in convection without much heating (my guess was 30-40*C).  Also I was expecting the temp to settle in a few minutes, not in longer than 1 hour.  Seems I was wrong.

It's a vintage resistor, couldn't find any datasheet.
It was used as a current limiter in series with the windings of a motor in 8'' floppy disk drives.

[Siwastaja]

Note that natural convection works by warm air rising and cold air replacing it. Therefore the cooling power (amount of thermal energy being removed per time unit) is proportional to temperature difference. The hotter it gets, the more watts can be removed; and by more than linearly.

You can try how forced cooling works by using a fan. You may be surprised how much more effective it is even when you have the exact same surface area available.

But if you can allow the resistor to get hot, then the difference to fan cooling is not that much anymore, because the heat removed with fan goes up only linearly with temperature. Hotter air will transfer more heat out, yes, but you only have the temperature to work with; air flow keeps the same, as defined by the fan.

Natural convection starts at zero cooling power because there is no airflow, and climbs up slowly, but starts to work at higher temperatures. Heat something up to a few hundred degrees and you can start feeling the air flow like a fan was running.

With radiation, the exponent is even bigger.

[Manul]

I think the area on your bench on which it is sitting is also getting slowly soaked by heat. That is also increasing time to reach equilibrium.

[RoGeorge]

Will experiment more to grow a feeling for thermal dissipation.

I've stopped the open air test at about 70*C, then tried again, this time with the resistor clamped to a big radiator with fins, black anodized, about the size a hard disk.  The resistor stays cold as a brick (30*C, only 5*C above ambient temp), and it settles in less than a minute.

Second thought about natural convection, I've noticed the air is also very dry today (indoor weather station shows 26% humidity).  My guess is higher humidity would have done a better job with natural convection, because a water molecule is heavier and can take away more energy when bumping into the resistor surface.

[RoGeorge]

I think the area on your bench on which it is sitting is also getting slowly soaked by heat. That is also increasing time to reach equilibrium.

That must be it, now the very slow time constant makes sense.  Underneath the resistor was a sheet of paper, then a termoresistant rubber antistatic mat, then some pressed wood fibers desk.

[james_s]

The temperature rise can be calculated, though it can take some work to figure out the expected heat loss through the various paths. Consider how hot an incandescent lamp gets, even a small 1.5W automotive indicator lamp will easily burn your fingers and can char materials that come into prolonged contact. A typical small hot melt glue gun is around 10 watts or less and is able to heat not only the surface area of the gun and nozzle but the glue sticks that pass through it. A watt is a significant amount of heat in a small space, a linear regulator dissipating 2-3 watts needs a finned heatsink and will probably get too hot to hold a finger on even then.

[RoGeorge]

Agree, but a resistor so big yet so hot at only 1.7W feels wrong.

How come that a Li-Ion USB powerbank can deliver 1A/5V without hitting that much?  1A would be enough to power 3 resistors like the 15\$\Omega\$/8W one.  The USB powerbank might be sligtly bigger, but not much bigger in size than 3 of those power resistors, yet the powerbank is covered in a plastic box, which must be pretty bad at evacuating heat when compared to an Al block.

I understand the basic physics, the difference between temperature and power and energy, the role of thermal resistance, specific heat, radiation vs conduction, etc. and I've measured that resistor myself.  Still hard for me to accept that I've poked my finger at it for 1-2 minutes and it was slightly warm, with no signs of much heating.  Then forgot about it and 30 minutes later was quite hot.  Then an hour later was too hot to keep a finger on it.

Resistors don't usually get hotter and hotter for 1-2 hours when kept under constant conditions.  They usually stabilize at a certain temperature in seconds.  I want to measure again tomorrow, maybe log the temperature, and try to measure the bench temperature around the resistor, too, as per Manul's hypotheses.

[Manul]

Agree, but a resistor so big yet so hot at only 1.7W feels wrong.

How come that a Li-Ion USB powerbank can deliver 1A/5V without hitting that much?  1A would be enough to power 3 resistors like the 15\$\Omega\$/8W one.  The USB powerbank might be sligtly bigger, but not much bigger in size than 3 of those power resistors, yet the powerbank is covered in a plastic box, which must be pretty bad at evacuating heat when compared to an Al block.

I think Li-Ion cell inside powerbank might have ESR of around 100 mOhm. So that's 100mW of power dissipation inside a cell. Very little. Now the boost converter outputting 1A @ 5V (5W) might have around 90% efficiency or more, so that's 500mW loss. So let's say 0.6W in total. Yes, the box is plastic and has bad heat conduction, but the heat flow is not high either, so in the end it cools quite well.

Edit: I don't know what I was thinking while writing, cell current is of course not 1A, more like 1.5A. But to be fair, that does not change much in this ballpark calculation.

[james_s]

How come that a Li-Ion USB powerbank can deliver 1A/5V without hitting that much?  1A would be enough to power 3 resistors like the 15\$\Omega\$/8W one.  The USB powerbank might be sligtly bigger, but not much bigger in size than 3 of those power resistors, yet the powerbank is covered in a plastic box, which must be pretty bad at evacuating heat when compared to an Al block.

Your power bank isn't dissipating that power, it's delivering the power, there's a big difference. The load is what dissipates the power and produces heat. Dissipation in this context is turning electrical energy into heat energy, and a resistor does that with 100% efficiency. If you have your mobile phone plugged into it, the phone will get quite hot in use and it has a huge surface area compared to your resistor, phones are carefully engineered to manage heat, it is one of the bigger engineering challenges. They also radiate a significant amount of the energy they consume as RF from the transceiver and light from the display.

[KubaSO]

For anything that generates heat to stop rising in temperature, it must be in a thermal equilibrium with the environment.

Thermal equilibrium is achieved when the heating power is the same as cooling power. In your case, cooling is mostly by convection, and a bit via conduction to the table.

The only way cooling happens is when there is a temperature difference. So, in order to dump any heat at all through conduction/convection, the resistor has to be hotter than environment. It doesn't matter how much heat it has to dump - it will always be hotter for heat to flow out of it, to get cooled. The more heat it has to dump, the hotter it will be, everything else being the same.

Since cooling rate is primarily controlled by temperature difference, heating rate is fixed, and temperature rise is by heating/cooling imbalance acting on the thermal mass of the resistor, you inherently get asymptotic behavior.

That resistor will be heating up essentially forever, just ever slower and slower. In practice it only depends on how accurate your temperature measurements are. In a precision temperature controlled chamber, with precision temperature sensing, you'd observe that resistor heat up for a day or more. Eventually it'd get buried in measurement drift and noise, but the temperature rise never stops - not unless the heat sink (here: air and table) temperature drops far enough.

So, what you observe is normal. It would be super strange if it was any other way!

I was expecting an Al block the size of a thumb to dissipate 1.7W in convection without much heating (my guess was 30-40*C)

See how big a pencil soldering iron's hot section is? It will be dissipating say 5-15W while the tip is at 250C or thereabouts. Your resistor dissipates a couple times less heat, but not much larger.

Get a 2W rated through hole axial resistor and see what it does when it is dissipating just 1W

[forrestc]

There's one other thing no one has mentioned.

As resistors get hot, their value changes.   That's what the ppm/c rating is.   Sometimes they have more resistance, and sometimes they have less depending on whether the resistive element has a positive or negative temperature coefficient.

The resistor you mentioned has 150ppm/C of change.  You've gone up roughly 44 degrees C, assuming 20*C room temperature.   This is 6600ppm of change, which is actually 0.6%.  So your 68 ohm resistor might have been reduced to 67.4 ohms, which will also increase the amount of heat dissipated.

[james_s]

A 0.6% change will have a negligible effect on the amount of heat produced. I think you would have to carefully control conditions in order to measure the difference between +/-1% energy going into the resistor, a slight change in room temperature or a draft from walking by the bench could swamp that. It is worth taking into consideration if your application depends on the precise resistance though.