Should resistors normally be too hot to touch

Thread Starter

sparkie78

Joined Nov 5, 2016
8
I'm new to electronics and I have a question about how hot resistors get. I'm turning an old computer power supply into a workbench supply. It would run for a few seconds then turn off. I read that some power supplies won't turn on or stay on if it doesn't sense a load and to remedy this install a resistor between the +5 volt line and ground.So I soldered in a 47 ohm 2 watt resistor between +5 volt and ground and it worked, the power supply stayed on. However, after just 1 minute, I touched the resistor to check how warm it was and quickly let go. It was too hot to touch! By my calculations, the power should have been just over 1/2 watt (0.532). Should a 2 watt resistor get that hot with just a 1/2 watt load? Or have I made a mistake in my calculation?
I used V / R = I, I x V = P
 

MrChips

Joined Oct 2, 2009
19,379
Use any of these three formulas to calculate power:

P = I x V
P = I x I x R
P = V x V / R

A 47-ohm resistor across 5V would dissipate 5 x 5 / 47 = ½ W

Double check the voltage and resistance.
 

jpanhalt

Joined Jan 18, 2008
8,066
Through hole resistors are often rated to 155°C.
upload_2019-9-29_15-30-16.png

I would expect your resistor to get warm, but perhaps not hot. By hot, I mean 65 to 80°C. At 60°C is where I first sense a little discomfort.
 

dl324

Joined Mar 30, 2015
9,111
However, after just 1 minute, I touched the resistor to check how warm it was and quickly let go. It was too hot to touch! By my calculations, the power should have been just over 1/2 watt (0.532). Should a 2 watt resistor get that hot with just a 1/2 watt load?
It depends on what too hot means. In general, hot to the touch means little to electronic components.

People often derate resistors to 50% of the manufacturer spec regardless of operating conditions. In most cases, this leads to an ultra conservative design.

According to a Yaego datasheet for metal film resistors, resistors can be operated at 100% of their rated load at an ambient temperature of 70C or lower. You don't need to derate to 50% until ambient temperature is over 100C.
upload_2019-9-29_12-55-9.png
 
Last edited:

MisterBill2

Joined Jan 23, 2018
3,983
Computer power supplies are a special case in that they do require a serious load on the low voltage section to stay on. And the resistor that is used as that load will get very hot because it will usually be drawing at least 3 amps, at 5 volts that is 3 x 5 =15 watts, and so it must dissipate 15 watts as heat. So it will be very hot indeed. That is why using that kind of supply as a bench supply is not the best plan, unless you have some load that can do something useful with that much power, such as a whole lot of LEDs, or a 5 volt powered fan, or a whole bunch of 6 volt light bulbs.
 

WBahn

Joined Mar 31, 2012
24,850
I'm new to electronics and I have a question about how hot resistors get. I'm turning an old computer power supply into a workbench supply. It would run for a few seconds then turn off. I read that some power supplies won't turn on or stay on if it doesn't sense a load and to remedy this install a resistor between the +5 volt line and ground.So I soldered in a 47 ohm 2 watt resistor between +5 volt and ground and it worked, the power supply stayed on. However, after just 1 minute, I touched the resistor to check how warm it was and quickly let go. It was too hot to touch! By my calculations, the power should have been just over 1/2 watt (0.532). Should a 2 watt resistor get that hot with just a 1/2 watt load? Or have I made a mistake in my calculation?
I used V / R = I, I x V = P
It depends a LOT on the specifics of the resistor you are using. Some power resistors are rated based on a max temperature of 155 °C and others are at 240 °C (and I'm sure these aren't the only ratings out there).

It's actually amazingly difficult to find thermal resistance data for power resistors, which I always found rather surprising and annoying.

The handful of data sheets that give temperature versus power generally indicate that the temperature rise is linear with respect to the power dissipation (for no additional heat sinking or forced cooling). So if you have a resistor rated at 155 °C when operated at 2 W at 25 °C ambient, then that's a temperature rise of 130 °C. If you operate it at 25% of that, then you would expect somewhere around 30 to 35 °C temperature rise taking you to around 60 °C. This is generally considered hot enough to produce second and possibly third degree burns in just a few seconds of solid contact. If you are using a resistor that is intended just to dump power and is not intended to stay particularly close to its nominal resistance, then it may be rated to operate at 240 °C, making the surface temperature at 25% power in the 80 °C range, which is enough to cause serious burns in a fraction of a second of solid contact.
 

WBahn

Joined Mar 31, 2012
24,850
It depends on what too hot means. In general, hot to the touch means little to electronic components.

People often derate resistors to 50% of the manufacturer spec regardless of operating conditions. In most cases, this leads to an ultra conservative design.

According to a Yaego datasheet for metal film resistors, resistors can be operated at 100% of their rated load at an ambient temperature of 70C or lower. You don't need to derate to 50% until ambient temperature is over 100C.
View attachment 187094
The derating depends heavily on the resistor type and specs. Some don't derate until 70 °C, but many others start derating right at 25 °C.

Part of this is driven by the specified resistor tolerance over temperature and life.
 

Reloadron

Joined Jan 15, 2015
4,573
Normally when doing what you are doing I would just go to my local Radio Shack and buy a 50-Ohm 10-Watt 10% Wirewound Resistor (2-Pack) which had the ceramic body. While Radio Shack is no longer with us the resistors are. Another good choice were the Ohmite Aluminium Housed Resistors. The beauty of the latter is the ease in mounting them to the power supply chassis making for easy heat sinking. They cost a few bucks more. While granted 10 watts is overkill they are both inexpensive and work just fine. Both are wire wound and tolerance is not an issue.

Ron
 

MisterBill2

Joined Jan 23, 2018
3,983
The wattage rating of a resistor will not tell you how hot it is going to get, but rather how much power it can convert into heat without damage that leads to failure. Temperature rise is determined by power input and heat removal. That same resistor with a large heat-sink and a good fan produced air flow will not get as hot.
 

WBahn

Joined Mar 31, 2012
24,850
The wattage rating of a resistor will not tell you how hot it is going to get, but rather how much power it can convert into heat without damage that leads to failure. Temperature rise is determined by power input and heat removal. That same resistor with a large heat-sink and a good fan produced air flow will not get as hot.
But how much power it can convert to heat without damage is primarily a function of the temperature that its core ends up operating at. If you take a 1 W resistor and heat sink the hell out of it and immerse it in a flowing coolant then you can dump some serious power into it without damaging it.

For power resistors intended to run in free air, the rating it based on the most power that it can dissipate without the core exceeding the rated max operating temperature. So, if you are running it in free air, you can estimate the temperature it will operate at based on those parameters. If you add any kind of heat sinking or other cooling mechanism, then you have to step back and take those into account for your specific situation.

Another way of saying this is that you can get a reasonable estimate for the core-ambient thermal resistance by taking the (max temp - ambient temp) specs and dividing by the power rating. This is probably why they never put this in the datasheet -- they figure it is too easy to calculate yourself. It's still annoying that it isn't just listed so that you can do quick selection discriminations without having to look at the spec details.
 

MrAl

Joined Jun 17, 2014
6,604
Hi,

If you think it is too hot drop a drop of distilled water on it see how long it takes to evaporate.

If it is too hot for your liking and you have 4 of those resistors on hand, put two sets of two in series then put those two sets in parallel. That raises the total power handling ability by roughly 4 times and since the surface area increases by 4 times the surface temperature decreases significantly.

Some manu's publish temperature data for their resistors on the data sheet for various configurations like horizontal, vertical, forced air @cfpm, etc.
 
Last edited:

MisterBill2

Joined Jan 23, 2018
3,983
Hi,

If you think it is too hot drop a drop of distilled water on it see how long it takes to evaporate.

If it is too hot for your liking and you have 4 of those resistors on hand, put two sets of two in series then put those two sets in parallel. That raises the total power handling ability by roughly 4 times and since the surface area increases by 4 times the surface temperature decreases significantly.
You could also put 4 of the 2 watt 220 ohm resistors in parallel, which is simpler mechanically. And 220 is a quite common value, probably easier to get.
 

MrAl

Joined Jun 17, 2014
6,604
You could also put 4 of the 2 watt 220 ohm resistors in parallel, which is simpler mechanically. And 220 is a quite common value, probably easier to get.
Hi,

Yeah lots of possibilities here. Also two sets of two in parallel, then those sets in series.
Of course we have the simple route too, two 100 ohm's in parallel makes 50 ohms at twice the power of each resistor which might be enough.

I also like to mention that if they are in an enclosed area that makes it more difficult because then we have to assess the heat rise int ehre and how it affects other components and how they affect it too. It might be near a plastic part that melts too.

The temperature rise is related to the power dissipated and the total surface area and if there is any air flow or other type of forced cooling.
Often resistors depend on convection cooling alone.

A while back i got some 3 Ohm 50 watt power resistors, the gold color ones with the chassis mount holes. The data sheet had shown temperature rise when they were bolted to a chassis of a specific size. I used them for a freezer defroster :)
 

Thread Starter

sparkie78

Joined Nov 5, 2016
8
The resistor type was a carbon film. I had a 10 ohm 10 watt ceramic cement resistor I scavenged off an old PCB and used that instead and everything seems to be working fine. Thanks to everybody who answered my post. Seems that resistors aren't as simple as I thought they were.
 
Top