Using a small resistor in some water instead of a power resistor?

[help_me_pick_username]

I don't have any large power resistors, I was wondering whether I could, instead of a power resistor, use a 1/4w resistor dunked in some water to dissipate some power? If this was even remotely possible, how much power would I be able to dissipate?

[Siwastaja]

You need to try it out, and it will heavily depend on the construction of the particular resistor you have, but I'd expect anything between just 1W and maybe even 5-10W.

If the voltage is very low (say < 1V) and the operation is short, electrolysis and corrosion isn't a problem. For higher voltages or permanent solutions, you need to properly coat it, and any coating will cause significant penalty in heat transfer; all epoxies and plastic-based materials have poor thermal conductivity. If the internal temperature rises too much, the coating material may be damaged slowly, finally letting water into contact with the leads.

For a properly epoxy coated solution, the end result may not be too stunning. It may be good for only, say, 0.5W, in the end, due to the excess thermal resistance added by the coating.

[AVGresponding]

You'd be better off using either auto coolant or mineral oil, but unless you have a flow it'll not help much, and ultimately the surface area of the resistor will still limit the maximum dissipation quite badly.
Even with good flow, to a radiator,  you probably aren't going to get better than 5x rated power.
Unless you're happy to see the coolant boil ofc...   

Edit: This sounds like a fun experiment!

[mikeselectricstuff]

[RoGeorge]

OK, but careful not to kick the bucket!   

[GerryR]

This is a 50 ohm dummy load used for RF transmitters:

[pk7639]

I can vouch for a 5W ceramic resistor being good for 100W if you drop it in your coffee the moment you realise you got the power calcs wrong!

[SL4P]

I’d be thinking that as the resistor heated up, you’d start getting air bubbles (insulation) forming on the surface.
The best/only reliable solution would have the ‘coolant’ flowing across the surface of the element.

[MLXXXp]

Do you have more than one 1/4 watt resistor of the value you need? If you have 4 of them, you can series/parallel them to make a 1 watt equivalent. You can do the same with more, and even use multiples or fractions of the value, to get as much power handling as you need.

[help_me_pick_username]

Do you have more than one 1/4 watt resistor of the value you need? If you have 4 of them, you can series/parallel them to make a 1 watt equivalent. You can do the same with more, and even use multiples or fractions of the value, to get as much power handling as you need.

I should have more than one... however, isn't there this thing when you put multiple resistors in parallel, the current will just go through the one that has the least resistance...? I only have 5% precision, so this could be an issue, right?

I can always do series though... Thanks for the suggestion!

[Kleinstein]

Precision is not that critical for power sharing. Even of some resistors are 10% off they get some 10% more or less than the average. Not much difference in parallel and series connection.

One may have to include the thermal coupling: so 4 resistors with 1/4 W placed close together can not dissipate 1 W. Especially with SMD resistors it needs quite some board space to get the nominal power rating.

[MLXXXp]

isn't there this thing when you put multiple resistors in parallel, the current will just go through the one that has the least resistance...?

No. If one is slightly lower, it will conduct slightly more current and use slightly more power but only proportional to the percentage of the difference. You might be thinking of diodes.

[MLXXXp]

https://phys.libretexts.org/Bookshelves/University_Physics/Book%3A_University_Physics_(OpenStax)/Map%3A_University_Physics_II_-_Thermodynamics%2C_Electricity%2C_and_Magnetism_(OpenStax)/10%3A_Direct-Current_Circuits/10.2%3A_Resistors_in_Series_and_Parallel

[help_me_pick_username]

You might be thinking of diodes.

I think I was thinking of diodes. Thanks for the clarification!

[schmitt trigger]

This can easily become Photonicinduction - worthy experiment.

[help_me_pick_username]

This can easily become Photonicinduction - worthy experiment.

What's photoinduction?

[Jwillis]

Immersing resistors to increase the wattage ? I would be concerned about  possible steam developing in any cracks or pours causing a small explosion destroying the resistor. I would be more inclined to use thermal adhesive and glue some heat sinks to the resistor if I didn't have enough to parallel as MLXXXp and  Kleinstein suggested.
 
Its Photonicinduction https://www.youtube.com/channel/UCl9OJE9OpXui-gRsnWjSrlA

[ejeffrey]

I once used a 1/4 W axial resistor as a 10 watt heater... Immersed in liquid nitrogen.  Of course as soon as the nitrogen boiled off it turned into a fuse.

[AVGresponding]

I once used a 1/4 W axial resistor as a 10 watt heater... Immersed in liquid nitrogen.  Of course as soon as the nitrogen boiled off it turned into a fuse.

  Good thing nitrogen is relatively inert...


Liquid cooled resistors are a thing. Here are a couple of links, for the curious:

https://en.wikipedia.org/wiki/Liquid_rheostat

https://www.danotherm.com/power-resistors/cooling-method/liquid-cooled

[SeanB]

I used to use a cup of water and a 2N3055 as a dummy load, using a resistance box to vary the base current to get the desired current flow. Was small, and the water would generally last longer than the 5 minutes I needed to do the testing, plus you could always add a little more if needed when running low. 5V rail, so not too much worry about corrosion, and in any case you could use old transistors that are not good for other uses, or where you replaced one of a matched pair.

[Zero999]

This can easily become Photonicinduction - worthy experiment.

What's photoinduction?

He was a crazy Youtuber who used to blow stuff up with high voltages.

Do you have more than one 1/4 watt resistor of the value you need? If you have 4 of them, you can series/parallel them to make a 1 watt equivalent. You can do the same with more, and even use multiples or fractions of the value, to get as much power handling as you need.

I should have more than one... however, isn't there this thing when you put multiple resistors in parallel, the current will just go through the one that has the least resistance...? I only have 5% precision, so this could be an issue, right?

I can always do series though... Thanks for the suggestion!

Yes, connecting resistors in parallel causes most of the current to go through the one with the least resistance.

Note that he said series in parallel. Suppose you have four, 1 Ohm resistors. Connect a pair of them of them in parallel, to make ¹/₂Ohm, now connect another pair in parallel to get another ¹/₂Ohm, put them in series and you now have a 1Ohm resistor with four times the power rating. You could also put two series pairs of 1R resistors in parallel: the effect would be the same: 1Ohm at four times the power.

Don't worry about the tolerances. Some of the resistors will be above an Ohm and others below, so it won't make and difference. In fact, the resulting resistor will have a greater precision, than each resistor, as the errors average out to some degree.

[Siwastaja]

Yes, connecting resistors in parallel cause the current to go through the one with the least resistance.

No, never say this. I know it gets stuck: it's a very common lie that is taught to us early on (speaking of Youtubers, I think Electroboom had a section about it on some of his videos). It has no merit in reality, not even as a simplification, because it is clearly just plainly wrong, and won't provide useful abstraction for any practical use.

I know you know what you are talking about, but phrases like this are extremely confusing, and the beginners must unlearn such widely touted lies.

Your words are clear: "the current" "through the one". This is just plainly wrong. Correctly would be, more current goes through the smaller resistance.

Of course, our complete electrical system is a bunch of things connected in parallel - classically even simple resistances, such as heaters. We all know that turning on a heater does not much affect the others.

You are right that tolerances are indeed not that meaningful, but this goes for both parallel and serial connections, so use whichever is easier for the particular set of resistor you have, and based on the ease of making the actual connections physically.

Unless the tolerances are really crappy, or the system is improperly dimensioned to be really on the edge without any margin whatsoever...

[Zero999]

Yes, connecting resistors in parallel cause the current to go through the one with the least resistance.

No, never say this. I know it gets stuck: it's a very common lie that is taught to us early on (speaking of Youtubers, I think Electroboom had a section about it on some of his videos). It has no merit in reality, not even as a simplification, because it is clearly just plainly wrong, and won't provide useful abstraction for any practical use.

I know you know what you are talking about, but phrases like this are extremely confusing, and the beginners must unlearn such widely touted lies.

Your words are clear: "the current" "through the one". This is just plainly wrong. Correctly would be, more current goes through the smaller resistance.

I appreciate the correction, but resent your implication of lying. It was a simple mistake. I omitted the words "most of". I've edited my post. Don't be so quick to accuse others of dishonesty.

[m3vuv]

why water and not oil??

[Siwastaja]

I appreciate the correction, but resent your implication of lying. It was a simple mistake. I omitted the words "most of". I've edited my post. Don't be so quick to accuse others of dishonesty.

Oh, I mean all the textbooks saying this are lying; this is where it gets stuck to our language from. Doesn't require active "lying" on your part, this wasn't intended. It just comes "automatically", I have said that too.

The authors of such textbooks may even know their stuff, but think that conceptually, a "convenient" lie is always a good idea, as a simplification. Sometimes it is, but it goes too far in this case. Hence, lying.