Dummy Load for Bench Power Supply - Resistors?

[DW1961]

I'd like to load a bench power supply using some sort of simple dummy load. I was thinking about a couple of 100 watt resistors. I can get two of them for 8 bucks.
https://www.amazon.com/Twidec-Wirewound-Resistors-Equipment-RX24-100W1RJ/dp/B07QY6365F/ref=pd_di_sccai_5/137-6696668-9948413?pd_rd_w=aY7wU&pf_rd_p=c9443270-b914-4430-a90b-72e3e7e784e0&pf_rd_r=P69W3XKZ8GKEE2Z97PQ4&pd_rd_r=133a564b-73a6-42d1-918e-03d37798984e&pd_rd_wg=haTE7&pd_rd_i=B07QY6365F&psc=1

I'm not sure about the Ohm rating or other things I would need to use a simple resistor for a dummy load, however. Is there a better way that is just as easy and cheap?

Is there anyway to run two identical resistors together for x2 power resisted?

[Benta]

Voltage? Current? 1 ohm won't get you far IMO.

[tunk]

I'm sure you already know that you also need to attach the resistor(s)
to a large heatsink. If you e.g. have a 30V/5A PSU, then you need a
150W 6 ohm resistor (R=U/I=30/5=6, W=U*I=30*5=150).

[Gyro]

I have a selection on self (radiation/convection) cooled tubular rheostats that I use. The chunkiest is 3.3R 10A... 330W @ 33V.

These are the sort of things that used to be used in school labs. Maybe try ebay. Note, the single turn rotary ones tend to have lower wiper ratings than end to end of the winding.

[DW1961]

Voltage? Current? 1 ohm won't get you far IMO.

I could probably do with 30V 5A but if it isn't too expensive 30V @ 10A would be all I would ever need.

[DW1961]

I'm sure you already know that you also need to attach the resistor(s)
to a large heatsink. If you e.g. have a 30V/5A PSU, then you need a
150W 6 ohm resistor (R=U/I=30/5=6, W=U*I=30*5=150).

It would only be running for less than a minute. Still? I should probably just build a cheap one using a small 1 x 4 and then a heat sink. Any ideas?

Would a 150W 8 Ohm work?

Also what about running smaller wattage resistors in a series? Several smaller ones are cheaper than a single large.

[Gyro]

If you're load testing a PSU you need to load it for long enough for it to reach a steady state temperature. A few seconds won't tell you whether it can sustain a given load current without frying.

[Benta]

30 V at 5 A is 150 W, at 10 A it's 300 W.
That heatsink is going to be one heck of a mother. You're more likely looking at forced cooling, either with fans or water.
A better idea than resistors could be heating elements in that power range. For example Diesel engine preheaters from a salvage yard. Should be cheap.

[Jwillis]

Even a cheap 180W electronic load from Banggood would give you a better indication of the potential of you power supplies and batteries. They're not bad . I have a couple of them . One I put a larger MOSFET and Heat sink on and one I put a just larger heat sink on and both will handle 150W easily . For anything above 150W they might be questionable . I have not tried either at that power. Built an   >800W for over 150W DUT .

[Wallace Gasiewicz]

You could go with the old Cantenna approach and use a big resistor 6-8 ohms and wire the resistor into a paint can with mineral oil.
Not related to the WiFi cantenna.
You could also put say 12 100 ohm 5 watt resistors in parallel in the can and end up with 8.3 ohms maybe DC to 50 MHz depending on the resistors.
12    5 Watt resistors in parallel give you a 60 Watt load
12  10 Wall resistors equals 120Watt load etc
Bigger resistors equals more watts as do more resistors in parallel.
These resistors are cheap and can be configured into any ohm load.
You can run these type loads for at least several minutes.
Heathkit made one and other manufactures still sell the same thing. These are 50 ohm loads customarily.

https://www.orcadxcc.org/content/cantenna_va7jw.pdf

Hams have been using this Homebrew load for at least 80 years for stuff up to one KW or more. I use one.
They solder one end of the resistors to a copper rod or maybe a 12 g wire and connect that to a RF connector. The can itself is the negative. as is the outer RF conductor.
You could use any DC connectors you like .
Customarily, there is a relief valve on top which is just a little hole covered by a stiff rubber flap.
Just don't leave it on for too long, it will heat up. Put it on a pie plate for any possible mineral oil escape. Transformer oil these days is better than mineral oil and is silicone. Good if you can easily procure some. Old transformer oil had PCBs.

[Gregg]

Automotive filament light bulbs can be series / parallel wired to get just about any load in your range.  I personally have a bunch of H4 headlight bulbs with the low beam filament burned out that I use for such purposes.

[DW1961]

If you're load testing a PSU you need to load it for long enough for it to reach a steady state temperature. A few seconds won't tell you whether it can sustain a given load current without frying.

It's true. I really do need a longer time. I'll just have to do what it takes.

https://www.amazon.com/Resistor-Aluminum-Wirewound-Resistors-Chassis/dp/B08G8JGW1M/ref=sr_1_43?dchild=1&keywords=150W+6+ohm+resistor&qid=1635802995&sr=8-43 Those type have their own heat sink. I could just run a 12 computer CPU fan over it in open air, right?

[DW1961]

30 V at 5 A is 150 W, at 10 A it's 300 W.
That heatsink is going to be one heck of a mother. You're more likely looking at forced cooling, either with fans or water.
A better idea than resistors could be heating elements in that power range. For example Diesel engine preheaters from a salvage yard. Should be cheap.

What about running x2 150 watt in a series?

[DW1961]

Automotive filament light bulbs can be series / parallel wired to get just about any load in your range.  I personally have a bunch of H4 headlight bulbs with the low beam filament burned out that I use for such purposes.

Will I need a welders glasses at some point?

[DW1961]

You could go with the old Cantenna approach and use a big resistor 6-8 ohms and wire the resistor into a paint can with mineral oil.
Not related to the WiFi cantenna.
You could also put say 12 100 ohm 5 watt resistors in parallel in the can and end up with 8.3 ohms maybe DC to 50 MHz depending on the resistors.
12    5 Watt resistors in parallel give you a 60 Watt load
12  10 Wall resistors equals 120Watt load etc
Bigger resistors equals more watts as do more resistors in parallel.
These resistors are cheap and can be configured into any ohm load.
You can run these type loads for at least several minutes.
Heathkit made one and other manufactures still sell the same thing. These are 50 ohm loads customarily.

https://www.orcadxcc.org/content/cantenna_va7jw.pdf

Hams have been using this Homebrew load for at least 80 years for stuff up to one KW or more. I use one.
They solder one end of the resistors to a copper rod or maybe a 12 g wire and connect that to a RF connector. The can itself is the negative. as is the outer RF conductor.
You could use any DC connectors you like .
Customarily, there is a relief valve on top which is just a little hole covered by a stiff rubber flap.
Just don't leave it on for too long, it will heat up. Put it on a pie plate for any possible mineral oil escape. Transformer oil these days is better than mineral oil and is silicone. Good if you can easily procure some. Old transformer oil had PCBs.

Wallace, thanks for all of that. Would lower wattage resistor in parallel run cooler than a single large resistor?

[beanflying]

Might be a bit over Engineered but it works  The Resistors are 50W and I built it as a high power decade as much as for a Dummy Load so 1,2,3 & 4  wired with bars so I can series/parallel or parallel two 5 pairs for maximum power. Copes easily and stays coolish at 7V and 15A

[ledtester]

You might find this video interesting/helpful:

Home Lab Equipment: Making sets of Power Resistors - TheHWcave
https://youtu.be/Fxp0nxphgYw

Includes a discussion of heatsink requirements.

Also mentions that a bank of power resistors is useful for testing AC power supplies.

[DW1961]

Got a question about doing this with resistors. After doing some reading, it looks like the resistor load tester will always pull it's maximum wattage from the power supply, or try to. If I tried to limit the current, for example, at the power supply, and the resistors were asking for more power, I'd get an OC on the PSU? In other words, if I built a resistor load tester for 30V/10A 300 watts, and my power supply was set to 150 watts, it would OC (or trip the protections)?

[ledtester]

Those smaller "100W" resistors definitely need a good heatsink... here's a pic comparing them to a 8-ohm 120W power resistor made by Dale in the 70s:

[wizard69]

That is one pretty load.

[beanflying]

Got a question about doing this with resistors. After doing some reading, it looks like the resistor load tester will always pull it's maximum wattage from the power supply, or try to. If I tried to limit the current, for example, at the power supply, and the resistors were asking for more power, I'd get an OC on the PSU? In other words, if I built a resistor load tester for 30V/10A 300 watts, and my power supply was set to 150 watts, it would OC (or trip the protections)?

You need to think of each single resistor being capable of dissipating X watts depending on how you wire them together. ** Also assumes they are properly heatsinked.

Below is some figures I did a while ago on my home made decade boxes (the top two are not relevant and not shown). The two lower figures are for the test setup shown above at  2.5  the limit is the 4 resistor in one of the legs being 3.4A for maximum 50W (I^2*R=50W) dissipation. So using this and basic Resistor theory the box will run at twice this current or about 7A throught the two 5 legs. If you then apply Ohms Law you get a maximum voltage of 7x2.5 or around 17V

The other configuration which is the current and voltage measurements you see above is all four resistors paralleled so around 0.5  with the limiting resistor being the 1 in this case so maximum current (I^2*R=50W) is circa 7A. In this setup the current/power is shared across the stack but the others will be handling less of it.

[TERRA Operative]

As a rough rule of thumb, those aluminium cased power resistors are basically halved in power rating when not attached to a heat sink.

[David Hess]

Cement power resistors have the lowest price per watt so I find the best deal on 5, 10, or 20 watt resistors in a value that I can series and parallel to get the resistance and power rating that I need.

[BrokenYugo]

Incandescent lamps are cheap.

[Wallace Gasiewicz]

A bunch of lower (not TOO low) resistors help a little in that the coolant surrounds more of the resistor. the idea is more surface area with a bunch of smaller resistors.
Probably unnecessary if you can find bigger resistors cheap.
Look on ebay and see the price point, I think it will favor the 5 or 10 watt cement resistors, probably 100 ohms.
Someone had a model of one of these using resistors soldered to can lids in two levels also. there were many resistors in each level, so he made the load with parallel and series. Then put it into a paint can with mineral oil.
You can also use a bigger can like a metal 5 gallon can but it becomes heavy.
These things are made for 50 Ohms but you can build one for whatever Ohm you want.
The can lids kept the resistors away from each other to maximize the coolant contact. Lots of people actually made and used and liked this thing.
Someone brought up incandescent bulbs:
Electric incandescent light bulbs work but resistance changes with different  voltages  when they heat up. This idea was also used by Hams a lot because it was cheap.
This works but it is not very stable resistance. Also would require a bunch of incandescents in parallel. But it does light up!!!!l
I think there is also a rush of current as the bulb light up.
AC incandescent bulbs work just fine with DC. Also heat sinking is less of a problem since they are made to run hot.

Remember Resistance is not Futile. Resistance is voltage over current . Refer to OHM.   HA!!!