Load Tester / Dummy Load

Geek

Well-known member
Joined
Aug 15, 2017
Messages
934
I was pondering about a dummy load for testing of various batteries. Looking around high wattage resistors seem to work out more expensive than lower wattage ones.

So I decided to do some tests to see how much power a 1/4 wattresistor can handle. I started with a 470 ohm metal film, being the closest I had to the max rated power at 1/4 watt. It did get warm, but it did not bother it, even after hours. That works out around 0.3 watt, just a little over its maximum.

The next was a 220 ohm metal film resistor - at 0.6w that's just shy of triple its rated power. It got hot, and turned a nice shade of brown, but it did not seem damaged in any other way, after 4 hours or so.

So, pondering to myself, I decided to test a 100 ohm carbon resistor. It caught fire instantly. But aside of being scalding hot, it also put up with this punishment for 4 hours, still measured 100 ohms!

So... aside from an amusing story, what does this all mean to me.

Well, how about instead of buying 2x20w 8ohm resistors, why not buy 100x 1/4 watt 470ohm resistors. It would be very easy to assemble, as resistors come neatly attached to tape, spaced just nicely. I also imagine it would dissipate heat very easily.

Is there any reason why this would not work?
 

ozz93666

Member
Joined
Feb 22, 2017
Messages
345
Yes that should work...

... should remember the wattage rating is not just about when the resistor will destruct ,it's about the resistor maintaining the specified resistance at that power ...

But extreme accuracy of resistance is not necessary for load testing...

Rather than wait for 3 weeks for delivery of resistors I have used scrap wire that was lying around ... a certain type of domestic extension phone wire is good , it's made of copper coated iron , and has a high resistance , each wire looks about 0.25mm dia , about 4 meters of this works fine , wound into a neat coil , it only gets moderately warm for 1A discharge (3.6V) , you can also easily change the resistance by adjusting the length ...

a winding from a scrap transformer will work well too , or an armature coil from a dead electric mains drill.

incandescent bulbs give a nice visual indication of discharge , either 12V car side lights , or torch bulbs , put them in series of parrell to get the resistance you required depends what you have in your scrap box..
 

Geek

Well-known member
Joined
Aug 15, 2017
Messages
934
ozz93666 said:
Yes that should work...

... should remember the wattage rating is not just about when the resistor will destruct ,it's about the resistor maintaining the specified resistance at that power ...

But extreme accuracy of resistance is not necessary for load testing...

Rather than wait for 3 weeks for delivery of resistors I have used scrap wire that was lying around ... a certain type of domestic extension phone wire is good , it's made of copper coated iron , and has a high resistance , each wire looks about 0.25mm dia , about 4 meters of this works fine , wound into a neat coil , it only gets moderately warm for 1A discharge (3.6V) , you can also easily change the resistance by adjusting the length ...

a winding from a scrap transformer will work well too , or an armature coil from a dead electric mains drill.

incandescent bulbs give a nice visual indication of discharge , either 12V car side lights , or torch bulbs , put them in series of parrell to get the resistance you required depends what you have in your scrap box..

Resistance of an incandescent bulb is too inconsistent. As the bulb heats up the resistance changes dramatically. Same for nichrome wire.

I am considering 220 ohm resistors submerged in oil.
 

Redpacket

Well-known member
Joined
Feb 28, 2018
Messages
1,239
Guess it depends if we're trying to do controlled & measured discharge vs "whack a large load on the pack" type test :)

Were 1W or 5W ones still too expensive?
 

Geek

Well-known member
Joined
Aug 15, 2017
Messages
934
Redpacket said:
Guess it depends if we're trying to do controlled & measured discharge vs "whack a large load on the pack" type test :)

Were 1W or 5W ones still too expensive?

Controlled discharge. Otherwise a globe would be perfectly suitable.

Too expensive? No. Just cheaper for 1/4 watt resistors.
 

Geek

Well-known member
Joined
Aug 15, 2017
Messages
934
Second test:


image_zrwvly.jpg


This load tester has 12v applied from a lead acid car battery. I know that the 270ohm resistors are operating way beyond their spec, yet they hold up just fine. Later when I full charge the battery they will discolour. But for the time being they handle 12v just fine. The whole array tests at 15 ohm.

This time I will run the test for 10 or more hours and see what happens. It is fused with a 2a fuse in case one of the resistors goes short circuit.

Wishing I had a decent video camera to do a time lapse.
 

dc443

New member
Joined
Dec 10, 2020
Messages
19
Curious what your results were.

Sorry for the thread necro but I'm strangely attracted to the concept of using simple mechanisms to measure battery or pack capacity.

Now I really do like the concept of incandescent bulbs as a load. But, a microcontroller would be needed in order to do any kind of coulomb metering since the current and voltage are both varying and need to be time-integrated.

Now... controlled discharge with a fixed resistance can make the curve easier to reason about, but I think you're still going to have the actual current decrease along with voltage while the cell/pack drains, so the amount of time it takes to reach cutoff voltage is still not totally linear to capacity as it were. It may be good enough to use for a relative measurement, though.

So I am hoping that it would be possible to have a very simple circuit with a flexible load (be it resistor banks or incan bulbs) that can be used to make inaccurate but usable output in units of time duration that can be used for comparative purposes to gauge capacity.

The thing i'm left scratching my head over is how to implement the low voltage cutoff in an accurate way. I think since I have a good number of relays on hand it'd be cheap to make, but in the general case, I'm not so sure.
 
Top