NTC in parallel.

Thanks to the excellent threads in this section, I am gradually learning about modding hotwires.

I am not up to speed on electronics but can use a soldering iron.

Now about NTC use. We want an ohm of resistance when cold so the bulbs
do not flash. Got that. The selected NTC must drop its resistance when hot so we get maximum voltage to the warmed up filament and best light output.
Got that.
Now where it gets confusing to me is how much the resistance will drop.
There seems to be a wide tolerance in the NTC ( SL12-1R010) .
So if I select a NTC that is one ohm cold and 0.060 ohms hot, what will these readings be with two NTC in parallel??. Also I have read about fixed resistors in the circuit as well,yet the author of the post did not explain the benefit of an NTC and fixed resistor in parallel.
Thanks

Anglepoise said:

Thanks to the excellent threads in this section, I am gradually learning about modding hotwires.

I am not up to speed on electronics but can use a soldering iron.

Now about NTC use. We want an ohm of resistance when cold so the bulbs
do not flash.

Click to expand...

Why one ohm? That should depend on the voltage of the batteries, as well as operating current on the particular lamp you're using. I suspect you'd want to pick a value such that the total current won't change much, IE you have a 12V 60W lamp, that runs at 5A when heated up, you'd probably want a resistor around 2 ohms (12/5 = 6A, assuming "zero" resistance for cold filament)

Now where it gets confusing to me is how much the resistance will drop.
There seems to be a wide tolerance in the NTC ( SL12-1R010) .

Click to expand...

I'd say your best bet is to buy several extra, then connect them to a bench power supply, and actually measure the voltage drops for yourself at different voltages. Particularly for things like parallel configurations.

So if I select a NTC that is one ohm cold and 0.060 ohms hot, what will these readings be with two NTC in parallel??. Also I have read about fixed resistors in the circuit as well,yet the author of the post did not explain the benefit of an NTC and fixed resistor in parallel.
Thanks

Click to expand...

There is no easy way to predict this at all, as NTCs are non linear, and vary wildly based on temperature. The following is purely conjecture:

Cold, the resistance will be 1/2 ohm with two in parallel.
Hot, it's difficult to say -- with two in parallel, eventual resistance shoudl be lower. However, since the heat dissipatd in each will be spread out more, temperatrue won't increase as much, which means the final resistance might be less than if there only one. My guess is that resistance of 2 identical NTCs in parallel woudl be still less than a single, but more than 1/2 (what you would see with "normal" resistors in parallel). You still may be better off though, as heat/high temps shoudln't be as much of a problem.

As for a fixed resistor in parallel, the only reason I could think to do that is if you want to have significantly less "cold" resistance. I woudln't expect it to change much as far as the "hot" resistance is concerned, though. In this application, I don't think you'd want that.

I don't think it would work very well to put NTCs in parallel because they would tend to get out of balance.

Suppose both NTCs start out with equal currents. If one of them got just a little bit warmer, its resistance would drop and it would take more current, increasing its temperature even more. So it would take still more current, get hotter still, take more current, and so on. At the same time, the increased current through one NTC would leave less current flowing through the other one, so it would get cooler, increase its resistance, take less current, get cooler still, and so on.

The net result is you would be likely to end up with the NTCs hot and low resistance and the other cool and high resistance. It might be possible to reduce this imbalance if they were tightly coupled thermally, but I don't know how this would work out in practice.

Thank you both for your insight.
Some good things to think about.