Maker Pro
Maker Pro

Current Limiting Thermistor Application?

Wizard Junior

Nov 21, 2009
9
Joined
Nov 21, 2009
Messages
9
So I have this circuit that draws about 7 amps that is powered by a 10 amp DC 13.8V Supply. At startup the load/current draw exceeds the 12amp instantaneous limit and puts the supply in overload protection. Know if I connect different legs of the circuit indivdually it does not surge and cause this overload. This though is unacceptable for my application and I also don't wish to buy another power supply of greater power.

So my question lies in the ability of limitting this surge as in series wiring a current limmiting thermistor on the load, or maybe a RC line suprresor of sorts? Any ideas?
 

Wizard Junior

Nov 21, 2009
9
Joined
Nov 21, 2009
Messages
9
I will add some further info...
The load is multiple car lights powered by a DC supply. The dash lights, head lights, turn signals, and marker lights. All are fused. Some of the lights are turned on and off by switches and also have dimmer controls connected. Its an art project so the exact lights to be on and there brightness will not be determined until installation.

The circuit works fine when connected with a car battery but can not be installed that way because the installation will have power removed in the evenings where it will not have the opportunity to be trickle charged. The installation is for about a month that all these lights will be on about 12hrs a day. So a DC supply was chosen as the source.
 

Resqueline

Jul 31, 2009
2,848
Joined
Jul 31, 2009
Messages
2,848
The problem is caused by the temperature coefficient of the lamps.
For a few milliseconds when first turned on they draw at least 5 times as much current as when operating.
There are several solutions:
1: Turn on the lamps one by one, headlights first. (If they still trip it then switch on several times quickly.)
2: Put a big capacitor (or a small battery) on the output of the PSU. Have a switch between the cap (bat) and the installation.
3: Put a 7A NTC surge current limiting resistor in series with the installation. (It'll get hot.)
4: Modify the PSU. (Soft-start, or delayed overcurrent protection.)
 
Last edited:

Wizard Junior

Nov 21, 2009
9
Joined
Nov 21, 2009
Messages
9
Thanks Resqueline for the advice.
As solutions one and two should work I am limited to one switch in total have turned on for this application. That switch is turning on many other things besides this circuit so an additional switch will not be possible.

#4 option will be my last resort because if for some reason there is failure of the PSU and it causes a fire or some other damage the supply will no longer be under warranty because it sounds as if both require internal alterations of the unit.

#3 is what I will try finding a surge limiting resistor. As for heat...I have no issues if it gets warm but hot I would like to avoid. So for clarity I would prefer it not reach over 180 degrees farenheit or lets say a temperature that could burn you if touched. If I double the value of the NTC surge current limiting resistor as in a 15 amp or more I would hope that this may be possible. Is this a feasible assumption or would something like 4 times the rated value be a better idea?
 

Resqueline

Jul 31, 2009
2,848
Joined
Jul 31, 2009
Messages
2,848
So you are commited to use a central mains switch panel if understand correctly?

The Negative Temperature Coefficient resistors work just because their temperature rises during operation so there is a limit to how oversized it can be and still work.
It seems they can reach up to 250 deg C, so you'll need to built it into a box then to get it under 82 deg C.
They have a lower & an upper continuous current rating. You'll want one that has an initial resistance of around 1 ohm (12V/12A) and an operating current range that incorporates 7-10A.
For example an 8A rated one won't work properly below 3A.
Googling ntc surge (or inrush) will provide several datasheets.
 

Wizard Junior

Nov 21, 2009
9
Joined
Nov 21, 2009
Messages
9
Yes I am committed to use a central main switch.

I was able to couple one set of lights that were always had full potential to a rheostat. Meaning I had the headlights on a rheostat and the turn signal lights fully on including other lights and the power supply at start-up went into surge overload. After some manipulation of the circuit changing the turn signal lights also to be connected with the head lights on the rheostat was enough of a change to allow start-up without an overlad. This is just a temporary fix while waiting for the NTC resistor to arrive...which will be ordered tomorrow.

Another interesting occurence that suprised me a little was when I originally tested the amp draw of the circuit it was without fuses, switches, and rheostats. The figure I measures was 7amps so I used that to source a power supply and got a 10 amp one. Now with everything on at its maximum mu current draw is slightly over 10 amps at 10.2. I figured on a slight increase in power but not as much as 50%. I suppose my circuit building skills are a little rusty. So this is a clear indication also why my supply is going into overload also. With the rheostat adjusting a few of the lights and the main amp draw below 8 at this time I am not overloading the circuit.

Thanks for your replies Resqueline and if anything changes or when the NTC resistor arrives and is installed I will keep updating.
 

Resqueline

Jul 31, 2009
2,848
Joined
Jul 31, 2009
Messages
2,848
Ok, good to hear you got it going, rheostats will negate much of the temperature coefficient of the lamps.

I'm not sure why you got a higher draw now than when you first tested it. Fuses & rheostats only contribute to a lesser draw, and only lighted swithes would add 0.1A draw per switch. The only thing could be if you tested it using a run-down car battery delivering considerably less than 12V, and now it gets 13.8V from the PSU.
The extra load could also be accounted for if you added a couple of turn lights afterwards.

But; if I understand the PSU manual correctly it is only rated at 7A continously, 8A for 25 minutes, and 10A for 15 minutes. So I suggest you keep some rheostats in there, or try to get hold of some lesser powered bulbs (there are many alternatives).
 

Wizard Junior

Nov 21, 2009
9
Joined
Nov 21, 2009
Messages
9
For know the rheostats are going to stay. As for the change in current I at this time account it to or contirbuted by a low input power. I checked and I was only getting 103VAC. The final location is not where it is being tested now so when installed it should have the appropriate 120VAC. So currently this project is being taken apart and moved. It will be installed next week. So hopefully with a NTC resistor installed(arrives Monday), proper input power, rheostats installed, and a safe transport all turns out well.
Thanks for your replies.
 

Wizard Junior

Nov 21, 2009
9
Joined
Nov 21, 2009
Messages
9
So the NTC resistor worked great in the circuit. The power supply though is not strong enough.
 

Resqueline

Jul 31, 2009
2,848
Joined
Jul 31, 2009
Messages
2,848
Sorry to hear that. It's unfortunate that it's sold as a 10A unit when in reality it's only 7A.
The increased mains voltage at the new site also means increased heating of the PSU.
Unless you can get hold of a better PSU I suggest you use two of them, or try to replace some bulbs.
If you have 55W head lights they might be replaced with 35W units.
If you have 21W turn lights they can be replaced with 15W bulbs, or even better; LED bulbs (they are 3W or less).
If you have 5W marker lights they can be replaced with 2W bulbs, or also LED bulbs (they draw 0.5W).
Dashboard lights can also be replaced with LED bulbs.
It won't take much to get below the critical limit of the PSU.
 
Top