Headlight Switches

[Thomas H.]

Re: Headlight Switches

A slow-blow fuse might be okay, but the best thing to do is find a good headlight switch. From what I am learning, I believe GM put a lot of thought into this design, and they did it just right. They could not have done it any better, given the technology at that time. I would love to know how new cars do it.-Dan-

Dan,

You touched on the slow blow fuse idea and you are correct in thinking that they are used in circuits where the initial power on inrush current is high. Fuse opening is based on what is called an I2T (or I squared T) curve. All fuse manufacturers publish the I2T curves for their fuses. The difference between the fast, medium and slow acting fuses is the period of time that elapses before the fuse opens based on the square of the overload current.

I can not say with 100% certainty but given the proliferation of microcontrollers in todays automobiles I would think that the manufacturers may employ some form of "soft start" algorithm when high inrush circuits are initially powered on. If they are using a MOSFET to switch the power to the circuit they can easily PWM the gate of the FET for a very brief period ramping up the voltage to the load limiting the inrush and saving the wear and tear on the fuse.

I have actually run tests where we witnessed fuses breaking down with repeated power on cycles to high inrush loads. Not only could you observe the fuse element deform, but the measured resistance of the fuse would increase limiting its ability to ride through the inrush event. Eventually the fuse would open even though it's current rating was well within the limits of the circuit being protected.

Lamps, motors, some LED circuits and items with a lot of input capacitance (some LCD displays) are all examples of circuits that will exhibit higher inrush currents at power up.

Tom

[Dan D.]

Re: Headlight Switches

Great Rich,

You will want to use a very low value resistor so as not to limit the peak current. I have a 0.1 ohm, 100 watt, 1% resistor that I plan on using. 0.1 ohm is a good value to use if you can, as it is easy to compute current from the voltage drop. You would need about 20 watts in order to handle the steady state 13.2 amps you measured with your high beams, and 1% for accuracy. Below is a picture, setting on my drill press, of what I plan on using, and a second picture showing the current state of my new load bank. As you can see, I have plenty of work to do yet.



-Dan-

[Thomas H.]

Re: Headlight Switches

Got a chance to take some inrush measurements on the 60 today. I connected a 100 amp (75mV) shunt between the battery - and the ground cable. Connected a scope across the shunt to measure the voltage developed across the shunt when the lights are switched on. Results:

Low beams only -

no highbeams.bmp

Peak is approx 26 amps and decays to normal operating current over a period of approx 200ms.

High and low beams -

with highbeams.bmp

Peak is approx 32 amps and decays to normal operating current over a period of approx 300ms.

I had better screen captures as I dialed in the trigger settings - but for some silly reason, the scope didn't store them like I wanted it to. Probably operator error.........

These tests were done without the car running and the battery voltage at 12.6 volts.

Tom

[Dan D.]

Re: Headlight Switches

Thanks Tom,

I guess the peak current is less than I thought it would be. I would like to know what the steady state current is. Also would like to know it with the battery voltage around 14V with the engine running, as this represents the actual conditions the switch will see when driving, and will let me better determine what the breaker current should be set at.

Do you think the narrow negative spikes are contact bounce in the headlight switch, and the dimmer switch with high beams? Guess it has to be that.

Are these the normal T3 type incandescence type bulbs (not halogen)? Rich M. measurements with halogens show double the current draw with high beams. This sounds right as you have 4 headlights instead of 2. With single headlights current draw is actually a little less with the high beams (repro T3s). The person at Lectric Limited confirmed this with me. High beams just focus higher up is all. It is easy to see why all cars went to 4 headlights in 1958. You have twice the Lumens.

I have been out of business for a few days. But I now have a new power supply and I have re-designed/rebuilt my resistor load bank. I had some family matters that occupied my time as well.

I have enclosed some pictures of my new load bank and test set-up now that I am back in business.

On test is an original 57 switch pulling 15 amps through the breaker. This particular switch, untouched and fairly corroded, will hold at 20 amps and open at 21. Very little hysteresis. I am surprised - the repro switches I have tested exhibit several amps and are hard to get an accurate measurement. This corroded 60 year old switch works better than any of the new ones I have tested so far.

More to follow.



-Dan-

[Thomas H.]

Re: Headlight Switches

Thanks Tom,

I guess the peak current is less than I thought it would be. I would like to know what the steady state current is. Also would like to know it with the battery voltage around 14V with the engine running, as this represents the actual conditions the switch will see when driving, and will let me better determine what the breaker current should be set at.

Do you think the narrow negative spikes are contact bounce in the headlight switch, and the dimmer switch with high beams? Guess it has to be that.

Are these the normal T3 type incandescence type bulbs (not halogen)? Rich M. measurements with halogens show double the current draw with high beams. This sounds right as you have 4 headlights instead of 2. With single headlights current draw is actually a little less with the high beams (repro T3s). The person at Lectric Limited confirmed this with me. High beams just focus higher up is all. It is easy to see why all cars went to 4 headlights in 1958. You have twice the Lumens.

I have been out of business for a few days. But I now have a new power supply and I have re-designed/rebuilt my resistor load bank. I had some family matters that occupied my time as well.

I have enclosed some pictures of my new load bank and test set-up now that I am back in business.

On test is an original 57 switch pulling 15 amps through the breaker. This particular switch, untouched and fairly corroded, will hold at 20 amps and open at 21. Very little hysteresis. I am surprised - the repro switches I have tested exhibit several amps and are hard to get an accurate measurement. This corroded 60 year old switch works better than any of the new ones I have tested so far.

More to follow.


-Dan-

Wow - someone has been busy.......... Nice set up.

Regarding the testing I did, I was also expecting to see more inrush then I measured. Some of that may be due to the overall resistance inherent in the original harness.

I could not run the engine while taking these measurements as the constant switching action of the voltage regulator would trigger the scope. The inrush based on a higher voltage can be extrapolated from the measurements I took. For the high beam test, Batt V(12.6) / Measured Current (32A) = 393 milliohms. (I'll have to take a headlamp into work and measure the filiment on our milliohm meter). Doing some more ohms law to find the current at an operating voltage of 14 would be 14/.393 = 35.6 amps.

I would like to do a bench test with just a single lamp and a battery, but that is a test for another day when I have more time to assemble a better test setup - similar to what you are doing.

Fun stuff.

Tom

PS: the car has repop low T3 beams that I installed and, I believe, the original T3 high beams.