Maker Pro
Maker Pro

IC Voltage Regulator Question

vwadad

Feb 8, 2013
3
Joined
Feb 8, 2013
Messages
3
Hello Forum Members,

I have an old Jeep that originally had a 6 volt electrical system. That vehicle was converted at some point to a 12 volt system. I am installing an original 6 volt fuel gage, and therefore need to drop the voltage in the fuel gage circuit (prior to the gage) to 6 volts. In order to do this I am using an NTE962 IC voltage regulator. The wiring seems simple enough, but I am having problems. With no load, Vin of 12V and the center terminal of the regulator grounded as required, there is indeed 6 volts as measured between the Vo terminal and ground. But with any load applied to Vo, the voltage immediately drops to about 1 volt. This happens when the fuel gage is connected (approx .5A), or when a small 6V bulb is the load. The NTE962 is attached to a piece of aluminum as a heat sink, so I don't think temperature is a problem.

Any suggestions?

Thanks, Brian
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
One:
The datasheet says "no external components required". But with regulators of this type it is generally a good idea to put a 10µF tantalum capacitor from input to GND and one from output to GND, placed as close as possible to the regulator. Otherwise the regulator is prone to oscillations which, depending on your measuring equipment, may show as supposedly low output voltage.

Two:
The datasheet doesn't tell to which potential the heat fin of the chip is connected. Check (using an Ohmmeter) this connection - it is most probably to GND. Then verify that the heatsink is not conneted to any other potential. Both the heat fin of the chip and the heatsink need to be on the same potential. Otherwise you will need an insulator (electrical, not thermal, e.g. mica) between the two.
 

john monks

Mar 9, 2012
685
Joined
Mar 9, 2012
Messages
685
Question: why not simply use a series resistor or a 6 volt Zener diode in with your fuel gage?
I was under the impression that the old fuel gage used a bimetal strip that heated up and moved a needle. So the only thing to be concerned with was the maximum current when you have a full tank. Am I mistaken?
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
The NTE962 seems to have a current limitof about 1A, this may not be enough for the gauge you have and the voltage drops as a result.
There were two types of fuel gauge, one was a moving coil gauge which responded very quickly, this needed a smooth supply.
The other type was a thermal gauge which responded slowly as described. I do not remember which the jeep had.
In later vehicles the thermal gauge was often used on 12V with a bimetallic stabiliser to give an average of 6V. You may be able to find one of these.

Measure the resistance of the gauge to chassis to determine the current required.

Dropping the voltage with a resistance would affect the calibration.
 

vwadad

Feb 8, 2013
3
Joined
Feb 8, 2013
Messages
3
Gentlemen,

Thanks for your responses. In this old (1948) Jeep the position of the float attached to the variable resistance sending unit in the fuel tank changes with fuel level, thereby changing the resistance in the circuit. There is not a thermal aspect to this design. Because the resistance of the circuit is not constant, the series resistor is not the best approach. However, it is my "Plan B" if I can't make the VReg work properly!

I do believe the gage draws no more than .5A.

I will investigate the capacitor issue that Harold brought up, and read up on Zener diodes.

Is there an easy way to connect the capacitors to the VReg? I can't see soldering them all together. Some sort of board to mount them to?

Thanks again,

Brian
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
Hve you checked that there is no thermal aspect? If the gauge operates rapidly, there is not. If the gauge creeps to its final reading, then there is. Thermal meters were used because the slopping of fuel in the tank made reading the level on the move very difficult.

You should measure the current used by the circuit. The difficulty here is finding a 6V supply, perhaps a lantern battery.

The way I would mount a regulator would be on stripboard (Veroboard). The board is perforated with holes spaced at 1/8in (2.54mm) and there are copper strips on the back connecting lines of holes. Most components have leads spaced to fit.. The copper strips can be cut with a drill but I do not think you will need to do this with your simple circuit. You can get screw terminals which can be soldered to the board.
Make the board a bit bigger than the circuit so that you have something to fix it to. I always make my boards too small bacause of my religion - I am a devout miser!

The regulator may need a heat sink to take away excess heat. Most regulators will shut down if they get too hot but the cooler they run the better. A finger test should be adequate.
1A at 6V drop = 6W. A TO220 style chip will stand about 1W without a heat sink.
The tab on the regulator will be live so the heat sink should be electrically insulated or the regulator should be fitted to it with an insulating kit.

We used to have a US navy jeep on the farm, when the fuel tank leaked we found that it was surrounded by a mixture of sand and rifle shells. Not a very safe seat!
 

vwadad

Feb 8, 2013
3
Joined
Feb 8, 2013
Messages
3
HelloDuke37,

You are probably correct that I should measure the current in the circuit. Perhaps my .5A estimate is wrong and the VReg is being maxed out. I will measure the current next week when I have some time, and let you know.

Thanks,

Brian
 
Top