Posted By |
Discussion Topic:
49 Mercury light switch
-- page:
1
2
|
|
TomO |
08-06-2011 @ 7:10 AM
|
|
|
Senior
Posts: 7250
Joined: Oct 2009
|
!2 volt conversions are custom conversions in that each guy decided how to do it. You may have one reducer for the cluster or one for each gauge. On the temp senders, because you jumpered around the left sender and the gauge went to cold, your right sender seems to be working correctly. If it indicates the wrong temperature, like always cold, it could be the that the wrong voltage is being supplied to it. If you have a volt meter, connect it to the wire from the gauge with the ignition on. If there is a reducer, the voltage will be neat 6 volts. If it is near 12 volts you will need a reducer. Your voltage readings indicate a large drop on the input of the light switch. You could make a temp connection from the battery to the input of the light switch with 12 gauge wire and then see if your terminals heat up. The easiest way to determine wire size is to purchase a wire stripper and crimp tool. They have the wire size next to the stripper hole. Strip a sample of the wire and look to see if all of the strands are there. If they are that is the size of your wire.
Tom
|
49FH8 |
08-06-2011 @ 4:37 AM
|
|
|
Member
Posts: 27
Joined: Jul 2011
|
The best I could tell is the left sender failed the test however I did not remove the wire, but clipped on to it. Both went to cold. The right sender seemed to hang around hot before slowly going to cold. I will retest removing the wires as you originally instructed and see what happens. I will also try the grounding suggestion. I was told when I bought the car the temp gauge used to work...for whatever that's worth. I will try to trace the wires to be sure about not seeing a voltage reducer. Is there typically one for each gauge or one for all the gauges. I did see several under the dash but have not traced all of them. One of them that seemed to have some crusty white stuff around the edge but I was standing on my head on the floorboard and the heat got to me before I could trace the wires. Wire size - How do I determine wire size without a micrometer? I have experience with house wiring 12 and 14 gauge and estimated the size based on that. And some experience with wiring on a motorcycle, but purchased the wire in a specific gauge. It seems difficult to determine gauge since the insulation thickness would skew the observation method. Wouldn't the wire get hot if it was insufficient in size? It is the input terminal on the switch that gets hot first. I appreciate the help. Thanks, Steve
|
TomO |
08-05-2011 @ 7:29 AM
|
|
|
Senior
Posts: 7250
Joined: Oct 2009
|
When you did the tests for the temp senders, what were the results? Did both senders fail the test? You should have a voltage reducer going to the gauges. I missed your reply about the wire size before. Supereal caught it. Your headlight wiring is too light for the amount of current that the bulbs draw. You should have a minimum of 14 gauge wire to the headlights and 12 gauge wire to the switch.
Tom
|
49FH8 |
08-04-2011 @ 5:08 PM
|
|
|
Member
Posts: 27
Joined: Jul 2011
|
I faired the heat and changed the left heat sender and no change with the temp gauge. Changed the left sender and still no change. I don't see a voltage reducer...can that be a problem. The temp gauge is most important right now. Thx, Steve
|
49FH8 |
08-02-2011 @ 6:17 PM
|
|
|
Member
Posts: 27
Joined: Jul 2011
|
I appreciate the help. The weather has been so hot and I haven't checked anything else yet. I will confirm the wire size and check the switch...I need a garage with AC. Thanks, Steve
|
supereal |
08-01-2011 @ 9:51 AM
|
|
|
Senior
Posts: 6819
Joined: Oct 2009
|
Sixteen to eighteen gauge wire is not sufficient to carry the load of even non-halogen headlights. Ordinary headlights draw a bit over 100 watts, and halogens about twice that.
|
TomO |
08-01-2011 @ 7:00 AM
|
|
|
Senior
Posts: 7250
Joined: Oct 2009
|
Steve, You did the test correctly. You have a voltage loss of .43 volts on the input to the switch. You may have a loose or corroded connection in the wires leading to the switch. Your light switch has internal resistance that is causing you to loose almost 1 volt. It should be around .5 volts with a 12 volt system. You can try operating the switch many times to see if you can lower the voltage drop across the switch. If that doesn't work, I would disassemble the switch, clean the contacts and lube them with a corrosion inhibiting grease. Now that you know how to do the test, move your probe to the head light connector. This will give you the total voltage loss to the lights. Because the terminals are getting hot, you should show more voltage drop at the headlight connector.
Tom
|
49FH8 |
07-31-2011 @ 4:23 PM
|
|
|
Member
Posts: 27
Joined: Jul 2011
|
Tom, Finally getting time to do test today. If I am doing this correctly, the results are: Lights on - Input -.43 volts Head lights on - Head -.9v Dash -.94v Tail -.97v Park -12.34v Park lights on - Head -12.32v Dash -.68v Tail -.68v Park -.71v Let me know what this means to you. Thanks, Steve
|
TomO |
07-28-2011 @ 5:47 PM
|
|
|
Senior
Posts: 7250
Joined: Oct 2009
|
Steve, I am sorry but I made an error in describing the meter lead hook up. I described it for a POS ground system and your is a NEG ground system. The COM lead of the meter should be on the POS terminal of the battery and you would probe with the POS lead of the meter. This will give you the voltage loss between the 2 points. A 12 volt sender is usually marked with 12 or 12V. I would not change the right sender if it is working. New parts are not always better than old working parts.
Tom
|
49FH8 |
07-28-2011 @ 1:38 PM
|
|
|
Member
Posts: 27
Joined: Jul 2011
|
Tom...I REALLY appreciate your help! I have 11+ volts at the input of the switch and therefore the same 11+ volts to the lights with switch on. 1/10v output with switch off. I assume I should not have 11+ volts on the input side. What now? Based on your instruction, the left sender is bad but planned to change both. How do I determine if the right sender is 12v. Thanks again for your help, Steve
|