Temp gauge pegged

[Luva65wagon]

On the Comet, which is going to a new home on Wednesday (they will announce that themselves, I'm sure) I have been trying to debug the temp gauge all day.

Here's the history: It (temp) wasn't reading and neither was the fuel gauge. Once I changed the gas tank and sender neither gauges worked still, so I have a collection of constant voltage regulators (CVR) and swapped out a couple until the fuel gauge came to life. Still the temp gauge was COLD all the time.

So I did the checks and could peg the temp side if I grounded the sender wire. Gauge is good. So I checked the sender and it was open.

Got a new sender from NAPA Sunday and installed it tonight and now it will peg the temp gauge if I hook up the sender wire. Thinking I had a bad sender I had another from a later head (1/8" pipe). Hook it up and it pegs it too.

I have other instrument clusters of various years and hook them up one at a time and all peg.

So, this has me totally stumped. I'm pretty sure the sender is the right one (according to specs), but not able to find what cold resistance is supposed to be. I ran it under really hot (insta-hot) water and resistance was close to hot value. Cold is between 220 - 350 ohms on all my cars.

I know something changed in '63, but not sure what. The sender went to 1/8 pipe from 1/4 pipe and resistance was slightly different at 220 degrees:

1962
http://www.napaonline.com/Catalog/Ca...464_0162101090

1963
http://www.napaonline.com/Catalog/Ca...178_0162101093

Anyway I search high and low for an explanation to this phenomenon, but right now I am stumped. I have yet to pull a CVR from one of my other cars, but I have two that do nothing and 2 that do the same thing. I'm assuming the two that are working OK with gas gauge are working, but I will check the voltage output tomorrow. Got too dark to continue.

I know I'm usually the one to answer this sort of question, but maybe someone else may have other senders laying around I can test with - or something else to add. I may have to add temp gauge under dash. This is the last thing I said I would solve.

[Jeff W]

Would it help if I get an ohm reading off my 63 head?

[Jeff W]

296 ohms on the 1965 head I have in the nice warm basement. I could check that without going out in the dark. I'm with you and think a defective CVR would cause the fuel level gauge to not work as well.

[Luva65wagon]

Well I spent a good 4 hours yesterday swapping CVR's on this thing and getting all sorts of odd readings. I also dug out my dad's old analog DVM I inherited and got it working again and did voltage tests. FalconParts.com has a nice diagnostic page on this and I didn't get readings they said I should get on any of the 5 spare CVR's I have. Finally, around 8PM I decided I'm gonna just have a look inside one of these critters. Suffice it to say all of them showed issues.

For those of you who don't know what we're talking about here, a CVR is a constant voltage regulator. It is crude. It is fragile. The way it works is they have this "U" shaped portion with one end of the U going to the gauges and the other end is a contact. That contact sits on another contact, similar to a set of points, and the other contact connects to +12 volts. Wrapped around the arm of the U that has the contact is a winding of fine insulated wire going to ground. When you first connect 12 volts to the thing it is going to feed 12 volts to the gauge, but that's when the function of the wire wrapping the U comes into play. It heats the thing up. This heating causes the U portion to contract and lift one end of the U, the contact end, off the other contact. Voltage to the gauge ceases. As does the voltage to the fine wire. It cools, retracts, and starts to heat again. Rinse and repeat, for ever and ever, amen. The result of this cycle is a voltage ON/OFF swing of about 0-10 volts when all is well. You can adjust the one non-U contact point to add more tension to the U side to increase or decrease the amount of time this thing is in the heat cycle - or putting out voltage. The end goal is an average on-off of about 5 volts to the gauges.

But let me reiterate - these things are crude. Every one I opened up showed shorts in the small wire, burned point contacts, corrosion.

Bottom-line is I need to replace the CVR on the Comet. Then I can trust the temp gauge, and do whatever adjustment to the fuel sender as needed to get it to read the right thing. With the best CVR I had, with the closest adjustment (still irratic) I could make to it - the temp gauge sort of works now, but with 10 gallons of gas in the tank it reads just off empty.



Interestingly these parts vary in price from $20 at Dennis Carpenter all the way to $70 at Dearborn Classics. I think I will buy 3 from Carpenter and replace all of them I have - or at least have the two spare that are new.

[Jeff W]

How about the solid state versions? Seems the way to go.

I think john henry said he would make some for around 23 cents - maybe time to check in with him.

[Luva65wagon]

I've seen those somewhere, but have not had time to look for them.

I know, it seems this could be done with a simple electronic part and I do recall hearing John mentioning that.

Parts already ordered.

[falcon cobra]

I can fix you up with a 5volt regulator that cost a dollar if it will work, it has three pins 12 volts in 5 volts out and a ground ..radio shack has them LM-7805..jh

[Luva65wagon]

But have we tried this with our gauges? Is it this "pulsating function" of voltage these gauges use - or a constant 5 volts? If it were as simple as this, could we not be going into business making these?

[Luva65wagon]

http://www.youtube.com/watch?v=SzLifw0W94g

This is a discussion video on creating one of these. Humm... I could build that in 15 minutes too!

[Luva65wagon]

Though I don't know why he didn't just plop it into the same can and use the can as heat sink. Probably enough surface area.

[falcon cobra]

they also make a 6 volt one if 6 volts is needed. I think I have a 6 volt in stock if anyone wants to try it...jh

[Luva65wagon]

This guy used an 8 volt version - why I have not found out. I'd like to fix this before she comes to get the car tonight, but may not get so lucky.

[falcon cobra]

The guy on that u-tube thing used a 7805-a [ 5 v ], I have a LM-317 which can make any voltage you need 1 to 12...jh

[Luva65wagon]

His video was using a 7805, but he apparently changed to a 7808 because he said the 5 volt unit, under load, wasn't sufficient. So I may swing by RS on the way home and see if I can duplicate just what he did.

[Luva65wagon]

The guy on that u-tube thing used a 7805-a [ 5 v ], I have a LM-317 which can make any voltage you need 1 to 12...jh

I see that you do that by applying resistors to vary the voltage according to some formula. Since we sort of know what the voltage should be, and because this guy already did the ground-work (and concluded the 8 volt version supplies the correct results under load), I'll try his way first and go from there.