Temp Gauge or Overheating??

[Green Machine 79]

Hi folks,

I have a 1979 C-20 5.7 350.

It sat on edge of my farm field for near 3 years before i finally got around to resurrecting it a few weeks ago. Just needed a new ignition coil and distributor cap to get it running.

BUT, now the temp gauge goes to hot within 60 seconds of starting up.

I switched out thermostat, Topping off coolant. No difference.

Changed out sending unit. (the unit offered both online and at oreillys for my particular make/model has a spade type connection for the wire, tho my original unit has the brass knob. But the connector piece that is on the wire does seems to connect well. Not sure if that could be a problem?? But otherwise the units look identical.)

But no change in Temp Gauge.

I also placed the sending unit in a glass of hot water while connected to wire but couldn't really tell if gauge changed much from C.

So I got into the gauge cluster and am getting the same reading on each side of the wire from sending unit to gauge.

If i remove wire from sending unit, my gauge drops to C. If i ground the wire it shoots up to H again.

The speed at which the gauge reaches high varies. This morning, having not started it in 24 hours, the gauge took about 4 minutes to reach high. That may be an average of what it takes tho often its much more instant than that.

After running it to hot, then stopping it for about 15 min. The gauge read only about 1/4 way up from C before starting. (Went to H within 30 seconds)

I ran it with radiator cap off and am not seeing much if any change in coolant level as it heats up. It's swishing around inside and is fairly hot to the touch by the time the gauge is past H.

Should i turn my attention to the radiator? I figure i should flush it today and while it's empty go ahead and take the thermostat housing apart and maybe boil check the thermostat since not every part I've bought from oreilly's has been functional.

Any ideas would be really appreciated. I'm running out of them. Thanks!!

[hatzie]

Can you get access to an infrared thermometer?
Knowing what the actual engine temperature is right next to the sender will help.

[Green Machine 79]

Ok. I'll try and get an infrared in a bit.. would I just shoot it an inch or so away from sending unit?

I just changed out a thermostat after testing it on the stove. And am in the process of flushing the radiator.

When I started it up for the flushing cycle, it ran cool a few minutes longer than it did earlier today, but then the gauge practically jumped all the way past H.

There hasn't been any other signs of overheating in it's performance but I also haven't driven it more than ten miles a couple of times..

[hatzie]

The original 1513321 sender is what folks call a Nail Head sender. It's obsolete and the common replacements don't have the same thermistor resistance curve.
The gauge will not point to the same location with the replacement sender that it did with the original but it should still sorta work. This kind of thing drives the Corvette restorers nuts. Letric Limited has made an exact 1513321 replacement for restorers but it'll cost much more than a parts store unit. If you still have the original 1513321 sender I'd hang onto it and maybe sell it on evilbay or just re-install it.

60 seconds is not long enough to actually overheat but 4-5 minutes is... barely... I made the assumption that the coolant isn't boiling. That's why I suggested the infrared thermometer. Take several readings anywhere on the head block or intake coolant crossover keeping away from the exhaust ports. You could point it directly at the exposed sender threads or the threaded plug in the RH cylinder head where the sender would go if you installed that head on the LH side. Boy that last sentence has a bad flow to it...
If the coolant isn't boiling I'd look over the sender wire for damaged areas that are grounding to the engine once the wire warms up and the jacket is more pliable. A piece of expanding sleeve slid over the wire would stop one of these type of shorts.

[Dead Parrot]

38 year old gauge + old sender = inaccurate and/or inconsistent reading. Aftermarket senders rarely match what the gauge is expecting.

Assuming a motor is properly full of coolant, it takes one several minutes to warm up from a cold start. Your cold to hot in 60 seconds pretty well means something is wrong with the gauge/sender/wiring.

On a stock motor, there should be a 2nd sender port on the passenger side that you could use for an aftermarket temp gauge. Above the back two plugs like the normal one is above the front two plugs.

[Tiger Joe]

i was going to suggest why not buy a cheapo mechanical gauge and install it to verify it is actually overheating.

also I would not doubt a bad sender right out of the box. On our 79 c30 we had a printed circuit going bad (didn't know it at the time) bought a new temp sender and it worked worse than the one we took out.

[RWJJR]

Your not gonna run hot 60 seconds after startup....

[wilkin250r]

I'll echo what everybody else is saying, while pointing to yet another piece of evidence that other's haven't mentioned.

If you ran with the radiator cap off, and you're "not seeing much if any change in coolant level as it heats up. It's swishing around inside and is fairly hot to the touch by the time the gauge is past H", then your temp gauge is not accurate. I can tell you without a doubt, if MY temp gauge was past H, and my radiator cap was off, it would be a geyser of magma-temperature coolant capable of flaying skin off (think of the face-melting scene from Indiana Jones and the Raiders of the Lost Ark). It wouldn't be "fairly hot to the touch", it would be boiling itself dry as fast as it could escape.

Now, I don't think that mean you're in the clear, you may well indeed still have an overheating problem. But so far the evidence indicates that your temp gauge is reading MUCH higher than actual. You're going to need another method to monitor engine temps.

[Green Machine 79]

Thanks you guys for the all the ideas and info! I really appreciate it.

While waiting for my I.R. Thermometer to come in the mail, I dug down to inspect the back of the gauge cluster and found it to be a bit dirty and rusted but the connections looked solid and overall not worth my possibly damaging them to 'improve'.

So then i followed the wire from the sending unit back to the firewall. In doing so, I gingerly placed it away from the motor as it ran along.

After I got the I.R.T., I start up the engine and within 3 minutes the gauge is ticking up to H, but this time doing a little dance along the way. Up two steps down 1 step, Up... Until it finally broke for off the scale H.

The temp readings around the Sending unit was between 176 - 204 max. The upper radiator hose was only 66.

I think the wire is grounding out with the heat of the engine..

To save much trouble for today, Could I just cut the original wire back by the fire wall and connect a new wire to the old? Im not sure if there would be too great a loss of current. But It sure seems like the easiest approach.

Thanks again!

[Green Machine 79]

Well, I wired a new connection between the sending unit and the fire wall with the same dancing gauge until off the scale, result.. strangely now my brake light is coming on and off occasionally.

I wonder if my sending unit is for a dummy light rather than a gauge. Yet it seems to be the only option on the market for my year and model..

Otherwise, perhaps I should buy a temp gauge that I run directly to the sending unit, bypassing everything else? Or perhaps a dummy light?

[Tiger Joe]

Maybe your printed circuit is bad on the back of the cluster. I had a bunch of odd issues with our 79 and ended up putting on a new printed circuit. Working good.

If you want to try repairing your own printed circuit you can go buy a rear window defogger repair kit. Has the lil bottle of “paint” that conducts electricity.

I would still just buy a mechanical gauge to verify your not actually overheating.

[Green Machine 79]

Thanks Tiger Joe! I'll look into that.. I ordered new light sockets anyway so I'll be back into it soon.

In the meantime, I got a mech gauge from Napa today. Got it in and took a drive. Gauge read right at 185 both in town and on highway. It went up to 190 once I pulled back in to the driveway and let it idle. So I think I'm good to go for now.

I read online earlier today that the sending units with the nail heads were built for gauges while the ones with 'tab' connecters are meant for switches. That doesn't explain why my only stated options according to part lists have tab connecters. But I plan to take hatzie's suggestion and contact 'Letric Limited'.

Thanks again everyone!

[hatzie]

The idiot light sender wouldn't move the gauge needle off the lower post til you were in an overheat situation.

The IR thermometer readings confirm what I already suspected. You are not overheating... there's something wrong with the gauge sender or power circuits or the gauge itself. At 204°F the gauge should sit almost dead centre.

The TEMP sender wire in the engine compartment is green with a Packard 56 Female terminal at the sender and Packard 56 Male at the firewall plug. Pretty easy to reproduce and install with no splices. You can re-use both plastic housings.

The TEMP sender wire could be damaged inside the cab. It's worth a look.

The other thing that could cause odd behaviour, besides a bad gauge, is the cluster power and ground splices about 6" from the Instrument cluster plug. I'd lean toward damage to the sender wire somewhere because a dead short to ground will peg the TEMP gauge at HOT.

Is the E-Brake returning all the way to the top? Not quite returning will cause the E-Brake pedal switch contacts to intermittently close to ground.
Does the BRAKE lamp go out when you unplug the pin terminal on the Combination (proportioning) valve switch? This switch could be failed or the Combination valve plunger has moved because you have a leak in one of the brake lines.

Is the exposed copper that mates with the GM printed circuit terminals on the cluster plug lifting off the plastic and making contact where it shouldn't?

[Green Machine 79]

Thanks again Hatzie!

Very interesting about the sender wire. I'll have to look closer for that. Wiring issues have always made me nervous and I've been lucky for many years. Ill have to look again at the cluster to know what your saying about the "cluster power and ground splices" for sure. (I'll prob wait til I get the bulb plugs in the mail since every time I disassemble the dash something gets a little less beautiful..)

As for the brake light, it was indeed not returning up fully. I crawled under and tightened the nut on the cable and it seems to be good now.

Thanks so much for all the help!

[hatzie]

Motornut posted a picture of those splices and another pic of the printed circuit plug area with the copper peeling off the plastic.
I'll see if I can find em... Hopefully Photobucket hasn't killed em.

Found it... Post #2 is the splices. Post #4 is the peeling printed circuit.
http://67-72chevytrucks.com/vboard/s...d.php?t=564588

[hatzie]

It'd be interesting to observe the resistance of the sender as the engine warms up. Just to get a handle on what's happening with the new sender.

The sender Ohmage is measured from the wire connection terminal to the threaded body of the sensor or the head.
I experimented a bit with a couple squarebody gauge clusters on my electronics bench using calibrated resistance decade boxes and a lab DC power supply.
One of the mid 80's TEMP gauges I tested took 110Ω between the sender and supply ground to drive the needle to 210° or so and 60Ω at to peg the needle at 250°.
If the sender is reading close to 60Ω at 204°F engine temp the sender is spent.