Overheating or bad gauge?

[Voyager]

Working off bugs on my new car.

My Smiths coolant temp fluctuates all over the place. It normally sits at 90, then spikes erratically to 110/120/130 in a matter of 2-3 seconds, then after a while, it comes back down just as quickly to around 90. I’m driving very conservatively while getting everything sorted out. Gauge problem?

The Smiths oil temp reads consistent 80.

Also - I have a “shiny” surge tank/expansion tank mounted to the front of the engine, but I don’t see an Over Flow Tank (black canister) anywhere. Problem?

Bob

[tcrist]

Sounds like you might have an air bubble still somewhere? Have you checked all of your grounds? Fiberglass cars need good grounds.

[Calbullet]

Smith's gauges are mechanical, they do not need grounds to work. erratic temps usually mean air is still in the system.

[Outwest34au]

Smiths gauges also come in electric form as well as the capillary type with bourdon tube, as does VDO and everybody elses. The sender will probably be 0-90 ohms or another common spec. To test the gauge earth out the wire to the sender on the block, the gauge should rise and stay steady, also check the sener with a multimeter as it warms up.
Firstly though, double check all earths as mentioned above.

[timsullivan]

Bob: My guess would be a bad sending unit or gauge as the sending units are very sensitive and can be damaged quite easily. I would make sure the cooling system level in the chrome expansion tank mounted in front of your engine is at the mid point of the tank with the engine cool. If the coolant level is low or not showing in the expansion tank the that clearly could cause a bouncing temp reading. I would also check the connection for the water temp sending unit which on your 408 is most likely directly behind the distributor, on the intake manifold. If the line leading to the sending unit was properly installed it should have a large loop in it such that the line extends straight upward from the sending unit at least 3 or 4 inches. If it doesn't have the loop and was installed with a sharper bend at the water temp sending unit the line may be damaged.

I wouldn't worry about the lack of a "puke tank" to catch cooling system overflow. If the coolant level is maintained at the mid point in the expansion tank, when cool, your system will rarely throw out excess coolant when the car is warmed up.

-Tim

[Blas]

The water level in your chrome expansion tank should be up to the divider inside the tank. Remove the cap (when cold) and stick your finger inside the tank, you will feel the hole in the baffle plate. Any extra coolant will be sent to your overflow tank (assuming you have one)
Smith’s gauges work or they don't. Accuracy is another story. The standard SPF temp gauges use capillary tubes.
A wildly reading gauge says you have just air in the intake manifold. (You hit steam then coolant hence the wild readings) You can try purging the system by opening the expansion tank (NOT when engine is hot) Start the engine, add fluid to expansion tank. The water temp sensor is usually in the front of the engine, look for a small brass tube wound with a steel wire. Loosen the brass fitting on the end of this brass tube and watch for air escaping along with a little coolant. When the air is gone you'll only get coolant. Tighten it back up. Having the front of the car higher than the back helps too when doing this.
Did you have the small hole(s) on your thermostat disk orientated up? If there are none on the thermostat disk, you can drill one. Careful with the thermostat orientation engine vs. radiator side when reinstalling the thermostat.

[Excaliber]

It's "steam pockets", of this I have little doubt. The system does have air in it, the water flow is not constant. The low water causes steam pockets to form. Those steam pockets are HOTTER than the surrounding water, thus spiking the temp gauge. When normal water flow returns, the steam pocket get's flushed, the gauge reading returns to normal.

The big clue here are the observed temps, going from 90 to 120-130 and back in a "spike like" fashion. Until you can PROVE the gauge is bad you better believe it!!! Ignore it's readings at your own peril, you wouldn't be the first to seriously damage his engine because he chose to dis-believe a gauge.

I've seen these spikes before, in my experience it's almost always related to a low water condition. Those pockets also greatly increase, at least momentairily, the system pressure. Pressure goes up, water is released to the overflow tank (if you have one, or to the ground if you don't) and given enough time it becomes a "catch 22". Steam, loss of water, more steam because you lost water, etc.

This is one of the primary reasons I'll be using Evan's Waterless Coolant, the very high boiling point at ZERO pressure, will not over pressurize the system and push coolant out. Steam pockets will not occur. If you have a water loss problem Evans over water may just buy you enough "time" to get home, off the highway or even save your engine from damage which may occur due to extreme overheating.

If you have something like a small head gasket leak thats one way you could be loosing water and not see or sense it's loss. Sometimes a head gasket leak does little more than over pressurize the cooling system, causing it to expel water. The motor may run great, with no signs of water intrusion.

[madmaxx]

Easy to test gauge. Remove sending unit carefully. Boil pan of water. Place sending unit in pan, compare pan temp thermometer with gauge.

I found mine to be withing .5 degrees accurate. That is the best way to fill the cooling system with the temp sending unit slightly loose to bleed air, then tighten.

[Voyager]

Guys,

Great info. Tell me if I’m understanding the fix to this steam pocket problem.

1. Ensure distilled water/coolant level in the chrome expansion tank is up to the divider inside the tank (cold engine).

2. Try purging the system by opening the expansion tank (NOT when engine is hot) Start the engine, add fluid to expansion tank. Loosen the brass fitting on the end of this brass tube and watch for air escaping along with a little coolant. When the air is gone you'll only get coolant. Tighten it back up. Having the front of the car higher than the back helps too when doing this.

3. Check to see the sending unit line has a large loop in it such that the line extends straight upward from the sending unit at least 3 or 4 inches.

Thanks

Bob

[Blas]

Sounds right...Keep us posted...

[Excaliber]

My radiator has a bleed valve at the top. BUT, the top of the radiator is lower than the top of the engine. Air can get trapped in the radiator and never move out to the higher part of the engine, so in my case I cannot bleed all the air out from anywhere on the engine.

My method: Park the car with the nose uphill at a good angle. Take off the radiator cap and start the motor. Monitor the engine temp by both the gauge and by sticking my finger in the surge tank. When it starts to get good and hot I know the t-stat has opened and I'm getting good flow from engine to radiator. Put the cap back on, open the bleed valve at the top of my radiator. I almost always see some bubbling as air escapes there. I fill my over flow tank about a third, go for a drive, the overflow tank fills up more with water being pushed out of the "over full" engine. As it cools down the overflow tank water is sucked back into the engine. After a couple of cycles I find the surge tank is now at the perfect level, about 2/3rd's full. The overflow tank is also at just about the right amount. Now I know how much to fill the overflow tank to as well as the surge tank on the motor when everything is cold.

Assuming you have an overflow tank and you find your having to add water to it pretty often (TO often) my guess is one of two things:
1. Water is evaporating from the tank (car sits a long time).
2. Water is being pushed out of the motor into the tank and the tank is overflowing. Eventually the motor sucks ALL the water out of the tank and back into the motor. At some point, it won't have enough water to suck back in and the system is now "low". You then add water to the surge tank as well as the overflow tank.

The question then becomes: Wheres the water going???

[Voyager]

I bled the temp gauge bung and went for a test drive. Steady 80-90 the entire time. You guys are magic. Next round is on me.

My punch list is down by one.

Thanks,
Bob

[Voyager]

I take it back. Went for a second ride today and the gauge went back to fluctuating back and forth for a while, and then it died. I’ll call the dealer tomorrow.

[Blas]

When you moved the brass tubing, you did so gently correct?
When you finished retightning the sensor end, and went for a ride, was the brass tube and its steel protective coil touching a spark plug wire or positioned so that a spark could jump to it from a wire on the cap?
That would kill it quickly...
You loose the gas inside the tube...a microscopic hole in the tube from the spark(s)
DId you get air air out of the connection?

[Voyager]

With 100 miles on the odometer, I sent the Superformance/Keith Craft engine car back to the dealer. The temp gauge now works, but the coolant temp fluctuation remains. Although not as frequently, the temp still spikes erratically to 110/120 in a matter of 2-3 seconds, then after a couple seconds, it comes back down just as quickly to around 90. Noticed it would drop sooner if I turn the heater on. Kept the fans on the entire time.

After a 15 min drive today, noticed the top of the radiator was extremely hot and the bottom was only luke warm. Air pockets or pump issue? Thoughts?

Sending it back to the dealer tomorrow…

[rustyBob]

The top will be hotter than the bottom .....hot coolent enters through the upper hose then flowes through the radiator to the bottom hose where it reenters the engine....your temp changes that show on the guage may just be the coolant "cycling" through the system....tstat opens hot coolent flows into rad. and when the "cooler" coolent from the rad. hits the tstat it closes and waits for the coolent to heat up again so it can "cycle' again....

[Blas]

Something that comes to mind, but probably not related, is the posibility of the impeller in the water pump being loose or with the veins of the impeller for a reverse rotation flow belt layout. But I assume your engine was fully dressed when purchased...Have you considered removing the thermostat to see if the problem goes away?

[Voyager]

Blas,

Thanks for the feedback. I’m still having the surging temp gauge issues.

Can’t seem to get that last phantom air bubble out (if it exists).

Can the capillary tube (wire) between the Smith gauge and sending unit cause the needle to surge if it is misrouted (too close to heat / plug wire / sharp bend)? Or does that just make it fail completely?

[Tommy]

Voyager,
You seem unwilling to give up on the idea that the gauge may be the problem. To safely resolve that issue I suggest one of two approaches. For about $20 you can buy an aftermarket temp gauge at any auto parts store and temporarily install it while leaving your stock gauge in place. If it does the same thing as the original gauge, then the gauge is not the problem. Another option, if the temp spikes occur while the engine is idling, is to take it to a repair shop or someone with an infrared temp sensor and have them see if their sensor spikes at the same time your gauge does. If you want an infrared sensor of your own, there is a Harbor Freight store that sells them not far from you.

It seems to me you need to resolve your doubts about the gauge so you can move on to solving the overheating problem if there is one.

[Voyager]

Tommy,

Great points. Appreciate your candor.

The 3-week old car is currently back at the dealer.

They said they tried 3 different Smith gauges - all spiked while test driven (doesn’t spike at idle).

They have also used their infrared temp sensor at idle. All readings normal.

I look to the dealer to resolve the issue at this time.