Overflow tank on Lexus fills up on warm days

Hey there, time traveller!
This article was published 01/04/2005 (7554 days ago), so information in it may no longer be current.

QUESTION — My Lexus has this problem only on moderately warm and warmer days. The radiator cap relieves to the overflow tank to the point that it then overflows. The cooling system temperature only increases when the system is low on coolant. The system has been flushed and replaced with new coolant, new thermostat and a new rad cap (seemed to resolve this initially). Since the fall, the problem is non-existent. I also used an infrared pyrometer and no unusually hot areas of the engine were found. Your help would be appreciated.

ANSWER — I can think of a few possible causes of your problem. Let’s start with the obvious. Perhaps the overflow tank has been overfilled. Normal expansion causes the hot coolant to be forced into the overflow tank and if it is too full it overflows. I doubt this is the problem however because I assume this has been checked. Therefore, I must assume that the engine is overheating.

Check the operation of the electric cooling fan. If the temperature sensor is out of range, the cooling fan may not be turning on until the coolant is too hot and already boiling. Most cooling fans turn on when the temperature reaches about 105 C and run for a minute or two until the temperature drops a few degrees. Also check for missing foam blockers that direct the air flow into the radiator instead of around it.

A very common problem is dirt and leaves blocking the airflow between the air conditioning condenser and the radiator. Both the condenser and radiator may look clean, but remove the top mount plate and look between them. I see some that are blocked almost solid!

If the cooling system has been refilled, you may have had an air bubble trapped in the system. This causes overheating and the coolant to boil out. It is difficult to release the air from many modern cooling systems because the filler is lower than the engine. There are often bleed ports on the engine that can be opened to release the air or a hose may need to be partially removed. Some shops are placing a vacuum on the cooling system to suck the air out before they fill the system, just like they do on the assembly line.

Finally, you may have a crack in the engine or a head gasket leak that allows combustion pressure into the coolant passages. There are dyes that can be used to test for carbon monoxide in the coolant (which is an indication of a combustion gases leak) without dismantling the engine. Because the system works well in cooler weather, I am guessing this is not the problem, but keep it in mind as a last resort. I would start by checking between the radiator and condenser for a blockage.

QUESTION — I have a 1995 Lumina Van with a V6, 3.8 L engine and 144,000 km. The problem is a noticeable engine shudder at 80 to 85 or 90 to 105 km/h when moderately accelerated. It was noticeable also at lower speed, however not as pronounced. Once this speed range is passed the powertrain runs smooth. I also noticed that the engine is sluggish to accelerate if the shudder is occurring. In addition, above the shudder speed range the engine runs smooth; however, it seems to accelerate and slow down without changing the gas pedal position. The shudder happens if it is wet or dry weather, 0 C or + 25 C outside temperature. Below -10 C it does not happen.

I changed the spark plugs and the spark plug cables without any improvement. Vacuum hoses seem to be intact without visible cracks. I would appreciate your opinion on this problem.

ANSWER — The problem is likely a lean fuel misfire shudder that typically occurs on a warm engine when the torque converter locks up and the engine load increases. Usually this is caused by a bad spark plug or plug wire. One may have been damaged, but look elsewhere first.

Check vacuum hoses again, especially the one going to the PCV valve for cracks or leaks. Make sure the PCV valve is the correct part number. There are many that look the same but have different airflow rates.

The problem could also be caused by a worn throttle position sensor. When operating in the worn spot, the fuel may be cut off or leaned out. A technician can check this out with a lab scope or graphing computer analyzer. A voltameter can be used but it is difficult to see the problem using only voltage readings.

Finally, the injectors may be dirty and restricting fuel flow at light throttle. Have them cleaned. If everything checks out fine, go back and check those plug wires again.

Jim Kerr is an experienced mechanic, instructor of automotive technology and freelance journalist. You can e-mail questions to Jim at the address below.

kerr.jim@sasktel.net