MGB V8 EFI Cooling Problem

[minorv8]

A question about your top hose: is the hose higher than your rad cap ? It´s hard to see in the photo. Mine did and it never bled the air in the top hose and consequently tended to run hot. A simple hose lead to the expansion tank cured that. The attached pic is a later version with the EFi manifold, with carb the top hose ran even more upwards. With this setup I thought that the bleed hose was no longer necessary. First test drive proved me wrong

Those with keen eyes will possibly note that this is not an MGB but even other makes have same issues. My rad is underneath the water pump pulley, I ran into space issues when I did this conversion 25 years ago.

[madmal007]

Yes, it's slightly higher as the rad cap would have originally fouled the distributor cap. Of course now I've converted to MS, the distributor cap is no longer required so I will move it to its highest point.

I would have preferred to have the expansion tank higher but it's at it maximum height now.

Many thanks for the observation.

Mal

[SuperV8]

Yes, it's slightly higher as the rad cap would have originally fouled the distributor cap. Of course now I've converted to MS, the distributor cap is no longer required so I will move it to its highest point.

I would have preferred to have the expansion tank higher but it's at it maximum height now.

Many thanks for the observation.

Mal

The way you have your cooling system setup means the height of your overflow/catch tank isn't important - could be anywhere in the engine bay. The Important part for your system is your radiator cap in the top hose which feeds the bottom of this overflow tank. This needs to be the highest point in the main pressurised part of your cooling system so any air in the system will collect there and get released into the overflow tank.

If you had an expansion/header tank system then yes it is very important this is the highest point in the cooling system.

You mentioned previously that you are using the standard MGB rad cap - but I don't think this is the correct one for the way you have the system plumbed. The rad cap in your top hose needs to be one with a double (rubber) seal in it, the standard MGB rad caps look like they have a single rubber seal designed for an expansion/header tank type system - NOT and overflow/catch tank system.
https://www.moss-europe.co.uk/cap-radia ... soc=108626


Additionally regarding the pipe you have looped back from the manifold into the coolant pump inlet pipe - this can be blocked off. With this plumbed in you have 2x thermostat bypasses (3 if you have the heater circuit open) which are coolant loops which miss the radiator. You only need 1 bypass circuit so that when the thermostat is closed you still have a small coolant flow.

Also have you checked your fan belt is tight?

[madmal007]

That's interesting, thanks. I'll get a new rad cap. Its the small format rad cap and its proving difficult to identify from parts drawings as they don't always specify size. I sourced the last one by taking it to a car parts supplier and getting them to find one.

I'll also look at blanking off the small pipe.

The fan belt is tight so I'm sure that's not the problem.

Cheers

Mal

[Ian Anderson]

Hi
Have you got the right tune resistor in?

The resistances change and it puts the ecu for Hotwire into another tune, for cats, no cat, different fuel etc.

Could the engine be running either rich or lean as both can cause additional heat to be generated.

You may need to see what emissions are coming out the exhaust to point you in the right direction

Ian

[madmal007]

The Dyno session included fine tuning the map with the lambda sensor so i think the emissions are fine.

I am checking the car temp gauge as I think it may be reading high and my lap top battery is shot so I can't get real time temp data from the MS or start any logs.

Hopefully with all the help and advice from the forum, I may have made some progress...

Mal

[DaveEFI]

I'm fairly certain the temp reading I get with MS is accurate. I entered the actual resistance of the sensor at O, 20 and 100C. But MS reports about 10c lower than the dash gauge. Of course the sensors are in different places.

[madmal007]

That's interesting, I've just fitted a new gauge, which I know is compatible with the sender unit, and that's reading about 10 degrees higher that the MS as well.

I've rewired the fans to come on as a pair (rather than MS for one and a manual switch for the other) and with the new alloy radiator, it sits happily at tick over with no overheating and the fans kicking in and out as needed, so all good.

I guess it was a combination of a bad radiator, incorrect temp gauge reading and only one cooling fan coming on plus various air locks and hose configurations etc.

Just waiting for sun to come back for an extended test drive.....

[Ian Anderson]

Are you running the gauges through a regulated voltage supply 10v or just from the battery/alternator at anything up to 14.1v?

Depending on what the gauge is expecting you could get those differences.

Ian

[madmal007]

I've had some experiences in the past, on a TR6, with voltage regulators so that was one of the things I checked first!

Mal

[DaveEFI]

On my car, the same voltage regulator feeds both the temp and fuel gauges. Both are hot wire meters. The fuel gauge is pretty inaccurate. It shows full long before the tank is full. And empty with fuel left. So do wonder just how accurate the temp gauge is.

[madmal007]

That's a good point. My fuel gauge reads quarter full on empty but I know the tank was modified to accept the Jag LSD and suspension so I know the sender unit needs looking at.

I went for a Smiths Flight water temperature gauge with "microprocessor controlled stepper motor needle movement" and it still reads 10 degrees higher than MS!

[DaveEFI]

Using an infra red thermometer, I did see some difference between the sensor body on the EFI sensor and the gauge one. Although not as great as the readings

[stevieturbo]

That's a good point. My fuel gauge reads quarter full on empty but I know the tank was modified to accept the Jag LSD and suspension so I know the sender unit needs looking at.

I went for a Smiths Flight water temperature gauge with "microprocessor controlled stepper motor needle movement" and it still reads 10 degrees higher than MS!

https://spiyda.com/fuel-gauge-wizard-mk3.html

[madmal007]

Thanks for that, looks exactly what I need.