Bleeding help. Non standard engine change

Bleeding help. Non standard engine change

Author
Discussion

tommobot

Original Poster:

648 posts

207 months

Saturday 10th October 2020
quotequote all
I've recently installed a BMW M52 (6cylinder petrol) into my defender.

As far as possible ive tried to match the OEM (south Africa spec) cooling setup where this engine is standard.

For the life of me, I can't get an air lock out of the system.

Bleed points, top of rad, coming out of engine before top hose, just before back into engine and before heater box up top.

Any tips, ideas, thoughts welcome..

tommobot

Original Poster:

648 posts

207 months

Saturday 10th October 2020
quotequote all
More photos







GreenV8S

30,192 posts

284 months

Saturday 10th October 2020
quotequote all
Impossible to guess what the circuit looks like. Can you draw it out?

Any idea where the air lock is? What symptoms are you getting?

Boosted LS1

21,185 posts

260 months

Saturday 10th October 2020
quotequote all
TR7 header tank. That's the good bit ;-). It's brass so shall polish up nicely.

tommobot

Original Poster:

648 posts

207 months

Saturday 10th October 2020
quotequote all
Temp gauge says overheating..

Top hose super hot..
Bottom hose ambient temperature.

Brand new rad, new water pump, new heater matrix and thermostat.

aka_kerrly

12,418 posts

210 months

Saturday 10th October 2020
quotequote all
My old 5 Series BMW was bugger to bleed after changing half the coolant system.

I found if you jack the front of the car up so that the radiator bleed point is the highest part of the coolant system. Take the bleed nipple out and put a funnel in to allow the coolant level to rise/fall a bit without overflowing making a mess and run the engine with the heaters on full heat for a good 10min keeping an eye on the engine coolant temperature, the fans and the coolant level at the bleed nipple.

tommobot

Original Poster:

648 posts

207 months

Saturday 10th October 2020
quotequote all


Pretty basic drawing, to accord with the photos.. the cooling system dips into the throttle body for an unknown reason (OEM)..

GreenV8S

30,192 posts

284 months

Saturday 10th October 2020
quotequote all
Sounds like you have low / no flow through the rad.

How is your pump vented?

Your drawing only shows one line to the header tank. The picture shows a small line from the top of the rad to the top of the header tank Is there a larger line out the bottom of the header tank back to the pump inlet?

normalbloke

7,450 posts

219 months

Saturday 10th October 2020
quotequote all
I had a Caterham with a Cosworth 2.0 YB turbo in it. It would only ever bleed properly and burp the final air if I used a coolant system pressure tester on the header tank and added a few psi while it was running.

tommobot

Original Poster:

648 posts

207 months

Saturday 10th October 2020
quotequote all
GreenV8S said:
Sounds like you have low / no flow through the rad.

How is your pump vented?
Oops, missed that off the sketch.. there a small nipple from the rad, feeding back into the header...

GreenV8S

30,192 posts

284 months

Saturday 10th October 2020
quotequote all
tommobot said:
GreenV8S said:
Sounds like you have low / no flow through the rad.

How is your pump vented?
Oops, missed that off the sketch.. there a small nipple from the rad, feeding back into the header...
That explains how the radiator is vented. If this is working, you should see water flowing through the header tank when you leave the cap off.

How is the pump vented? There needs to be a route out the top of the pump assembly to let the air out. Often, the bypass hose provides this. In that case you need a path for the air to get vented from the top of the bypass so that it doesn't get pulled straight back into the pump when you start the engine. Typically that would be into the stat housing and through the stat to the top hose via a jiggle pin. It's hard to visualise this from a drawing and photo, but with the car in front of you it should be possible to see the route the air will take to get out of the system. It needs to be going uphill or with the flow of coolant at all times.

tommobot

Original Poster:

648 posts

207 months

Saturday 10th October 2020
quotequote all
Oh god.. that's a bit confusing..

The water pump arrangement is as per OEM... This engine was in another defender before working perfectly, and I've almost had it working properly before.. but think that was pure luck..

I was thinking of getting a metal thermostat housing and putting 2 bosses in it to act as bleeders....

E-bmw

9,214 posts

152 months

Sunday 11th October 2020
quotequote all
aka_kerrly said:
My old 5 Series BMW was bugger to bleed after changing half the coolant system.

I found if you jack the front of the car up so that the radiator bleed point is the highest part of the coolant system. Take the bleed nipple out and put a funnel in to allow the coolant level to rise/fall a bit without overflowing making a mess and run the engine with the heaters on full heat for a good 10min keeping an eye on the engine coolant temperature, the fans and the coolant level at the bleed nipple.
Exactly this assuming you don't have an actual problem.

It was the only way I could completely bleed the air out of my e39 with the same engine.

Obviously it was all done with the expansion cap off & repeated top-ups both at the tank & the bleed point.

tommobot

Original Poster:

648 posts

207 months

Sunday 11th October 2020
quotequote all
My setup is a bit different to the 5 series, but I assume jack it up.. leave the rad bleed point open and header tank open aswell?

Evoluzione

10,345 posts

243 months

Sunday 11th October 2020
quotequote all
normalbloke said:
I had a Caterham with a Cosworth 2.0 YB turbo in it. It would only ever bleed properly and burp the final air if I used a coolant system pressure tester on the header tank and added a few psi while it was running.
This is the way to do it. Pressurise it or apply vacuum, kits are cheap on Ebay.

smn159

12,646 posts

217 months

Sunday 11th October 2020
quotequote all
Have you checked the actual temperature with an independent temp gauge? I've chased 'overheating' problems before which have turned out to be problems with the gauge itself.

tommobot

Original Poster:

648 posts

207 months

Sunday 11th October 2020
quotequote all
Haven't checked it against something, but pretty confident if not utterly confident it's an airlock..

Tried again today... No temperature raising in top hose, throttle body hoses v.hot... removed top hose from housing slightly and water started flowing into rad..

Still nothing at bottom hose. Stone cold.
Took out temp sensor/bleed valve on the bottom hose. Could physically see the stationary cold water.

Also no heat to heater hoses arghhh...

Pressure kits, are we thinking this sort of thing?

Think I'll drain tomorrow and refil via heater hoses....

Was trying today with front end on Ramos, header tank closed and a funnel in the rad bleeder full of water etc...

GreenV8S

30,192 posts

284 months

Sunday 11th October 2020
quotequote all
Is there pressure in the top hose or just stationary hot water? If there's no flow or almost no flow, that suggests there is no coolant circulating or the stat isn't opening. An air lock in either component would account for that.

You really need to look at the layout of the water pump and stat housing and understand how they are connected. It should be pretty obvious once you see how the bottom hose connects to the pump inlet. The first thing to determine is whether it has a top stat (connected to the top hose) or a bottom stat (connected to the bottom hose).

tommobot

Original Poster:

648 posts

207 months

Monday 12th October 2020
quotequote all
GreenV8S said:
Is there pressure in the top hose or just stationary hot water? If there's no flow or almost no flow, that suggests there is no coolant circulating or the stat isn't opening. An air lock in either component would account for that.

You really need to look at the layout of the water pump and stat housing and understand how they are connected. It should be pretty obvious once you see how the bottom hose connects to the pump inlet. The first thing to determine is whether it has a top stat (connected to the top hose) or a bottom stat (connected to the bottom hose).
The thermostat is connected to the bottom hose... as shown below.

How can I actually tell as such if there is pressure in the top hose? - Resistance to squishing?

At first there was no flow whatsoever, whilst running I slightly slipped off the top hose from the housing, a load of air came out and you could feel the water rush into the hose, and the hose instantly got hot, as did the rad and the water could be seen pumping through the rad bleed nipple.

I couldn't make it come back up the bottom hose, as this remained stationary.

It seems like its almost worth taking the stat out to test and see if it circulates!?


Generic M52 image:

tommobot

Original Poster:

648 posts

207 months

Monday 12th October 2020
quotequote all
Oops, just realised I forgot to put the link into the pressure bleeder thing...

https://www.ebay.co.uk/itm/General-Radiator-Pressu...

This sort of thing?