Try cracking just the highest fitting. And similar to manually bleeding the brakes, can you close the fitting off before you stop blowing? Do this several times until coolant comes out. If you get this far and are satisfied there aren't any air bubbles, mark the fluid/float level in the expansion tank. Leave the cap off, cover with paper towel to keep bugs out and come back 1/2hr later. Did fluid level increase? If not, try running the aux pump bleed procedure. (Edit, does e46 have aux pump?)

I'm just brainstorming here. Maybe its not an issue with bleeding the system. Granted I'm boosted but not turbo so no oil/coolant lines here. Is this how all the turbo guys plumb? I originally thought you had your lines backwards but there is no set flow direction on this turbo, right?

My background is testing jet engines. I think you could have a pressure differential issue going on here. Its possible the coolant is boiling because it's not flowing, either while driving or after shutdown. This will create air in the system. It may flow fine while the engine is running but stagnate when you shut down. (It should flow, read below.)

IMO quickest way to diagnose would be to test the pressure at the lower cooling hose and the block drain tap, ideally having two gauges or transducers to measure the pressures during startup/normal/spirited driving. If there's a pressure differential you have flow. I understand this is not a trivial task.

Almost illegible..

You could test the hot shutdown flow by using a mechanics stethoscope on the lines, listen for flow, or boiling...

I think you may end up with an aux water pump on that line, with a timed shutdown procedure haha

 

With your current setup, I believe, when thermostat is shut, engine coolant is circulating internally to the engine, drawing coolant from the lower radiator hose, through the turbo, and into the block, or in reverse order, it is midnight here and the cooling system has been doing my head in over the last two days lol. This opening of the engines internal coolant circuit, regardles of direction, will also circulate coolant at a rate through the radiator. ***will update in 8 hours to be factual after o sleep

When the thermostat opens, coolant should flow through from the engine block, through turbo, and into the lower hose.


The heater lines form the second coolant circuit, both heater lines originate from the pump inlet and outlet. I wonder if tapping the heater line before the heater switch (lower port), for return, and tapping the lower heater hose at the expansion tank (feed), would solve your issue? If not completely, possibly a non return on both lines would hold the system bleed in the turbo while the engine is stopped. Heater circuit utilises the expansion tank for some return flow while the heater switch is closed, so it may be something to investigate????

 

After some more thought, returning to your top radiator hose may also correct flow, in place of the bottom radiator hose. Tapping as close to the engine block as possible.

 

Thanks for all the brainstorming! I have read your remarks over and over trying to grasp what I'm really working with.

Dave, you mention it might be flowing from the block to the turbo, and back in the rad hose when the thermostat is open. I believe I have flow, perhaps when I shut the car off this could be improved by reversing my lines? I assumed the bottom rad hose was a positive pressure, because its hard when I grabbed it, but I can now see that it could be negative. Is the block drain positive? I guess the radiator is the "transition" point of pressure? I also see you run a coolant line off your chrome thermostat? How does that work and whats your setup like(bottom mount but still interested).

What really gets me is that the turbo is higher than the radiator cap, so if any air gets in the system, i'm not sure it gets coolant. Alpine, yes, I blow and then close the top nut after air stops coming out, just like a brake system. And yes, after shutdown I planned to get some flow, that's why I clocked my turbo. When I shut the engine off I get a "gurgle", I can hear the water/air flowing around. I will be away for the next week, but I will definitely look into testing the flow as you mentioned. Thanks

Every turbo car I have worked on has a reservoir, aka swirl pot, above the turbo, take Subaru for example. They have a normal rad cap, but once the system is bled you only add fluid to the top reservoir(turbo mounted on the top/back of engine). So, in my case, what stops air from pocketing in the turbo while running? The only thing that gives me hope is that we have a "pressurized" system.

thanks for all your thoughts, I really appreciate it.

Makes me want an air cooled turbo lol

 

I think the swirlpot is a fix for the issue, but not the root cause. When you shut off the engine the coolant in the turbo rapidly expands as it heats, it's density decreases so it will naturally want to go upwards, just like a hot air balloon. This is the driving force behind the flow. But it's already at the top of the loop, "nowhere to run, baby!"
Might as well take the coolant lines off if you are going to run a swirlpot

 

So just run a swirl pot not connected to the engine coolant system at all is what your suggesting?

I'm not sure how to integrate one into our cooling system to be honest. I have a CSF radiator but still the factory expansion tank and such. Maybe off the heater lines as Dave noted?

 

Sorry, I was being sarcastic. I think you need an inline pump, that also circulates after shutdown. I say this because the swirlpot will "filter" out the bubbles while you drive, but you will still get bubbles (boiling) everytime you shutdown. And one of the biggest reasons for a coolant cooled turbo is to help cool the bearings after you turn the car off.

Another test you could easily do is two different shutdowns. First, drive around in boost and then with the engine and turbo hot shutdown (then try bleeding once it's back to ambient temps). The second would be drive easy not in boost, let the turbo cool down for a couple minutes idling then shutdown. Again when its back to ambient bleed the system from the highpoint and see if there's any air.

*Also, listen on the hose near the turbo using a mechanics stethoscope or screwdriver. You will hear a difference between the two I'm fairly confident. The gurgle you explained is the coolant boiling.

 

Lol ok gotcha, I had no idea. But yeah I have driven and then bled the system before, it always has air in it when I crack the fittings, which is actually why I made this thread. The sound being water boiling does make sense as well.

Either way, would I be able to integrate a swirl pot using the setup I have now(rerout some hoses, run it between the block drain/lower rad hose and the turbo)? Honestly not sure how to route it, and with the water not cooling the turbo on shutdown, I really need to fix this.


Sent from my iPhone using E46Fanatics mobile app

 

I'd like to add something small like this, not sure how to plumb it though. Mounting it somewhere high in the engine bay wouldn't be hard though.

Here's one on a E46Fanatics drift car, not sure how it's plumbed though


Sent from my iPhone using E46Fanatics mobile app

 

The Instagram user above just messaged me back about his pot. He is running a m52 iron block, bottom mount turbo, so I don't think I have these ports, but this is how he runs his:

"Mine is plumbed like this, hose coming from the top of the head to the top of the pot, hose coming from the coolant pipe on the side of the motor (comes out of the back of the waterpump) going to the bottom of the pot"

I wonder how this could be done on the M54


Sent from my iPhone using E46Fanatics mobile app

 

I have two more drawings to mark up for coolant flow, but am having trouble showing flow on the pictures I have. Hope these ones help somewhat.

 

Ok great that really help.

So a swirl pot off the block and heater lines and head coolant temp sensor hole could work well, and is what the drift guy suggested.

However, he does his to get air out of the coolant in the head to keep hot spots to a minimum. He runs a bottom mount turbo, so doesn't have my issue.

To fix my issue, won't I have to run it in line with the turbo (sitting above the turbo however) so air can escape? Perhaps from the block drain, to the pot, to the turbo, and back to the rad hose? Or on the red hose? That way it still flows through the turbo but air is removed in the pot, instead of trapping in the turbo.


Sent from my iPhone using E46Fanatics mobile app

 

since you have a "T" fitting coming off of your radiator hose already then you can do something like this with a Bosch VW or MB or BMW auxiliary water pump.

 

Ok, that looks cool. Please excuse my ignorance, but what is that box with the numbers in it?


Sent from my iPhone using E46Fanatics mobile app

 

oh its the relay to run the power through rather than direct.

 

Should I maybe use a pump and a swirl pot?

 

I consider my bypass to the thermostat on the lower hose side to be a mistake which I am currently rectifying. I am not saying to use a pump or swirl pot or vice versa, I am limiting my comments because I am not decided with my own system. But, it would be in my opinion be a mistake to run a pump inline with your current connections, block to lower radiator hose. You will ultimately be bypassing the thermostat, and sending cooled water directly into the block at a high rate. This will also cause your radiator to have constant flow, which is usually controlled by the thermostat.

 

The line going into the block can simply be rerouted to the upper radiator hose. I usually try to keep things simple, but I think a pump is necessary in this case.

Is this a serious thread? I can't believe you have a boosted car and don't know what relay wiring looks like.

 

Yeah I built my car with the help of the internet in my garage. And yes, I didn't know what it was, partly because I didn't think it would be necessary in this application, and I am not familiar with many wiring diagrams since I hate wiring things. Eventually I'll get around to learning more about the finer points.

In other news, the whole swirl pot idea seems to be to get air pockets out of the back of the head on most cars.

It seems if I took coolant from the coolant temp sensor ("t" this junction) from the back of the head, and plumb it to a swirl pot mounted high, I could kill two birds with one stone. I could limit possible air pockets, and have a good coolant feed to the turbo. I just need to figure out if it would flow back better into the block or the lower rad hose naturally. I would also need to figure which side of the plumbing I should run the coolant through the turbo.

I agree that a pump would be nice, but if I can get the coolant moving the right way it shouldn't be needed, plus wires

Thanks for all the help

 

Some of my flow paths that I marked are proving not correct. Head temp gauge will be a good take off, and top radiator hose or block should work. Though top hose would be my pick.

 

Ok Great. I already have the block barbed so I may just go that route first as a proof-of-concept.

I have to see if I can get a T in there by the head with the M54 manifold when I return home next week. Anybody know if anybody sells adapters to t it? Likely usually used for a second temp gauge.

Gotta find out what thread it is... https://www.summitracing.com/parts/vdo-240850/overview/

http://www.mscn.com.au/oil-pressure-tees

 

One more thing, should I use an overflow tank, kind of how the expansion tank works? Not sure if the expansion tank will work now that this will be the highest part in the system.