Vuti

Has anyone ever wondered why FM, Begi and even miataturbo FAQ recommends to run coolant lines for your turbo all wrong?



The main idea of a water cooled turbo is to be able to cool the turbo after shutdown with thermal siphoning. For this to work the water and heat needs to rise up. This is not the case with most of the setups, for example taking feed from the front thermostat housing and returning it to the mixing manifold. Water boiling at the turbo, oil clocking, bearings destroyed?



The right way by Garrett:
1. Rotate (clock) the center housing to 20° from horizontal, either direction, once turbo is installed
2. The input water (colder side, from the engine’s cooling system) should be plumbed in to the lower of the two ports after the housing is rotated.
3. The hotter output water leading back into the engine’s cooling system should be plumbed into the higher port and allowed to travel “uphill” all the way back to where it meets the cooling system. No up/down kinks or “traps” should be present in this return line.

You can read all about it from the Garrett white paper: http://www.turbobygarrett.com/turbob...er_Cooling.pdf

I'm done with my FMII original setup, you can hear water boiling in the turbo after shutdown and no wonder since the lines go all f*cked up.

I'm making new lines, the feed will be taken from the block next to the oil feed. This ensures nice uphill to the turbo for the feed.
But the coolant return is a bit difficult, the only solution I have come across is AbeFM reroute (pic below) where coolant return goes to back of the head after the thermostat.


(AbeFM reroute, notice the turbo coolant return to back of the head (reroute))

I hope this opens a nice flame free discussion on the topic?

Braineack

iTrader: (62)

this thread is stupid.

this is wrong. why was this thread created when the entire basis was wrong?

ScottyP3821

iTrader: (2)

I always figured after shutdown the coolant is going to want to move away from the hottest part of the car. So after shutdown the hottest part of the car will have the hottest water and will want to move to a less hot region thus causing water pressure and causing the water to move through out the cooling system. I might be wrong thats just how I picture it in my head.

sixshooter

iTrader: (12)

I understand Garrett's idea but when OEM auto manufacturers don't follow these guidelines and their turbos routinely last in excess of 100k mi in the car of the average dimwit, I don't find myself all lathered up about making changes to mine. Many OEMs have a 100k mi warranty and don't implement this idea.

I allow adequate cool down time, change my oil more regularly than normal, and use better oil than OEM requires. I also don't expect I will need my turbo to last over 100k mi, as I will likely upgrade to newer technology at some point. A rebuild kit for my turbo is ~$42.

18psi

iTrader: (76)

1) OEM's run it and have run it this way for ages with no problems
2) Your "fix" is not in any way better than the alledged "flawed" routing. Pretty much same
3) If you ACTUALLY want to do it properly, you will run a similar setup as Subaru and put a little tank a bit higher than the turbo so that water from it can actually pour through the turbo after shutdown

Vuti

Don't know how things are made in the US but here in Europe most of the OEM turbo applications I have seen (and I really have seen a few) are doing this the right way. Some manufacturers are even using electric pumps to circulate water through the turbo because natural thermal siphoning can't be done.

Don't get me wrong, I'm not trying to insult anyone here J
Just doing some RD on the subject... Because it seems that there's room for improvement.

18psi

iTrader: (76)

No one is insulted. But all those fittings and (I'm guessing aluminum) lines are just more failure points that you're adding to the setup, and since your water feed line has the turbo as the highest point, and the return line goes up after going down, I'd say you've failed in your R&D

Leafy

iTrader: (1)

IIRC Fiat is using an electric pump for the turbo water. And the only person I know who owns a new fiat (is it weird that I know more people who own x1/9's than 500's?), has already had the pump **** the bed under warranty, and it throws a check engine light.

Vuti

Which is better?
1. Feed from the thermostat housing (water level over the turbo water in) vs. feed from the block near the oil feed (level under the turbo)
2. Return to mixing manifold (way under the water out on the turbo) vs. back of the block (about the same level)

Not mentioning eliminating the up and down kinks on the hoses?

18psi

iTrader: (76)

I don't understand what you're saying:
-to me that picture you posted of your setup clearly looks like the feed is about same level as turbo, and bends up then bends down to get to the turbo. So water wouldn't keep flowing through there. Maybe its the angle of the picture, but are you saying that from source on the head to the inlet of the turbo it is completely downhill?
-also your return goes down, then up, then down again.

So you're saying that your very very marginally better routing is going to solve the alledged flawed routing design?

I don't think so. I think your design is almost as bad.

Braineack

iTrader: (62)

1. who cares.
2. who cares.

18psi

iTrader: (76)

If you're gonna say our routing is flawed, and attempt to fix it, you're gonna have to actually fix it LOL

So either implement a reservoir tank above the turbo or install an electric pump or runn a bottom mount with uncompromised lines going directly downhill

Otherwise you're just pulling a Hyper

Vuti

Oh please, I'm talking all the time taking the feed from the block near the oil feed.



On AbeFM's picture on my first post I'm only trying to illustrate the idea for the coolant return higher in the system. In his case back of the head.

The water feed on the block is the lowest point where you can take the feed after the pump (high pressure). The highest point where you can return your coolant is back of the head after the thermostat (reroute) or your upper water hose (low pressure).

18psi

iTrader: (76)

You have it backwards I think.
The mixing manifold is where it returns, and water neck is where it feeds.
So in that original picture it is flowing from highest (head), to turbo, to mixing manifold, which is exactly what you're doing in your routing, except from the back of the head.

Am I missing something?

The only ones of us that draw coolant from the block are the pre-96 guys, most of us use the routing in original picture (mixing manifold/water neck)

ScottyP3821

iTrader: (2)

It just seems like something too difficult to calculate results. If you plan on doing what your talking about we will have to wait some 100k+ miles to see if any benefits? my motor will probably go before than and Ill be on to a different build. Too many variables for me.

ScottyP3821

iTrader: (2)

What do you guys think of my set up. Its routed by the throttle body.

18psi

iTrader: (76)

You're essentially doing what OP is doing, just using longer routing lol

ScottyP3821

iTrader: (2)

Haha thats what I kind figured.

Vuti

Forget the picture already.

I'm sorry but you missed the point completely and thought it backwards.

1. Car is running
When taking the feed from the block, water is high pressurized by the water pump, it then goes uphill to the turbo and continues to back of the head after the thermostat where is low pressure (return). Constant uphill.

2. After shutdown
Cold water from the feed (block) rises up thought the turbo with thermal siphoning and exits at back of the head. Constant uphill.

WIN

Greasemonkey2000

iTrader: (3)

Self declared winners, even if they are truly the victor, are RARELY celebrated or praised.