Cracked heads

[krede]

Cylinder heads that is..

I know this topic has been up before.. both here on the froggy forum and on the Murena forum I also attend.

But there seem to be lots of different views as to why the cylinder heads crack so frequently.

Everything from poor ventilation in the engine bay (Murena) from poor alloy and fabrication has been mentioned...

Others claim that they only crack if overheated from lack of water.. etc etc... and some say that its the turbo charging that's the cause

That they crack is undeniable.. My murena has had a new cylinder head (before I bought it), and I know of at least two more that has as well...not to mention some of you guys...

So.. just wanted to know.. what's the general opinion .. why do these cylinder heads crack?

[V-M]

I would say there is 3 main reasons for cracks, and usually they goes in 2 and 3 cyl cause they get most heat:

1. Bad cooling circulation design (effects easily if running low water)

2. Bad aluminium material (a lot of bubles and crap inside of aluminium)

3. (dam forget 3:rd one... I'll add this when I remember it )

V-M

[krede]

1. Bad cooling circulation design (effects easily if running low water)

Indeed I believe that was what happened to Hans Murena (I think you know him).. He lost the coolant doe to a leaking coolant pipe, and before he could get off the road it overheated and cracked the head.

I have bought an electric water pump that should hopefully improve cooling flow... especially at low revs, and by continuing running a while after the engine has been shut off to prevent heat soak.

[Bean]

All quite possible...

But from what I've heard and what I think is a significant contributor is the coolant level. The bottle on the 505 Turbo is at the same level or lower than the head - so the chances of not having a completely coolant full head at all times *might* be an issue. Wouldn't take much for a hot spot to form if a bubble or whatever forms which then lead to failures down the road due to it happening over and over.

So in my cars I always raise the bottle in the bracket as high as the hood allows. For my project I'm going to use a different coolant overflow tank mounted to the firewall.

With proper bleeding of the system, and the coolant tank well above the head - it should ensure the head is always absolutely full of coolant at all times. I also run Redline Water Wetter to make sure maximum heat transfer happens between coolant and aluminum.

That's my thoughts on the matter anyway...

Rabin

[krede]

Hey Rabin.

My mate Hans has fitted the standard murena coolant header with a level sensor, so he will never run out of coolant ever again.. And I plan to follow his example by using the header from a Mercedes GD which has adequate capacity, similar connections, and a build in level sensor..

[Bean]

Yes - I know the older VW's have a high mounted tank as well - but the size is smaller than I'd like and no level sensor.

It'd be nice to find a nice OEM solution like the one you mention and then wire up the level sensor signal the 505 Turbo already has in the radiator to work on the overflow tank.

I'm fairly confident that this will go a long ways to preventing the cracking issue - or at the very least it's drastically minimizing the possibility of it happening.

Proper cool down / warm up procedures are also key. IE let it warm up before caning it, and let it cool off after a good caning...

Rabin

[N9TE]

It's pretty much a given that they're not the most robust heads out there. However, there are a couple of things to consider.

You get ZERO overheat events. Iron block and aluminum heads don't allow for this. At a minimum you're looking at a blown head gasket.

To ensure this, you have to make sure the cooling system is at factory spec. That doesn't mean what they put in there still exists. You have to make sure the electromechanical clutch fan works properly. You also have to bleed it "by the book". Plenty of topics on both subjects. Hoses also need to be in excellent shape.

Your cooling system is at least as important as your oiling system. Few people treat it with the importance it deserves.

Proper warm up and cool down. Easy. DON'T hammer it until the OIL is at normal operating tempature. Water will get there first, but wait until the oil gets here also. Oil plays an important role in heat extraction of an engine also. You have to allow for the proper expansion of the iron block and aluminum cylinder head before you start bouncing off the rev limiter.

Cool down. Easy. Just drive it sanely before shutting down the engine. Do this 5-10 miles before you turn off the car.

Too often people assume "more is better". While true for money, sex & guns, it is not always true for engine systems. A high flow fan or higher flow this or that is not better. There have been many cases where coolant rushes through the system at a pace that doesn't allow for the maximum amount of heat to be transferred to the coolant. Don't do this. Just get your cooling system 100% factory in performance. If you want to tinker around with "extra this" and "more that", measure your results. Get a pyrometer/thermo-couple and mount it into the head. Preferably two or three of them.

[Bean]

Great advice Koll...

Only thing to consider though is that Krede is building a turbo motor out of an NA engine - so it's not factory to start.

Turbo will have increase cooling needs, so ideally you need to have it comparable to the 505 Turbo cooling system.

I'll also pipe in that factory is not always the best. You can make improvements on factory systems, but like Koll mentions - it needs to be well planned and thought out so you don't go over board.

I talked with a guy that had put every fancy heat coating into his engine build up (pistons, bearings, combustion chamber, valves etc etc), monster cooling system, and after he got it running - it ran so cool it would never hit operating temp! Had a bugger of a time tuning it and had to do a bunch of stuff to get and keep heat in it...

Rabin

[krede]

I agree that the cooling system rarely get the attention it should.

That many cars keep running in spite of this, must be that manufacturers build in a lot of extra capacity into the systems.

The Murena used to have a reputation for overheating (as did the mr2).

This was mostly based on rumours and assumptions that " mid engined cars overheat", when in fact it had much more to do with poor maintenance.

As for the capacity, a stock Murena cooling system should be able to cope with a n9t engine. That is at least what people that has done such a conversion tell me.

But I will probably fit an oil cooler to give it a hand.

[Pugsley]

Too often people assume "more is better". While true for money, sex & guns, it is not always true for engine systems.

Very true in those cases

[Johnny]

I got my hands on a Chrysler-Simca 180 head from 1977 which I believe is the original design.

As you can see in the picture below, it doesn't have a support under the exhaust port, I think this might have something to do with the cracking. From what I've read on the net, the Chrysler/Simca version of these engines didn't have a problem with cracking heads.

According to this site: Translated French Chrysler site, It seems like it was a quite reliable engine that could exceed 300 000km if maintained properly. The head gasket could fail at 150000 km (could be due to owners not letting the engine warm up and cool down properly), but nothing about cracking heads.

I have a little theory about this,

Metal that get hot expands. And if you restrict that expansion, you'll get thermal stress. When the engine cools down again, the metal goes back to its initial state. So what you get is a periodic thermal load, which in turn leads to metal fatigue and a crack starts to propagate. I think the reason that there was no support initially was to let the metal expand instead of restricting it's movement and get more thermal stress.

What do you guys think of this?

[V-M]

With turbo heat load in exhaust side is bit different thing than N/A engines. IMHO

I have also that same 180 head and block in my carage.

V-M

[Johnny]

Yes of course, but the N/A Matra Murena seems to have the same problem with cracking as our engines.

[Bean]

Bringing this back to add another possibility to the discussion:

In researching how to fix cracked heads I read one builder's thoughts that the cause for cracks propagating from the valve seats was caused by valve seats being pressed into the head casting with too much of an interference fit.

Throw in the different expansion rates, and the possible coolant flow issues, and it makes pretty good sense to me that the seats themselves are a contributing factor.

Rabin

[bryancohnracing]

In regard to Coolant Expansion Tanks, my solution can be found here:

http://www.505turbo.com/forum/index.php?/topic/2114-winter-maintenanceservicetrack-preparation/page-2#entry15339

Also, I made a T-Stat housing bleed valve that works GREAT! See Post #36 on the same page.

The expansion tank is a $40 Speedway Motors part, found here:

http://www.speedwaymotors.com/Speedway-Coolant-Expansion-Tank,1840.html

I'm sure you can find these worldwide.

The T-Stat bleed is a steel tire valve. I've used them to make bleeds in other radiators and all have lasted without fail, leaks, etc. They will stand up to the heat as wheels get hotter than coolant due to tire friction and brake temperatures. As air is a fluid like water and a tire valve will hold air pressures high air pressures (say at least 75 psi or more) and cooling systems only run in the 15-25 psi range its no issue.

I use a high pressure cap, 22-24 psi.

See the entire build thread for all kinds of fun stuff!

Good luck!

Edited December 29, 2012 by andrethx
first link fixed to go directly to post #27

[Bean]

Damn - I've been searching for coolant swirl pots and here I find the exact thing that should work perfectly for what I want right here... Totally forgot that you did that tank - and it should be exactly what I want.

Curious if you had a specific reason for doing the 22-24 psi cap as I think the stock cap is supposed to be14 psi... Found a Gates rad cap on the Speedway site that was 13 psi which I thought would work fine.

BTW: I've been studying the coolant routes on the N9TE motor, as I wanted to make the new tank a proper swirl pot that allowed the system to vent air naturally. Current plan is to run the one hose off the coolant tank that goes down to the coolant pipe into the bottom of this expansion tank - and the side port down to the coolant pipe. The current rad hose connection to the current plastic tank will get "T"'d into the bottom fitting as well - but still not sure where to best take that hose.

That should allow any trapped air in the system an easy route out, and the expansion tank itself mounted to the intake manifold most likely - but I'd have to noodle around with that once I get it. (I'll have to build up the wish list a bit and get a bunch of things at once)

Any comments on that idea for placement of the tank?

Rabin