Cracked head... any ideas?

[JoeMag]

Gents -- Have a head that cracked. any ideas why? 964 head on 3.8 jugs

[JoeMag]

more pics.

[stownsen914]

I think it was MBruns who has commented on 3.6 heads cracking on his 3.8 race motors

[MBruns]

We cracked 3 sets on the race engine at around the 30hr point, we welded many areas that we thought needed material when building them, we ran them un welded still cracked top to bottom anyway, we have been running the Xtreme castings and have over 150 race hours total since and not one issue, I think for severe service they are not up to the task, If I remember correctly Bills castings were much heavier comparing bare heads.
Mike Bruns

[Catorce]

Look how porous that casting is for starters......

[stownsen914]

Originally Posted by MBruns

We cracked 3 sets on the race engine at around the 30hr point ... I think for severe service they are not up to the task.

Mike - have you seen similar issues with modified 911 heads other than 964/993?

[MBruns]

All the other heads seem to hold up very well, the turbo heads tend to get soft after a few seasons or long races due to EGT and higher head temps, On the stock 964 they are getting long in the tooth and have seen a few of them giving up too.
Mike Bruns

[VFR750]

Is it related to the large intake ports on the basic casting? Wall thickness getting too thin?

[stownsen914]

Originally Posted by Catorce

Look how porous that casting is for starters......

Yeah jeez. The amount of porosity in that pic is impressive.

[Mahler9th]

I am not an expert. I have had two 993 heads crack in the past 15-20 years. I have heard of others, like those mentioned by Mr. Bruns. I am pretty sure that Jeff Gamroth has seen some, and my friend/engine builder has seen others besides mine. I think at least some of those were 964 heads.

There is a thread on this in these Forums I started years ago.

Some folks believe in welding, others do not. Some have tried and abandoned it.

In my case, the cracks occurred immediately after racing events with very high ambient temps... around 100 degrees. Several of us around here have tried to implement ways of ducting as much cool air directly into the engine fan... I think there is a thread on this as well.

We haven't done any real experiments measuring temps on the backside of the fan... might give that a try since I can now log. But we all figure directing as much air flow to the fan cannot be a bad thing. In all local cases with which I am familiar, we run 3.8-style rear decklids.

I decided years ago to just not run my car when ambient is north of ~95 degrees.

As Mr.Bruns stated, these heads have been around for a long time... and 3.6 heads are getting really hard to find.

In summary, my similar inquiries via these forums and directly with expert builders (like Mr. Bruns, Jeff Gamroth, Jerry Woods and Mat Lowrance) and machinists like Ted Robinson and the late Mike Stimson, suggests that there is no single cause or preventative measure with 3.6 head cracks.

[JoeMag]

Thanks for all the comments. Did not realize this was common. few other comments

- I originally had a very bad tune on the car, had detonation issue wo that's why you're seeing what looks like porous surface (it was welded up).
- regarding head temp, i have found that head temp on my car is fairly consistant at ~300F. Run AIM dash. Interestingly it does not seem to change much with ambient temp.
- This is second life of these heads on race motor. Was on someone elses motor and i bought them. My motor has about 60 hrs on it.

[Mahler9th]

Sorry to hear about the bad tune. Not likely relevant, but just in case...

Most tuning processes use dynos, but some tune on track. I use the latter approach as it is favored by the gentleman that builds my engines.

In any case, oxygen sensors are usually employed.

What I have started to see is a range of supply chain "compromises" with oxygen sensors.

As many know, supply chain challenges exist with automotive parts. I am not in the camp that blames countries like China, India, Mexico, Brazil and others. I feel that this is really a result of the intersection between capitalism and technology.

In any case, I have heard and seen evidence that suggests that you have to be very, very careful with 02 sensors if you are relying on their readings for tuning.

Last year, at the suggestion of my friend/engine builder and his colleagues, I switched from an AEM AFR device to a Ballenger Motorsports product. This link describes one of the things I learned from them:

BEWARE! Fake NTK Sensors are Flooding the Market! | Ballenger Motorsports

After speaking with them and a few others, it seems somewhat clear to me that it can be very difficult to distinguish fake from real.

For tuning, I will only use 02 sensors I buy from them (they take precautions-- if they did not they would have no business). They have been an excellent service provider thus far. I am sure there are others.

My current hypothesis that using resources like Amazon and EBay may more risky as their supply chains may be bigger targets for bad actors.

[Mahler9th]

Here is a pic of one of my cracked 993 heads (from 2009):



The engine at that time was a 3.8 with high compression JE's and RSR cams.

If I recall correctly, it was north of 102 degrees ambient.

Both times I had cracks were at events with high temps or at the next following a hot weekend.

I think there are some other pics in the other thread.

Good luck with your repairs.

[JoeMag]

That’s interesting... that’s exactly where the crack was progressing in mine.