The 355-Exposed? Covered Up? or Overlooked ????

I dont know, but a lot of things on a car fail. The more performance you gain, the more difficult it becomes to try to keep it together. How hard were these cars driven, how well were they maintained, etc., those questions add into it too. I have cracked headers on my 308, its rather common on them too. But they arent making 109 HP per liter and running as hot either. This is why I am suggesting you expecting a lot. I dont think other than a Lambo, there are any other production cars with tube headers. And I dont know of any header manufacture that would warrantee a header longer than three years. The only other place you commonly see them is on aircraft, and all of those crack sooner or later too.

 

Good info there bruce!! Though I think the 355 has similar lifters as the mondial/348 in that they have a 'tube' to slide in to insulate them from any lateral loads. Here is a pic I borrowed from lusso64's 30k mondial thread.
Image Unavailable, Please Login

 

Acronyms ................ I need definitions

 

Wait no longer!

 

Here's the valve guide analysis, thanks to Dave for providing all these parts. As you can see from the picture, the guide color gives a good indication of the % Cu. At this point I'm not going to draw any conclusions from this, this is just another piece of the puzzle. From a chemistry standpoint it seemed like Ferrari took a step backward on the 355, until hardness testing revealed it had the highest hardness of the bunch (slightly). I haven't analyzed a new post '98 355 guide yet but will add that when we get one. After doing some reading and talking with automotive engineers one critical point was stressed - clearance between the guide and valve are absolutely critical and must be 'zero'. So one thought is that perhaps some 355's had the perfect 'zero' clearance and some didn't, resulting in a scattered failure rate. Then switching to a more durable guide made the assembly more robust to handle an ever so slightly wider range of clearances. (basically echoing what Bruce just stated above) That's all I'll theorize, and let the guys with infinitely more experience than me to comment!
Image Unavailable, Please Login
Image Unavailable, Please Login

 

The 330 mid production to 330 late production makes a big change in zinc composition. If late 330's had fewer valve guide failures than mid production 330's then we need to look at Zinc?

Interestingly, 355's have similar levels of Zinc to the mid production 330...while the 348 valve guides have levels of Zinc like the late production 330.

Valve guide failures aren't an issue for the 348.

What about mid production 330's?

 

Yes much of this hinges on the actual failure rate, which we can only get anecdotal evidence of from the independents (unless we can get an insider to post actual data!!!) Interesting to me was the late 330 was identical to the 348 challenge, evidently they liked something about that one. And the late 330 and 348 challenge are bronze, while the others are really just high grade copper alloys (having more than 96% copper)

 

Nice work Jay, thank you for your efforts with this.

Knowing the composition and hardness is one facet of the equation. Add to this the load and pressures being exerted on the guides as yet another factor. The 355 uses 6MM stems where none of the others did. The 355 is running RPM's and valve acceleration rates exceeding the others, running FAR hotter than the others, and running a hydraulic lifter maintaining zero (or near to that) clearance. The environment the 355 stem is exposed to and the loading on the smaller stem far exceeds what the other models are seeing in average use.
Brian Crall and I have long theorized that the initial assembly clearances varied car to car and was likely one factor in the high failure rate. With the small stem diameter, heat and RPM present in the 355 any "extra" clearance would cause a wear rate that would be exponential in nature. The 355 guides were simply not hard enough to handle the loads they were exposed to for very long.
When the stem diameter dropped from 7mm to 6mm and the RPM and heat rose by (?)% the "hardness of the guide" should have followed in direct proportions. They did not.
As the stem started walking around in the guide the seal was no longer able to control oil flow allowing a great deal of oil down the stem and into the combustion chamber. Oil mixed into the combustion process lowers the effective octane rating of the fuel being burned in that cylinder and thus causes another array of problems.
Along comes the need for a "harder" guide and the steel replacements become available. Personally I have yet to see a failure of any of the steel guides. As Bruce wisely stated, this is an area that requires perfection in assembly and machining of both the guide fit and the seat to valve contact. I have seen many repair orders from both dealers and independent shops that state "Lap Valves" after guide installation. I assure you these will fail in short order and we will start to see burned valves as a result of the failure to properly machine both the seats and the valves after the guide replacement. We are talking about seconds of a degree in stem angle making all the difference in the world here. Even with the finest seat cutter machines it will require a very experienced machinist (that gives a damn) to hold to these tolerances. With that issue addressed will the same person be willing to accurately measure 40 valve stem heights before the machining process and accurately machine the stem ends back to the original assembled height. Remember, these have hydraulic lifters with a very narrow acceptable stem height range. Glass beading combustion chambers is a thing of the past (should have gone away when aluminum heads came about in my opinion) and few shops own a soda blaster to properly clean these heads. The 355 heads are riddled with very small oil flow ports to supply the hydraulic lifters that will not blow out clean with compressed air. All the threaded plugs need to be drilled and replugged after a complete cleaning if the heads get anywhere near a bead blaster.
Did the machine shop pre-bore the guides to a very thin wall thickness, heat the heads in an oven to expand the aluminum and then gently press out the thin walled guide resulting in no gauling of the guide bore? Nope, I have had to machine a number of heads for oversized OD guides to date.
There are so many factors involved in analyzing the failures of these components. I had to moth ball my $20K seat cutter and rely on my machinist who invested in a $130K cutter with digital readouts that is required to properly do this job. The seat angles and widths are so critical and so small it requires this type of investment in tooling and expertise.
When we start to see a high failure rate of the valves or steel guides we will have to be careful about condemning the components and look closely at the expertise that went into the last machining process.
Bruce said it best, DO NOT CUT CORNERS here!

Dave

 

Dave, the critical tolerances/machining are amazing. That could easily explain the situation.

On another note, if *materials* have any share of the blame, I'd think that we'd be able to figure that aspect out by examining the performance of mid production 330 valve guides versus late production 330 valve guides.

Was there any noticeable difference in valve guide failures between mid to late 330's?

 

No, none that I have noted.

I think the steel guides would have been appropriate from the get go in terms of hardness considering the 6mm stem and the engine parameters. The reduction in supported surface area from 7MM to 6MM is quite drastic. With that said, I do share Bruce's concerns about the material selection but I have seen nothing to support these concerns.
I felt just fine putting steel guides in my 1951 Ferguson tractor. To date all of the uses I have been exposed to where steel guides are present have a redline of under 6K.
Material advances come about way faster than I can keep track of today. They may be just fine and I am judging them on my generic knowledge of the material from 15 yrs ago.

It must be mentioned at this point that these are only my latest theories / thoughts and are not meant to be stated as fact. I do have a fair amount of recient tests and data that once reviewed seem to support these ideas.

Dave

 

Thanks.

Based on your comment above that the 330 mid production had no known difference in valve guide failures from the late production 330's, and considering the material composition of mid to late 330 valve guides compared to 348 valve guides and 355 valve guides, then we can conclude at this point that:

MATERIAL COMPOSITION IS NOT AT FAULT in 355 valve guide failures.

 

You must explain your logic and reasoning as it makes no sence to me.

How did you factor in surface area (supported) and operating conditions and reach this conclution?

I offer up thoughts I have based on as much data as I can obtain. Believe what you want, everyone will draw their own conclutions based on the debate but for goodness sake do not state things as fact.

Unless you have additional data that is not being layed out, you can conclude NOTHING from this so far.

Dave

 

My logic is this:
1. the chemical composition as posted above for 330 mid production is strikingly similar to 355 valve guide composition, but the 330 mid production series (per your comments) had no unusual valve guide failure rates compared to late production 330 VG's.

2. the chemical composition as posted above for 330 late production valve guides is strikingly similar to 348 Challenge valve guides which did not have unusual valve guide failure rates.

CONCLUSION: material composition itself (not assembly, not tolerances, not design shape, not stem size, not heat/conditions/pressure, just the chemicals involved) is NOT to blame for the primary reason of 355 valve guide failures.

 

I'm not so sure that the cars are similar enough in design and operation to come to that conclusion. For example, we know that the 355 exhaust temps are much higher than the other cars; and max revs of 8500 are much higher than a 330. Would that not influence valve speed? Also, there may be a few other variables not being considered, such as valve material composition, that could influence the choice of valve guide material.

 

My point is more simple. The *materials* are not at fault based on the comparison to mid and late 330 valve guide materials, 330 valve guide failures, and 348 valve guide materials/failures.

Could "design" be at fault? Yes. Possibly. Could assembly? Yes. Possibly.

Could poor cam timing setting? Yes. Possibly.

...But we can rule out the materials per se based upon the data posted in this thread.

 

Dave, do you have any info on the 355 guide that was tested,e.g. bore(each end, and the middle), mileage of car, int or exh(from color it appears to be int), and any noticeable wear of valve or seat, any carbon build-up on backside of valve?? I know you probably don't have any of this , but if you do I would like that info.


They were bell mouthed on both ends like all of the failed guides. Std, bad guide carbon deposits. The guide to stem clearance on this engine was bad enough that the seal was unable to follow the stem. Std seat scuffing from the valve sliding into a seat position.
I have my machinest bore the ID to less than 1MM wall thickness prior to removal so to save the bores in the head. I have had to machine dozens of OS guides to clean up "quickie" guide replacements and this has eliminated any gauling issues. I only remove 1 valve these days just for data collection as I have FAR too much time invested in this already. On this head the valves were dancing around ready to change holes.


Jay, sorry for the delay in responding to your analysis post. I do want to say a big thanks for the work. Because of your efforts it is now possible for the first time to actually see the factory receipe.

Boy, isnt that the truth! Without Jay's efforts we would be far behind the curve. So much of this effort is just to dot the i's which is something I need before making major changes to a Ferrari design. Accurate data like what Jay has provided doesnt come cheap! A big thanks again.


My point is more simple. The *materials* are not at fault based on the comparison to mid and late 330 valve guide materials, 330 valve guide failures, and 348 valve guide materials/failures.

Could "design" be at fault? Yes. Possibly. Could assembly? Yes. Possibly.

Could poor cam timing setting? Yes. Possibly.

...But we can rule out the materials per se based upon the data posted in this thread.


I have read it over and over and it still doesnt make sence to me.
What works for one engine does not mean it will work for another of a completely different design. Material composition for a 330 running at low RPM, cold exhaust temps (compared to the 355), and large diameter valve stems is completely inadiquate for use with a 8500 RPM, 6MM valve stem.
Cam timing has nothing to do with it, the guide doesnt care when the valve opens.
Design may be a fault to some degree but if the geometry is correct all should be OK.
Assembly is, as I stated before, still a thought but cant be verified. It is the one likely theory why some lasted longer than others that is left on the table after many others have been proven wrong.

Material specs have to be properly matched to the operating conditions to obtain a proper cycle life. In this case they were not, but there are still two more tests that I am doing to prove my theory and verify my new design is up to the task. The key word is proof, anything less constitutes experimentation with a customers car which is not acceptable in my book. Hell, I bought a second 77 308 just so I had a base line (try explaining and justifing that to your wife!) on designs I am working on, the dyno only tells part of the story. There are a couple of problems with the 355 engine that have to be sorted and redesigned. I intend to own a 95 B soon and I want the long term reliabilty (read values) to be up to what I think they should be.

No matter how you switch the words around, materials used to make the guides that are not matched to the task or operating conditions is still a material fault.

Dave

 

Wow!! What about a material that would better suit the operating environment of the 355? This would not be good if those 'other' potential causes you listed were to blame

 

As far as keeping the heads cool has anyone looked at the engineering of the cooling system? Things like water jackets or water pump capacity? Are there any bottlenecks in the system? More efficient water flow is what I'm getting at here. I know from past experience with another car I was able to get a 30% increase in waterflow in the heads by machining out some flashing that was left over from the manufacturing process. Yes this created a 30% increase in flow something I think we would all be interested in if this was a possibility.
You experts out there, are there any gains that can be made in that area?

Steven