Nice George! How about the one below? It took us a while from the picture to figure it out.
1 - 20 of 24 Posts
I ran this photo and others past a few wise engine builders. Turns out the radius, seen in the background journal, has some harmonic chatter marks in it from when the crank was cut and polished. These are machine tool marks. If you look at the other side of the fracture, it begins at the root of one of the chatter marks in the radius. Stress risers are always a problem, no matter where they are.
That's exactly where my 1600 crank broke, and from what I've read, how they all break. Starts at a journal radius and goes straight through to another journal radius.
Andrew
Andrew
I'm sure there are other ways to break them Andrew, but I consider this one the innocent owner syndrome, where the crankshaft grinder and polisher felt he might get away with dull tools and a poor polish at the radius. You will see more like this in time. It might be fine with a 5,000 rpm GM daily driver engine, but it does not work with Alfa's in my experience.
I was in the PCA for some years and have all the "UpFixen Der Porsche" collections. Harry Pellow ("The Maestro") was a renowned, charismatic 356/912 engine builder here in the Bay Area til he died 10ish years ago. Before turning to Porsches, he was a nuclear engineer, handy machinist, and pretty qualified. He claims to have built on the order of 500 Porsche engines, and loved to tear them down, find the problems that caused self-destruction, and write them up in his books and in PCA tech sessions. His books are a hoot if you don't get bogged down in his personality.
All that said, he claimed every 356/912 crankshaft failure he ever analayzed (not simple broken conrods, etc.), and there were hundreds, was exactly on this pattern. And most of those were due to a poor radius on the journal edge. Some were mere fatique, but that's where he said the crack starts, either way.
On my 1600, I looked very closely at the journal radius and could see it had not been recut when the crank had been ground undersize. There was a partial curve of the original filet, then an abrupt 90* drop-off where the crank grind cut to the new dimension, but did not fair in the old filet. On both main and rod journals.
I'm a believer.
Andrew
All that said, he claimed every 356/912 crankshaft failure he ever analayzed (not simple broken conrods, etc.), and there were hundreds, was exactly on this pattern. And most of those were due to a poor radius on the journal edge. Some were mere fatique, but that's where he said the crack starts, either way.
On my 1600, I looked very closely at the journal radius and could see it had not been recut when the crank had been ground undersize. There was a partial curve of the original filet, then an abrupt 90* drop-off where the crank grind cut to the new dimension, but did not fair in the old filet. On both main and rod journals.
I'm a believer.
Andrew
Forged crankshafts have some intentional flex to them so they don't need to be massive. If you add any stress-risers, at all, you are asking for trouble.
So Gordon,I ran this photo and others past a few wise engine builders. Turns out the radius, seen in the background journal, has some harmonic chatter marks in it from when the crank was cut and polished. These are machine tool marks. If you look at the other side of the fracture, it begins at the root of one of the chatter marks in the radius. Stress risers are always a problem, no matter where they are.
How much were the journals cut?
In summary: What are the lessons learned?
Thanks,
b
I believe this was a .020 cut, not more. The lessons learned are that ONLY a shop with a reputation and history for top quality work on forged crankshafts should be used. All crankshafts being considered for any cuts should be crack checked prior to any work, as well as checked for straightness before any work. Correct radius cuts with proper tools are part of the job, as well as a final polish to completed dimensions in all machined areas. After work is complete, the crankshaft again needs to be crack checked, and qualified straight. If re hardening is done, it needs to be done by a shop that either manufactures forged crankshafts for the racing industry, or does harding as a regular part of their undersize recutting business.
The difference in costs between the best shop you can find, and someone who claims to be able to do the work, can be the build cost of your entire engine.
I hope this helps Bill.
BTW: RJR just called asking for your phone number. Better e-mail or call him.
The difference in costs between the best shop you can find, and someone who claims to be able to do the work, can be the build cost of your entire engine.
I hope this helps Bill.
BTW: RJR just called asking for your phone number. Better e-mail or call him.
Sound advise from one who knows. Folks please take note.
Gordon,
Have you ever seen a standard journal balanced crank with light weight balanced piston / rod / flywheel assemblies attached to it break like this.?
b
Have you ever seen a standard journal balanced crank with light weight balanced piston / rod / flywheel assemblies attached to it break like this.?
b
Not yet. However if you got high enough engine speeds with good rods and pistons, you might set up some nasty vibrations. MANY years ago it turned out to be possible to crack the main bearing webs out of GTA blocks without much obvious damage to crank, rods or pistons. Naturally, if the engine ran like this more than a few seconds it became aluminum scrap. Enough were shut down when the oil pressure dropped to diagnose the failure, and Alfa (Autodelta) redesigned the blocks, both race and street, with different webs.
With the older blocks the rule was not to exceed 7500 rpm for more than absolutely necessary or you might suffer a failure. I believe these were from accumulated stress build up, as the previously stressed blocks that showed no sign of failure, would fail later in racing at lower speeds. Early on, until the "new" blocks arrived, we tried both steel and aluminum bottom end cradles. The machine work was daunting to both fit these and make them work at that time. Six bolt main fastening was the extreme version like the GM Chevies. This was a heck of a job, and I have not seen one of these blocks that has survived, though one or two are probably out there. The whole bottom of the block was re-engineered by guess-and-by-gosh.
It is possible this was caused by the light, 2 ring pistons, and lightened rods. The Autodelta GTA cranks were left as massive as possible, as obviously, any flex was the end of the story.
With the older blocks the rule was not to exceed 7500 rpm for more than absolutely necessary or you might suffer a failure. I believe these were from accumulated stress build up, as the previously stressed blocks that showed no sign of failure, would fail later in racing at lower speeds. Early on, until the "new" blocks arrived, we tried both steel and aluminum bottom end cradles. The machine work was daunting to both fit these and make them work at that time. Six bolt main fastening was the extreme version like the GM Chevies. This was a heck of a job, and I have not seen one of these blocks that has survived, though one or two are probably out there. The whole bottom of the block was re-engineered by guess-and-by-gosh.
It is possible this was caused by the light, 2 ring pistons, and lightened rods. The Autodelta GTA cranks were left as massive as possible, as obviously, any flex was the end of the story.
Hi Gordon,
You have piqued my interest yet again. You refer to webbing, so you obviously do not mean the change to HD rear bearing mount. I wonder whether these old GTA blocks were the ones numbered 00502/A and the later changed ones were the 00532 series. I assume this change was done also on other 1600 blocks. I was never aware that there was actually a change in webbing design at some point in time on those blocks. Don't really want to be impertinent, but you don't have a picture of the change by any chance.
You have piqued my interest yet again. You refer to webbing, so you obviously do not mean the change to HD rear bearing mount. I wonder whether these old GTA blocks were the ones numbered 00502/A and the later changed ones were the 00532 series. I assume this change was done also on other 1600 blocks. I was never aware that there was actually a change in webbing design at some point in time on those blocks. Don't really want to be impertinent, but you don't have a picture of the change by any chance.
Not yet. However if you got high enough engine speeds with good rods and pistons, you might set up some nasty vibrations. MANY years ago it turned out to be possible to crack the main bearing webs out of GTA blocks without much obvious damage to crank, rods or pistons. Naturally, if the engine ran like this more than a few seconds it became aluminum scrap. Enough were shut down when the oil pressure dropped to diagnose the failure, and Alfa (Autodelta) redesigned the blocks, both race and street, with different webs.
With the older blocks the rule was not to exceed 7500 rpm for more than absolutely necessary or you might suffer a failure. I believe these were from accumulated stress build up, as the previously stressed blocks that showed no sign of failure, would fail later in racing at lower speeds. Early on, until the "new" blocks arrived, we tried both steel and aluminum bottom end cradles. The machine work was daunting to both fit these and make them work at that time. Six bolt main fastening was the extreme version like the GM Chevies. This was a heck of a job, and I have not seen one of these blocks that has survived, though one or two are probably out there. The whole bottom of the block was re-engineered by guess-and-by-gosh.
It is possible this was caused by the light, 2 ring pistons, and lightened rods. The Autodelta GTA cranks were left as massive as possible, as obviously, any flex was the end of the story.
I don't really know the numbers, but can get the time frame at 1966. The rear main got a rework with the 101 1600 Veloce engines, but the upper web was somehow stiffened about late 66 to 67. The Autodelta replacement blocks were unnumbered. I wish I had thought at the time that these details would be of future historical importance, and carried my Cannon 35 mm around with me the way I now carry my Cannon digital camera everywhere. You might send Karen Dale McGowan, <[email protected]> an inquiry about this subject as she was with Ron Neal at PROTOTYPE engineering after the AUSCA years. Ron was still the master with these engines at the time Karen worked there. She may have more information, but probably, like me, no photo's. I will forward these posts to her E-Mail as she may have comments that differ from my own.
I hope this helps.
I hope this helps.
Thanks for the input Gordon. Just know that the early blocks were all 502/A and the later ones 532 and that some of the factory cars with earlier chassis numbers later ended up with some 532's. I believe that the blocks were only numbered if the entire engine was built up at the factory but the replacement blocks were all unnumbered. Some special engines seem to have an AD with a number rather than the regular number stamped.
I have a late car with a 532 motor in it. I was always curious what was different. The story with the webbing makes sense given some of the early crank failures on GTA's during races, e.g. Businello at the 6h Sandown and others.
I wish I would have looked more carefully the last time it was apart and compared to the other 1600 blocks I have around - and taken some pictures too ...
I have a late car with a 532 motor in it. I was always curious what was different. The story with the webbing makes sense given some of the early crank failures on GTA's during races, e.g. Businello at the 6h Sandown and others.
I wish I would have looked more carefully the last time it was apart and compared to the other 1600 blocks I have around - and taken some pictures too ...
I don't really know the numbers, but can get the time frame at 1966. The rear main got a rework with the 101 1600 Veloce engines, but the upper web was somehow stiffened about late 66 to 67. The Autodelta replacement blocks were unnumbered. I wish I had thought at the time that these details would be of future historical importance, and carried my Cannon 35 mm around with me the way I now carry my Cannon digital camera everywhere. You might send Karen Dale McGowan, <[email protected]> an inquiry about this subject as she was with Ron Neal at PROTOTYPE engineering after the AUSCA years. Ron was still the master with these engines at the time Karen worked there. She may have more information, but probably, like me, no photo's. I will forward these posts to her E-Mail as she may have comments that differ from my own.
I hope this helps.
This is probably of no help, but examining the webs in some engines I have had apart, whatever the difference, is not easily apparent. Perhaps if I had seen two, side by side, it might be noticeable. The basic architecture must be very similar.
As this change was not limited to just the GTA blocks, that would seem to suggest all late 1600 and all 1750 blocks were strengthened in the web area. Perhaps weighing a stripped early 101 1600 block and comparing that to a later, but seemingly identical 105 1600 block, might show a weight difference. I don't know.
As this change was not limited to just the GTA blocks, that would seem to suggest all late 1600 and all 1750 blocks were strengthened in the web area. Perhaps weighing a stripped early 101 1600 block and comparing that to a later, but seemingly identical 105 1600 block, might show a weight difference. I don't know.
Don't feel so bad about not having caught it then
This is probably of no help, but examining the webs in some engines I have had apart, whatever the difference, is not easily apparent. Perhaps if I had seen two, side by side, it might be noticeable. The basic architecture must be very similar.
As this change was not limited to just the GTA blocks, that would seem to suggest all late 1600 and all 1750 blocks were strengthened in the web area. Perhaps weighing a stripped early 101 1600 block and comparing that to a later, but seemingly identical 105 1600 block, might show a weight difference. I don't know.
Gordon,
And so the moral of the story is: Thou shall not grind Alfa crankshafts lest they become prone to failure.
On the block issue: your saying that there were feature changes made to the block. What about alloy changes to the casting aluminum that was used or post casting operations (i.e. normalizing / stress relieving). Did any failures occur on mature blocks with many thermal cycles on them or mostly green race motors??
BTW: I happen to have a pristine late 105 - 1600 block out in the garage with a standard crankshaft, head and internals to match. I can take fotos if there is something to compare it to. Will check the numbers on it as well if there is any interest.?
I also have a 1750 block / standard crank / standard head as well.
b
And so the moral of the story is: Thou shall not grind Alfa crankshafts lest they become prone to failure.
On the block issue: your saying that there were feature changes made to the block. What about alloy changes to the casting aluminum that was used or post casting operations (i.e. normalizing / stress relieving). Did any failures occur on mature blocks with many thermal cycles on them or mostly green race motors??
BTW: I happen to have a pristine late 105 - 1600 block out in the garage with a standard crankshaft, head and internals to match. I can take fotos if there is something to compare it to. Will check the numbers on it as well if there is any interest.?
I also have a 1750 block / standard crank / standard head as well.
b
Cranks can be easily ground and refinished, but ONLY by a shop that knows forged racing crankshafts.
"On the block issue: your saying that there were feature changes made to the block. What about alloy changes to the casting aluminum that was used or post casting operations (i.e. normalizing / stress relieving). Did any failures occur on mature blocks with many thermal cycles on them or mostly green race motors??"
I believe there may be measurable alterations. I have not measured. They may alter the weight of the stronger blocks. Just a guess, again, I do not know.
There have been casting alloy changes, but not in the period of the blocks that failed and the later, tougher ones. The failure of used blocks was determined to be stress failure, more specifically, accumulated stress failure. Once they were stressed by very high engine speeds, they could fail at lower engine speeds at some later time. The structural damage was not visibly apparent. There were no dimensional changes (warped block) that occurred before failure. This may have been a problem with a single casting run. At this late date, there is no way to tell. It was not a failure that would turn up on a street engine of that time.
I too have various blocks, but no way to really look into what may have been the cause or solution back in time. You would need a failed GTA block of a particular series, and another of a different series that had been stressed without failure for any valid comparison. Very unlikely 45 years later!
As always, just my opinion.
"On the block issue: your saying that there were feature changes made to the block. What about alloy changes to the casting aluminum that was used or post casting operations (i.e. normalizing / stress relieving). Did any failures occur on mature blocks with many thermal cycles on them or mostly green race motors??"
I believe there may be measurable alterations. I have not measured. They may alter the weight of the stronger blocks. Just a guess, again, I do not know.
There have been casting alloy changes, but not in the period of the blocks that failed and the later, tougher ones. The failure of used blocks was determined to be stress failure, more specifically, accumulated stress failure. Once they were stressed by very high engine speeds, they could fail at lower engine speeds at some later time. The structural damage was not visibly apparent. There were no dimensional changes (warped block) that occurred before failure. This may have been a problem with a single casting run. At this late date, there is no way to tell. It was not a failure that would turn up on a street engine of that time.
I too have various blocks, but no way to really look into what may have been the cause or solution back in time. You would need a failed GTA block of a particular series, and another of a different series that had been stressed without failure for any valid comparison. Very unlikely 45 years later!
As always, just my opinion.
Thanks Gordon.
BTW: I emailed RJ, got no responce so it must not be that important?
b
BTW: I emailed RJ, got no responce so it must not be that important?
b
1 - 20 of 24 Posts
- This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Alfa Romeo Forums
A forum community dedicated to Alfa Romeo owners and enthusiasts. Come join the discussion about performance, parts, modifications, troubleshooting, maintenance, and more!
Full Forum Listing
Recommended Communities
