When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
We're building a new engine for Joseph's race car.
His "old engine" (not very "old", BTW) died a terrible death from severe detonation. The crankshaft survived with only minor gouges from the "loose" pieces that flew around inside the engine....all of the journals were perfect and it was still perfectly straight. All of the pistons were overheated/pounded to the point where they distorted and cracked. The upper half of one connecting rod "left" the scene, through the oil pan. The piston pin that was attached to this rod was also completely gone....ripped completely out of the piston. Another connecting rod had a crack down the middle (from the piston boss), the result of the incredible pounding.
While it is impossible to have pieces made to survive severe detonation, we did have the pieces analyzed to see if any improvements could be made. While there were no significant changes suggested, J&E made the "original" pistons and we now use CP Carrillo for piston design and manufacturing.
Here's some pictures of their new design:
Shorter skirt....wrist pin is in the same location, in relationship to the top of the piston. More modern "box" design piston is stiffer Side comparison view of the new "box" design piston
Semi-retired, as of Feb 1, 2023.
The days of free technical advice are over.
Free consultations will no longer be available.
Will still be in the shop, isolated and exclusively working on project cars, developmental work and products, engines and transmissions.
Have fun with your 928's people!
Since I don't have the wrist pin left over from this engine that "expired", I can't tell if the wrist pin broke first and the connecting rod broke second (although the crack in the second connecting rod certainly points to the rod failing), I'm going to error on the side of strength (although I hope this type of detonation never occurs again) and increase the wall thickness of the wrist pin.
Worth noting, the box piston, although it looks like it should be lighter, is also heavier (by a fair amount). I had CP Carrillo add material to the dome of the piston to increase the strength, in this area, too.
Just a matter of adding more Mallory to the crankshaft......
Interesting. Will this engine have piston oil squirters or not? Wet sump or dry sump? If wet sump, any plans to pull a modest vacuum to the crankcase. What are you planning for the redline rpm? Any adjustments to the conrods or bearings with the slight increase in the piston assembly weight?
Interesting. Will this engine have piston oil squirters or not? Wet sump or dry sump? If wet sump, any plans to pull a modest vacuum to the crankcase. What are you planning for the redline rpm? Any adjustments to the conrods or bearings with the slight increase in the piston assembly weight?
No plans for squirters. This particular engine has always been wet sumped. I'm going to try and talk the client into a dry sump, this time around.
RPM's are something we need to look at, confirm, and record in the future. Neither engine got run extremely high in the rpm range under acceleration, according to the driver....and I trust what he says. No idea what happened on downshifts...the tachs were certainly not accurate or fast enough to be of any use, in a downshifting situation. (The tachs in both of the cars that had engine issues were stock 928 tachs out of junk cars....unknown if they were ever checked for accuracy.) One of the first things Carrillo asked for (both times) was a copy of the rpm data. Neither engine had any data gathering capability, so the following discussions were a bit embarrassing.
Carrillo: "A $3,000 go cart has data being recorded, these days. Your customers are running mega dollar Porsche race engines and they can't afford something that records data?"
Carrillo: "You know that generally means they don't want the data to be recorded, right?"
Carrillo: "You seriously want us to tell you something about failed connecting rods from a race engine with zero RPM data?"
Carrillo: "We can look at the material. We can check the material for hardness. We can see if the material was cracked when it was forged. Other than this, we can't really tell you much about why it failed, without knowing the stresses it was subjected to."
Trust me, any/every race engine that I have anything to do with will have the capability to gather data, from here on out....I felt like a complete idiot.
Piston speed, with a 3.750" stroke crankshaft reaches "critical" speed (and this comes straight from Chevy Racing, when they were running Trans-Am engines equipped with Titanium rods and super light pistons) at 7800. They used to limit these engines to rpms lower than this, to make them "live". I'd be happy if the engine never went over 7200-7400. Carrillo is fine, with this.
Pretty amazing, to me, for the incredible forces the pistons and rods were subjected to, the rod bearings (not Porsche bearings) looked like the day they were installed. No dimensional changes. No sign of pounding. No loss of crush. The rod bearings and oil definitely did their job, very well.
Carrillo did a design study of the connecting rods. They also cut apart the connecting rod with the crack down the middle. "The connecting rod failed from fatigue." I asked what that meant, on a rod with less than 12 total hours on it. "It wasn't made wrong, the material is perfect, the crack wasn't in the forging. There is nothing wrong with the machining. It broke from fatigue." I asked if they had any recommendations for improving the strength of the rod. "There's nothing wrong with the design. We have no recommendations to change the connecting rod. The rod is actually "overbuilt" by a significant amount." I asked if we could make the "pin end" stronger, since we have had two failures in this area. They said that they certainly could, "However, this was not the problem."
I insisted.
The "new rod" for this engine has a slightly thicker strap, suitable for use in a high rpm diesel engine, according to Carrillo.
Rotating weight went up 3 grams. Reciprocating weight went up 16 grams.
Makes sense then. The weight increase is very small in any case. Should work better than the old piston assemblies for sure.
Thinking in terms of theory, very high rpms would require a very light piston assembly. A light piston assembly, in turn, needs oil spray cooling. In 928, however, trying to run very high rpms with piston oil squirters gets one into oiling problems, mainly because crankcase gas flows (pistons pump a lot of gas everywhere). To mitigate that effect of squirters, one would need at minimum a crankcase vacuum and while one pulls that one might as well go dry sump in a race car engine. The changes that are needed for a 928 to run very high rpms are therefore pretty extensive.
With a wet sump and atmospheric crank case, piston oil squirters don't seem to play well with high rpms in the 928 crankcase. Without piston oil cooling, the piston can't be very light. It needs mass to hold the heat. This in turn makes very high rpms hard on the rods and crank. My opinion then is that in a wet sump 928 without piston cooling oil squirters one shouldn't even shoot for very high rpms. Instead, one should try to make the power at high but not very high rpms.
Just my theory.
Originally Posted by GregBBRD
No plans for squirters. This particular engine has always been wet sumped. I'm going to try and talk the client into a dry sump, this time around.
RPM's are something we need to look at, confirm, and record in the future. Neither engine got run extremely high in the rpm range under acceleration, according to the driver....and I trust what he says. No idea what happened on downshifts...the tachs were certainly not accurate or fast enough to be of any use, in a downshifting situation. (The tachs in both of the cars that had engine issues were stock 928 tachs out of junk cars....unknown if they were ever checked for accuracy.) One of the first things Carrillo asked for (both times) was a copy of the rpm data. Neither engine had any data gathering capability, so the following discussions were a bit embarrassing.
Carrillo: "A $3,000 go cart has data being recorded, these days. Your customers are running mega dollar Porsche race engines and they can't afford something that records data?"
Carrillo: "You know that generally means they don't want the data to be recorded, right?"
Carrillo: "You seriously want us to tell you something about failed connecting rods from a race engine with zero RPM data?"
Carrillo: "We can look at the material. We can check the material for hardness. We can see if the material was cracked when it was forged. Other than this, we can't really tell you much about why it failed, without knowing the stresses it was subjected to."
Trust me, any/every race engine that I have anything to do with will have the capability to gather data, from here on out....I felt like a complete idiot.
Piston speed, with a 3.750" stroke crankshaft reaches "critical" speed (and this comes straight from Chevy Racing, when they were running Trans-Am engines equipped with Titanium rods and super light pistons) at 7800. They used to limit these engines to rpms lower than this, to make them "live". I'd be happy if the engine never went over 7200-7400. Carrillo is fine, with this.
Pretty amazing, to me, for the incredible forces the pistons and rods were subjected to, the rod bearings (not Porsche bearings) looked like the day they were installed. No dimensional changes. No sign of pounding. No loss of crush. The rod bearings and oil definitely did their job, very well.
Carrillo did a design study of the connecting rods. They also cut apart the connecting rod with the crack down the middle. "The connecting rod failed from fatigue." I asked what that meant, on a rod with less than 12 total hours on it. "It wasn't made wrong, the material is perfect, the crack wasn't in the forging. There is nothing wrong with the machining. It broke from fatigue." I asked if they had any recommendations for improving the strength of the rod. "There's nothing wrong with the design. We have no recommendations to change the connecting rod. The rod is actually "overbuilt" by a significant amount." I asked if we could make the "pin end" stronger, since we have had two failures in this area. They said that they certainly could, "However, this was not the problem."
I insisted.
The "new rod" for this engine has a slightly thicker strap, suitable for use in a high rpm diesel engine, according to Carrillo.
Rotating weight went up 3 grams. Reciprocating weight went up 16 grams.
Makes sense then. The weight increase is very small in any case. Should work better than the old piston assemblies for sure.
Thinking in terms of theory, very high rpms would require a very light piston assembly. A light piston assembly, in turn, needs oil spray cooling. In 928, however, trying to run very high rpms with piston oil squirters gets one into oiling problems, mainly because crankcase gas flows (pistons pump a lot of gas everywhere). To mitigate that effect of squirters, one would need at minimum a crankcase vacuum and while one pulls that one might as well go dry sump in a race car engine. The changes that are needed for a 928 to run very high rpms are therefore pretty extensive.
With a wet sump and atmospheric crank case, piston oil squirters don't seem to play well with high rpms in the 928 crankcase. Without piston oil cooling, the piston can't be very light. It needs mass to hold the heat. This in turn makes very high rpms hard on the rods and crank. My opinion then is that in a wet sump 928 without piston cooling oil squirters one shouldn't even shoot for very high rpms. Instead, one should try to make the power at high but not very high rpms.
Just my theory.
Unfortunately, "I don't know" is the best responce I can make.
Looking at the amount of "cooked" oil that was deposited on the bottom of the pistons, in the latest failed engine, there can be little doubt that the pistons were extremely hot. And if just the pistons had failed, perhaps taking away more heat would have helped, some. The fact that rods failed makes me think that no amount of heat removal would have "saved" this engine.
Perhaps worth noting, the amount of "cooked" oil deposited on the bottom of the pistons varies quite a bit, indicating higher piston temperatures (which agrees with which pistons were damaged the most), from one cylinder to the next. The intake manifolds, on both of these engines, were never tested for individual air flow....so individual cylinder mixtures were/still are a complete unknown. (The manifolds were never more than a sales mock-up, as far as I can tell.) Therefore, It's not surprising that some cylinders would have been hotter than other cylinders. However, all 8 pistons have cracks in the tops...indicating severe pounding.
The damage to the connecting rods is not from heat....the wrist pins are not "blue", but a very light straw color. And the surviving rods have no oil residue "cooked" onto them.
Additionally, none of the pistons actually melted. There were signs of deformation from heat on a couple of the pistons and some "pitting" in the top surface of some of the pistions, but no "holes" from actual melting.
My conclusion was that the "pounding" from detonation was the direct cause of failure, not actual heat.
A couple of "real life" events made me question the "need" for piston oil squirters:
Once I directly observed the huge volume of oil that is present in the oil filler neck area of the block, when a 928 engine is running at 6,000+ rpms, I really questioned if piston oil squirters (adding more oil upwards) were necessary...or even desirable.
Additionally, whenever we ran Mark Anderson's dry dumped 928 engine over 6,000 rpms (under load for extended amounts of time), his dry sump tank volume would drop by several inches, indicating large amounts of oil being held in suspension inside the engine.
More study is necessary before I can draw any conclusions.
05-20-2019, 04:40 PM
New shiny parts
