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.
Greg, calling me names isnt going to end well for you as well. so, lets just make this respectful. I actually like you, and have appreciated all your help over the years, so lets keep this on topic.
Ok, lets go back to the beginning for a sec. Mark's rods were NOT chevy offset, they were 928 offset rods that he bought from someone for a 928 project. he had a crank made to fit the stock 968 pistons and those rods. as you said. (i cant go back and find your quote) but you said, anyone that uses a chevy offset rod in a 928 is looking for a HUGE problem.
are you disagreeing with me (forget that i said lying... we are trying to reconstruct history here from 20 years ago), then you should know this is what Mark told me back then about his piston (968), rods... custom 928 offset, and crank.. moldex , made to his stroke change to make it all work in a stock 928 block.
( i get the entire story about the damaged crank of mark's after road america's intake failure and your new crank and A rods, that were pulled due to the other "failure" you had due to those rods, and replacement to the ones in question (the tapered H rods) .
now, again, if you have a chance to take the picture again of a piston and rod combo that is the same as the JOE FAN piston.rod picture i posted, do it with the rod off center 0.060 as i did and you will see it will be identical to the picture i posted. your "perspective" pictures were not only from more of the side, but angled in orientation as well. it doesnt take a college graduated engineer to see that ether.
again, i measured a piston with a total available space of 0.260" , 0.130" on each side if centered. if it was off the 0.060" i see in the picture of the joe fan piston in the engine, then that could be a problem, no?
if it is perspective, then sure, it would not be an issue. BUT, again using the same "perspective" i was able to replicate what i see and it looks like there is too much offset, even if you take in to account movement at the big end of the rod on the crank.
Let me attach your pic and then the pic again to see why you are a little off here.
the other thing is that in both failures, the piston , if damaged by detonation would have stopped in the bore, causing the crank to pull the rod apart at the strap or pull the pin out of the piston, as it rotated at high speed and force.
Originally Posted by GregBBRD
Mark:
Like I said, ignore until you say something completely stupid, again.
From the above statement, I know you have no idea how much .060" looks like, in your head.
.060" is the thickness of an American quarter!
When the rod is off .060", the "short" side dimension will actually be touching the piston pin boss!!!!!
Every single engine that I built for Mark Anderson, before his last engine, used "Chevy offset" connecting rods....supplied by Mark Anderson. On every single previous engine, I had to machine the inside of the piston boss, on the "short side", so the connecting rod did not rub on the piston boss!
The picture you keep referring to, isn't anywhere near .060" off! It simply off from the perspective of where the picture was taken from.
Greg, in order to show "perspective" you must use the same angles in comparison and you didnt. it is very obvious from the pictures below to see this.
1. my picture of a mock up of 060" off center
2. the actual picture of the piston in the block (look the same?)
3, my picture of .020 off center
4 . your picture of .020 off cemter.
5. your picture of the piston centered but twisted with a side angle to show "Perspective".
A couple of notes, in blue
One would think that the impact of downshifts on the rod and piston assembly would be almost completely be captured by the rpm. If the rpms don't match, then I am sure that it's hard on clutch, transmission, and the rest of the driveline. But I don't see an obvious physics based reason why anything but the rpm would really matter to the rod and piston assembly in that situations. Either revs go above the redline or they don't. There's more to it, than just rpms. Inertia and change in direction become a major factor. Pistons and rods, in normal operation, are "pushing" on the crankshaft. On a downshift with a high rpm delta (with long splits between gear ratios, like in a 928) the pistons are no longer driving the crankshaft. The crankshaft is being forced to slow down very quickly, by the gear ratios, which "pulls" on the connecting rods and pistons much harder. Think of it this way: The piston and rod normally have inertia in one direction (up), which is stopped and forced the other way (down) by the combustion process, in normal engine acceleration. If the crankshaft suddenly slows (from a non rpm matched downshift), the piston and rod have the exact same inertia (going up), but the crankshaft is instantaneously slowing....which pulls back on the rod and the piston.....radically. This almost doubles the forces on the connecting rod and the piston.
Greg, that really isnt how it works (or how physics works) when you dont match engine revs, there is NO force that slows the pistons down, there is a reduction of force. this slows the engine crank down, it intern slows the pistons and rods down as well. then, normally , a force that acts on the piston and then the rod to the crank , accelerates the speed of the crank, will then have a force originating on the crank pushng the piston UP and pulling it down as the clutch is released on a mis matched RPM downshift.. the wheels , not the combustion (through the transmission and driveline, clutch, and crank) will drive the rods and pistons faster.
the accelerative force can be greater on the engine, than what the engine can develop on its own due to combustion. So, any forces on the engine that were near the limits under combustion at any RPM, are goig to be greater. your 500hp engine, suddenly sees (example) 1000hp forces for the downshift that is not rev matched if the WHEELS drive the engine to the matched RPM. make sense??
High intensity knock causes compression load, and by my understanding that's not how the rods failed here. They didn't buckle. So to the extent that the knock is relevant, it's because it has pounded the bearing and caused a bearing failure before the rod failed. The bearings for all the cylinders, not just the one with a failed rod, might be informative. Perhaps an incorrect conclusion. Actually, there is way too much of the connecting rod missing....to know the rod did not buckle. (The entire "beams' are gone.) And the piece remaining on the crankshaft is badly twisted, presumably from hitting other engine parts. The particular rod bearing I use is extremely robust....and as long as there is an oil film between it and the crankshaft, the rod could certainly buckle and the rod bearing could be unharmed.
Greg, if there was knocking, you would see damage in the rods and there wasnt any . the rod had to of broken AFTER the piston pin was pulled out of the piston. there would be nothing to pull the piston pin out of the piston, if the rod was broken in half... the rod most likely broke after it was disconnected from the piston by flailing around in the block. .
Originally Posted by ptuomov
“BUT, if overrev forces were the issue, wouldnt it break the smallest part of the rod,not the FATTEST/strongest part of the rod???”
Kibort — Why do you think the fat part of the rod is fat and the skinny part of the rod is skinny? Think about how the weight of upstream beam contributes to the force.
it's mostly the same thickness but goes wide to at the bottom to fit the journal, and you cant have it fatter, or it will hit the side of the piston as the piston goes down the H rods are pretty much the same thickness at the top as near the bottom. tappered H a little less at the top vs the bottom
Originally Posted by ptuomov
Kibort - The acceleration of the piston is a different animal from the acceleration of the engine rpm.
yes, i agree. and maybe you are right... But think about 0 to 5000rpm in 0.2 seconds (on a stalled engine, clutch release at 50mph in 2nd gear) vs the forces at 5000rpm at cruise... which has higher forces on the connecting rods? are they the same? I have to think about that one for a sec.
"There's more to it, than just rpms. Inertia and change in direction become a major factor. Pistons and rods, in normal operation, are "pushing" on the crankshaft. On a downshift with a high rpm delta (with long splits between gear ratios, like in a 928) the pistons are no longer driving the crankshaft. The crankshaft is being forced to slow down very quickly, by the gear ratios, which "pulls" on the connecting rods and pistons much harder. Think of it this way:The piston and rod normally have inertia in one direction (up), which is stopped and forced the other way (down) by the combustion process, in normal engine acceleration. If the crankshaft suddenly slows (from a non rpm matched downshift), the piston and rod have the exact same inertia (going up), but the crankshaft is instantaneously slowing....which pulls back on the rod and the piston.....radically. This almost doubles the forces on the connecting rod and the piston."
The piston acceleration equation has two terns. The first is proportional to the engine speed (radians per second) squared and the second is proportional to engine acceleration (radians per second squared). Quantitatively, when the engine speed is high, the second term relating to engine acceleration is unimportant. At 1000 rpm and comparing an engine with extreme acceleration vs. engine which isn't accelerating, the plot of piston acceleration against crank angle does show a visible difference between the two cases. However, at high engine speeds such as 7000 rpm, the impact of the engine acceleration on piston acceleration is just a rounding error -- it's so small that it's safe to ignore. Therefore, for connecting rod failures, one just needs to know the engine rpm and the engine acceleration doesn't matter.
I don't think there's any practically relevant way to get the piston acceleration forces to double due to engine accelerations, even at idle (when the forces are uninterestingly small anyway).
I couldn't find a free calculator on the web that would show this, but this thesis has a page (68) on it:
The typical engine data of Table 4.1 has been used to demonstrate the effects of engine acceleration and crankshaft mechanism offsets. The piston acceleration was found to be affected on some conditions by the crankshaft acceleration which depends on crankshaft mechanism parameters and acceleration amplitude. Although affected, the error is quite minimal on engines that do not have immense accelerations. A typical value for acceleration has been used from a 2-stroke V6 Mercury marine engine. These engines can accelerate from idle to maximum (around 10000 RPM) in a fraction of a second which was evaluated to be in the order of 25000 to 30000 rad/sec2; thus this engine’s acceleration can be used in an example to illustrate the effect of a large crankshaft acceleration. The piston acceleration is affected only at low engine speeds by the crankshaft acceleration due to the fact that the velocity is the main source of the piston acceleration.
\it's mostly the same thickness but goes wide to at the bottom to fit the journal, and you cant have it fatter, or it will hit the side of the piston as the piston goes down the H rods are pretty much the same thickness at the top as near the bottom. tappered H a little less at the top vs the bottom
Take a look inside the machined slot of the H beam and tell me if you still think that the cross-sectional area doesn't increase as you move from the small end to big end.
Take a look inside the machined slot of the H beam and tell me if you still think that the cross-sectional area doesn't increase as you move from the small end to big end.
i absolutely agree. but it does look fairly constant at near the end going toward the bottom. then, it widens quite a bit. however , you would think that the most stress is near the end, no? they cant make it thicker, or it wouldnt clear the piston sides, right? Plus they try and get as much weight out of the piston side end as possible. im not sure im getting your point. mine was, if the piston stops in the bore as it looks like it did on this engine blow up, as well as possibly mark's engine before, why wouldnt the small end be the most vulnerable? either pulling the pin out of the piston or breaking the strap or even breaking near the small end. breaking.
pics are the H rod, the tappered H and a comparison of the two, plus the broken rod
forces caused by collision of a freewheeling rod to the bock. near the bottom , where it is fat, seems to be caused by the tremendous forces of it hitting the block .
in thinking about it, you are right. I hope greg leans a little physics here to help with his theories.
However, i do think with rapid increases of crank speed (as you mention , piston accelerations are NOT effected to a discernible amount) the accelerative force can effect the crank and the block in ways that might put greater stress on the all things that hold them in place. i guess you can think of the pistions travel as analogous to an rotational, inertial load. the centrifugal force dosnt change at different rates of acceleation, only by the angular velocity. what does change is the force on what is ever accelerating the rotating mass. (i.e. the crank and components being driven by the wheels/transmission and short shaft, in a 'non-rev -matched -downshift)
Originally Posted by ptuomov
The piston acceleration equation has two terns. The first is proportional to the engine speed (radians per second) squared and the second is proportional to engine acceleration (radians per second squared). Quantitatively, when the engine speed is high, the second term relating to engine acceleration is unimportant. At 1000 rpm and comparing an engine with extreme acceleration vs. engine which isn't accelerating, the plot of piston acceleration against crank angle does show a visible difference between the two cases. However, at high engine speeds such as 7000 rpm, the impact of the engine acceleration on piston acceleration is just a rounding error -- it's so small that it's safe to ignore. Therefore, for connecting rod failures, one just needs to know the engine rpm and the engine acceleration doesn't matter.
I don't think there's any practically relevant way to get the piston acceleration forces to double due to engine accelerations, even at idle (when the forces are uninterestingly small anyway).
I couldn't find a free calculator on the web that would show this, but this thesis has a page (68) on it:
The Chevy offset issue has been mentioned here. Can anyone tell me the C-C between a pair of cylinders for the Chevy small block? It probably is .920" or 23,4mm but I like to have it confirmed. For the 928 block the cylinder C-C is 25,0mm.
Åke
I spent a little bit of time thinking about this stroker issue. The 968 has 1.75mm wide top ring, right? And 968 has a 88mm stroke. I'm curious how Porsche got the 1.75mm top ring to work with 88mm stroke and 6200 rpm peak power engine speed? My little spreadsheet says that those thick rings would be at risk of flutter already at slightly over 6000 rpm. The same spreadsheet works pretty well across other engines, putting the peak power rpm at about 90-95% of the flutter critical rpm. The same spreadsheet says that a 95.25mm stroker that makes peak power at 6500 rpm and shouldn't go to ring flutter before 7000 rpm should probably have something like 1.2mm top ring width.
Model,GT,GTS,968
ring width (mm),1.50,1.75,1.75
stroke (mm),78.90,85.90,88.00
stroke/ring width,52.6,49.1,50.3
sqrt of ratio,7.25,7.01,7.09
flutter critical mps (m/s),18.1,17.5,17.7
flutter critical engine speed (rpm),6894,6117,6044
peak power engine spreed (rpm),6200,5700,6200
rpm ratio,90%,93%,103%
Well, the engine is out and almost completely apart....I didn't bother removing all of the pistons and rods.
Long story short....every piston is cracked across the domes, and all through the pin bosses. #2 piston and rod were going to be the next cylinder to fail. The piston is so distorted that it sticks .030" out of the hole, instead of being flush with the top of the cylinder (or slightly down in the bore or out of the bore, depending on the engine configuration/design. #8 and #2 pistons are pitted on the tops. (Others may be pitted, I didn't pay very much attention to the other six pistons, other than noting the cracks in the tops.)
#2 connecting rod has a crack about 3/4" long right through the bottom of the pin bore and down the center of the rod.
I called Jerry Rouch at Carrillo/CP (these particular pistons were made by J&E and Jerry worked for them for many years) and started to tell him what happened and was going to ask him if he would have his engineers look at the piston/rod. Right after I said, "Every piston is cracked", he interrupted me and said "Turn the piston over and tell me what color the underside of the dome is." Well, the underside of all the domes look like someone spray painted them black....over the entire surface. I told Jerry this and he said "Well, you don't need to bother sending us the pieces, the engine was detonated....severely. The pistons almost reached the forging temperature. The fact that they cracked, instead of just melting, tells you that. The connecting rod is being "split down the middle" because of the severe hammering of detonation....the pin is literally trying to split the rod into two pieces. Truthfully, I'm shocked the pin didn't break." (It is quite possible that #8 might have broken the piston pin, BTW.)
Like I said, from day one.....this engine was not designed to run any combination of street 91 octane fuel. This was a race engine, with race engine compression, and was designed and built for 110 octane fuel, minimum.
End of story, here. I don't know exactly what it is going to cost to repair/rebuilt this engine, but I'm guessing the final bill would buy a truckload of the proper fuel....
Note: I'm going to post a more detailed analysis on my web site, with pictures....it will take me a week or so....I'm crazy busy.
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!
I've been studying pre-ignition a little bit because that's a big threat to my turbo engine. Pre-ignition is when the charge ignites before the spark fires. In contrast, knock is when the mixture is burning and the unburnt end gas ignites in an unstable fashion. One can have both pre-ignition and knock, that is, the charge may start burning before the spark fires and then the residual end gas ignites.
Knock causes extreme pressure waves, but those have a high frequency or in other words the pressure alternates very rapidly. The piston has a lot of inertia, so knock doesn't usually hurt connecting rods. It knocks out ring lands, though. Mahle piston book section 7.5.1 shows the results from 105 hour induced knock test. The piston crowns got pitted, and ring lands eroded. However, the piston temperatures (measured at the top ring groove) didn't really go up that dramatically. It run hotter, but not much hotter. Adding pressure relief groove above the top ring in the top land creates enough of a buffer in that study to limit pitting above that relief groove. All signs point to knock damaging the engine directly because of force, not because of temperature.
Contrast knock with pre-ignition. Pre-ignition causes its damage thru sustained pressure (300 bar!) and heat, and can lead to knock on top of that (superimposing the high-frequency pressure oscillation knock pattern on top of that). While Mahle could test knock for 105 hours in a running engine, pre-ignition can overheat the piston in tens of seconds. Furthermore, if the pre-ignition isn't accompanied by knock, the 928 S4 knock detection system can't detect it. But even if it could, if the mixture is igniting before the spark, delaying the spark doesn't do anything to stop pre-ignition. So it's the silent killer from the ignition system's perspective -- it can't be detected easily and even if it could be, stopping it requires things like closing the throttle, opening the wastegate, and/or spraying water into the intake port.
Since pre-ignition is the main threat to my turbo engine, I've been trying to figure out how to reduce the risk. Fuel octane is of course one thing, as low fuel octane leads to knock, knock leads to ring flutter and oil in the combustion chamber, and glowing oil droplets in the combustion chamber can cause pre-ignition. Low fuel octane and knock also breaks and overheats spark plugs, which then cause pre-ignition. Wrong heat range spark plugs cause pre-ignition, because too warm (for the application) spark plugs turn into glow plugs. So those are some things that I'm trying to avoid.
The mention of cracks in the pistons were interesting. Exceeding either temperature or pressure limits can initiate cracks. According to Chapter 6 of "Engine Failure Analysis", excess temperatures usually cause cracks in the piston crown while excess pressures cause cracking in the wrist pin bosses. Sounds like the engine failure in question had both kinds of cracks, which would to me suggest both excess temperature and excess pressure. Oil burned to the underside of the piston also indicates excess temperature. I'd bet it was pre-ignition, given the observations.
I've been studying pre-ignition a little bit because that's a big threat to my turbo engine. Pre-ignition is when the charge ignites before the spark fires. In contrast, knock is when the mixture is burning and the unburnt end gas ignites in an unstable fashion. One can have both pre-ignition and knock, that is, the charge may start burning before the spark fires and then the residual end gas ignites.
Knock causes extreme pressure waves, but those have a high frequency or in other words the pressure alternates very rapidly. The piston has a lot of inertia, so knock doesn't usually hurt connecting rods. It knocks out ring lands, though. Mahle piston book section 7.5.1 shows the results from 105 hour induced knock test. The piston crowns got pitted, and ring lands eroded. However, the piston temperatures (measured at the top ring groove) didn't really go up that dramatically. It run hotter, but not much hotter. Adding pressure relief groove above the top ring in the top land creates enough of a buffer in that study to limit pitting above that relief groove. All signs point to knock damaging the engine directly because of force, not because of temperature.
Contrast knock with pre-ignition. Pre-ignition causes its damage thru sustained pressure (300 bar!) and heat, and can lead to knock on top of that (superimposing the high-frequency pressure oscillation knock pattern on top of that). While Mahle could test knock for 105 hours in a running engine, pre-ignition can overheat the piston in tens of seconds. Furthermore, if the pre-ignition isn't accompanied by knock, the 928 S4 knock detection system can't detect it. But even if it could, if the mixture is igniting before the spark, delaying the spark doesn't do anything to stop pre-ignition. So it's the silent killer from the ignition system's perspective -- it can't be detected easily and even if it could be, stopping it requires things like closing the throttle, opening the wastegate, and/or spraying water into the intake port.
Since pre-ignition is the main threat to my turbo engine, I've been trying to figure out how to reduce the risk. Fuel octane is of course one thing, as low fuel octane leads to knock, knock leads to ring flutter and oil in the combustion chamber, and glowing oil droplets in the combustion chamber can cause pre-ignition. Low fuel octane and knock also breaks and overheats spark plugs, which then cause pre-ignition. Wrong heat range spark plugs cause pre-ignition, because too warm (for the application) spark plugs turn into glow plugs. So those are some things that I'm trying to avoid.
The mention of cracks in the pistons were interesting. Exceeding either temperature or pressure limits can initiate cracks. According to Chapter 6 of "Engine Failure Analysis", excess temperatures usually cause cracks in the piston crown while excess pressures cause cracking in the wrist pin bosses. Sounds like the engine failure in question had both kinds of cracks, which would to me suggest both excess temperature and excess pressure. Oil burned to the underside of the piston also indicates excess temperature. I'd bet it was pre-ignition, given the observations.
I, too, have read volumes about both knock and pre-ignition. It's really interesting stuff.....I never really had any idea of the forces involved.
While it is sometimes tough to tell the difference and (certainly both can occur), one thing is for certain, in this particular engine....it detonated.
Pre-igntion overheats things. Detonation is like a huge hammer striking the piston.
Overheating will certainly overheat/damage the piston, but it takes some serious hammering to "split" a Carrillo rod down the middle.
Well, the engine is out and almost completely apart....I didn't bother removing all of the pistons and rods.
Long story short....every piston is cracked across the domes, and all through the pin bosses. #2 piston and rod were going to be the next cylinder to fail. The piston is so distorted that it sticks .030" out of the hole, instead of being flush with the top of the cylinder (or slightly down in the bore or out of the bore, depending on the engine configuration/design. #8 and #2 pistons are pitted on the tops. (Others may be pitted, I didn't pay very much attention to the other six pistons, other than noting the cracks in the tops.)
#2 connecting rod has a crack about 3/4" long right through the bottom of the pin bore and down the center of the rod.
I called Jerry Rouch at Carrillo/CP (these particular pistons were made by J&E and Jerry worked for them for many years) and started to tell him what happened and was going to ask him if he would have his engineers look at the piston/rod. Right after I said, "Every piston is cracked", he interrupted me and said "Turn the piston over and tell me what color the underside of the dome is." Well, the underside of all the domes look like someone spray painted them black....over the entire surface. I told Jerry this and he said "Well, you don't need to bother sending us the pieces, the engine was detonated....severely. The pistons almost reached the forging temperature. The fact that they cracked, instead of just melting, tells you that. The connecting rod is being "split down the middle" because of the severe hammering of detonation....the pin is literally trying to split the rod into two pieces. Truthfully, I'm shocked the pin didn't break." (It is quite possible that #8 might have broken the piston pin, BTW.)
Like I said, from day one.....this engine was not designed to run any combination of street 91 octane fuel. This was a race engine, with race engine compression, and was designed and built for 110 octane fuel, minimum.
End of story, here. I don't know exactly what it is going to cost to repair/rebuilt this engine, but I'm guessing the final bill would buy a truckload of the proper fuel....
Note: I'm going to post a more detailed analysis on my web site, with pictures....it will take me a week or so....I'm crazy busy.
Well, that sounds like the reason.......amazing what seeing vs guessing can achieve. so, im just curious how that much damage can be done without hearing or feeling it running poorly, I mean , knocking sounds like there is a rattle can in your engine. you know, we both heard gregory's engine when it came down from Devek. was the exhaust too loud to hear it? i guess , there is no other reason, is there? so, the mixture of 110 and 91 was not a good enough average octane. that's about 100. I'm running 91, at 11:5 compression , but i also have 200less HP so im sure that's a factor too. so you are saying that engine required 110?
anyway, sorry to hear and i apologize if i was riding you hard about the piston offset. it just seemed unlikely that 100 octane could cause such issues..... sure a little denial mixed in for sure.. ..... by the way, then what killed Marks original engine? same thing? did you see any of his pistons showing cracking? and one other things. how can you have such bad detonation and not have the rod bearings be pounded out? ive heard that is the tell-tale sign.
thanks for the report. very expensive lessons learned here.
I think ill start mixing 100 and 91 to be sure im not pushing the knock envelope!
omg I just realized this was Todd motor, what a shame , so it was supposed to be run and tuned for race gas , dang I am so sorry Mark,
NOT todd's motor.. that is in my car and has 120 race hours on it and countless street miles!
all on 91 octane but down on HP vs Mark and Joe's engines from GB
06-24-2018, 04:17 AM
I just realized this was Todd motor, what a shame , so it was supposed to be run and tuned for race gas , dang I am so sorry Mark,
