Re: O\'l silvers tear down
To RiceEater:
A connecting rod (conrod) is a structural column supported on both ends by hinged joints that can not support a significant torque. Therefore, the load is ideally a pure tension or compression load with the forces applied at the hinge supports (the wrist pin and bearing) It transmits a load from one end to the other. The load must be applied along the centerline (longitudinal axis) of the column for maximum strength. When a column is overloaded, it buckles. The maximum load an undeformed column can handle is many times that of a buckled column. Or, a buckled column can handle only a small fraction of the load of the undeformed column. If the load on a column gets close to the maximum a column can bear, the column can catastrophically fail, it will completely collapse. This is why column-loaded members are designed with huge factors of safety, or margin. Not just a few percent of the maximum possible load, but multiples, like 20x, 30x, etc. But connecting rods are not pure columns. Unfortunately, this is not an ideal situation, and non-axial loads are present, which screws things up. The conrods can also be bent from these non-axial loads, which is very different from buckling.
Head gasket issue:
I suppose that head gaskets are not supposed to fail anymore? Why does that make me think of "permanently lubricated bearings" that aren't, semi-boneless ham that is and isn't? How many times have you had to replace those things that are supposed to never need replacement again? Silly me...
To tdimeister:
From your post: Boundless, with all due respect it seems obvious to me with your last post that you're no mechanical engineer but posing as one. I challenge you to describe to me: a) how you figure tensile overload caused the failure of the con-rods that we see; and b) how a tensile force of that magnitude could have been generated, as you said correctly, during the intake stroke alone.
Here is the answer to a)
From my earlier post: From the photos, the connecting rod pulled the wrist pin so forcefully that the wrist pin bosses of the piston were torn right off the piston, a classic tensile failure mode.
In simple English this says, "that the wrist pin bosses of the piston were torn right off the piston, a classic tensile failure mode." Do you see the word 'conrod' here? It clearly states that I was addressing the wrist pin bosses of the piston, no mention of conrod. Therefore, there is no need to explain tensile failure of conrods since I wasn't talking about tensile failure of conrods. You used your Mulligan. Reading comprehension issues won't be dealt with so politely anymore.
Here is the answer to b)
The photo clearly shows that the wrist pin pulled the bosses off the piston. The piston is not seized or binding in the cylinder at all, according to the reports. I maintain that this occurred on the intake stroke when the piston is pulled down the cylinder by the wrist pin. The failure may have occurred towards the end of the exhaust stroke when the deceleration of the piston was high, resulting in high wrist pin forces that cause tensile loads in the wrist pin bosses of the piston. But in that case, the piston is usually slammed into the head and there was no damage to the piston top or the head, so that couldn't have happened. So, the only other way it could have happened was when the wrist pin pulled the piston down some time after the beginning of the intake stroke. The piston mass and some friction between the rings/piston/cylinder caused a force in one direction (up) and the wrist pin exerted the force in the other direction (down). These forces were directed away from each other resulting in a tensile load on the wrist pin piston bosses. This pulled the wrist pin piston bosses apart. These excessive forces are generated by over revs. The higher the RPMs, the higher the acceleration/deceleration forces generated by the mass of the piston. (F=ma) That should be enough. If it isn't, go ask someone else.
So now you know, or should I ask if I am correct?
Okay, I used the easy equation instead of the hard one. I wanted to avoid another piece of information that further shoots down this over press scenario. But, in demonstrating your alleged prowess, you opened another door that you need to close. You calculate that with 35 PSI boost, the pressure at the end of the compression stroke will be 2890 psi. The injection pump is rated at 200 bar or 2900 PSI. The instant combustion starts, the pressure in the cylinder will spike and easily exceed the pressure capability of the pump. How can a 2900 PSI pump pump into a cylinder that is at a pressure greater than the 2900 PSI pump rating capability? If the pressure in the cylinder gets above the pressure capability of the pump, the pump can't pump into the cylinder. Don't even think of trying to say that the pump shoots all the fuel in there real fast and then the fuel burns and raises the pressure. That's not how it works. I'm looking forward to your answer great swami, especially since your calculations show pressures several thousands of PSI greater than the 2890 PSI, all well in excess of the pump pressure capability.
So you did some calculations that produced some pressure figures. They don't substantiate that "That, ladies and gentlemen, is what MANGLED those rods. I'm so sure of it! It's a classical failure mode!" Please explain how those pressures that you calculated mangled the rods. Exactly what is the classical failure mode? You didn't mention that.
It is very difficult to conclusively prove that over press & buckling caused the deformed rods with out actual pressures and conrod geometry. But, certain failure modes can be ruled out by the fact that certain things don't jibe with the hypothesis. I'm sure I have done that. Specifically, the bent rods could not have been deformed (buckled) by over press. It sounds like the over press was an isolated incident. Deforming a steel/iron rod that much so fast would have fractured the conrods. This magnitude of deformity took a long time to achieve, certainly not in the time the BOOM took place as described in an earlier post. The pump could not pump full loads of fuel into a cylinder that was over turbo'd. You guys have shown that in the event of a turbo spike, the cylinder pressures will be well over 2900 PSI, the max. pressure of the TDI pump. This can effectively attenuate or shut down the pump. Do you think that the pump pressure is designed to compensate for turbo spikes, excessive cylinder pressures, or over fueling to protect the engine? This eliminates over press as the primary cause of the deformed rods. Prove that the turbo over press alone (with fat injectors) can cause cylinder pressures that are egregiously (>>30x) in excess of normal operating turbo & cylinder pressures with normal or even 90 hp fuel loads. You gotta prove this while supporting the over press hypothesis. Remember, cylinder pressure much above 2900 PSI effectively attenuates and may even shut down the pump.
See if you can do it credibly and professionally without attacking someone personally to direct attention away from your own weakness and inability. If you really know what you are talking about, you shouldn't be confrontational, threatened or defensive. This isn't a competition, unless you want it to be. Does, "I challenge you to describe to me: a) " look familiar? It's just a place where people can ask for help and those that are nice enough can politely offer it. If you want to behave like that, go to work.
So, tdimeister, tell me about yourself. From the way you present yourself, you must be a post-post-post doc, forgive me please if I underestimated, and your tact and people skills must make you a Sr. VP of a very large & high powered engineering organization, again, please forgive me if I have underestimated. Tell us why we should blindly believe all the unsubstantiated speculative things you have to say simply because you say so.
From your post, you used a freshman thermo equation, readily available from most reference sources, to calculate cylinder pressures. You failed to connect those calculated pressures to the mangled rods. Here is the relevant passage from your post:
from tdimeister: "Do the same with 35 PSI of boost, that's 35 ON TOP of 14.7 PSI, and applying my magic equation again, and I get a figure of 2890 PSI!! Throw fuel into the equation and let it burn, and the peak pressure rises to SEVERAL THOUSAND PSI OVER my original example! That, ladies and gentlemen, is what MANGLED those rods. I'm so sure of it! It's a classical failure mode!"
Right now all you have is unsupported speculation that has a gazillion holes in it. Your pressure calcs did nothing. There is no basis from which to say the pressures you calculated are excessive and caused the conrods to deform. None whatsoever. This isn't a competition, it's about sharing and learning and some of us have a lot more to learn than others. You are quite entertaining although you really aren't saying anything of value. You make it look like you know what you are talking about, but: Where's the beef?
So tell us great swami why you are so sure of it, even though a sieve has less holes than your conjecture, and YOU LIKE TO YELL AND USE A LOT OF EXCLAMATION MARKS TO ATTEMPT TO GET YOUR UNSUBSTANTIATED POINT ACROSS!!!!!!!
To DavinATL:
I like your post. I admit, I was lazy and didn't go into detail. If I did, I'm sure you would have gotten the points. I'll explain the things you didn't agree with in more detail. But, I have plenty experience in this area. I have spent many years of my career conducting failures analyses in the aerospace & automotive industries, mainly concentrating on engines or engine equipment. I also am involved in motorsports and deal with team owners and engine builders that participate in nationally sanctioned racing organizations. I've seen plenty of bent and broken parts along with great details regarding the surrounding circumstances. Also, I did quite a bit of research in combustion dynamics. I'm not one of the armchair guys. Good luck and congrats on the phd. When can I call you Dr.?
First, let's agree that the failure of the #2 piston is probably due in a large part, to the deformation of the rod. The rod problem allowed the piston wrist pin boss to be compromised, leading to its demise. The weakening of the boss reduced the max. RPM that the assembly could tolerate.
Also, I really don't care about the last gasp runaway situation. The damage that could be valuable to understand is the deformation of the rods which obviously happened long before the engine died.
So, let's start with a further explanation of column loading, buckling of columns, design margins, over pressure, and fat (.205) injectors. See the above discussion about columns for RiceEater. For example, let's say we have a column that can support a maximum load of 100#. If the load exceeds 100#, the column will buckle and collapse unless the load is removed or dramatically reduced. The load required to continue the collapse of the buckled column is just a tiny fraction of what was required to initiate buckling. So if it took 100# to initiate buckling, it might take only 30# to continue the failure collapse. Note: I'm talking load and not prescribed axial deflection. (BTW: I was a teaching assistant and ran Strength of Materials labs. I conducted a gazillion column buckling tests on well instrumented samples and equipment.) I maintain that if an over press condition caused the conrods to buckle, there would have been a complete catastrophic failure because once the buckling was initiated, the over press load was not attenuated and therefore far in excess of what would be necessary to destroy a buckled conrod. This didn't happen, so I maintain an over press did not cause this failure - bent conrods or eviscerated piston. Now let's look at how you say the over press occurred. You claim that the fat injectors can pump that much more fuel into the engine and that with the over press spike went boom. Let's look at a stock TDI motor. At maximum output, (90 hp), the fuel flow consumption rate is X #/hr. If you double the fuel flow rate (2X) and assume it could be burned effectively, the engine would have an output of 180 hp. Now, redline is limited to the same max. RPM, so this increased output came from an increase of only torque. (I believe you said ol' Silver never went above 3,500 RPM) (HP = Torque x RPM) This means that cylinder pressures were doubled for the same RPMs. Now, I don't think that the .205 injectors could double the fuel flow rate over the stock injectors and increase the HP that much. The actual increased fuel flow rate is some percentage significantly less than 2X the stock injectors. But let's say the fuel flow was doubled and the air flow was adequate to burn all that fuel. The pressures on the pistons would be doubled, the torque would be doubled and all would be okay as long as we didn't let that extra torque make us zing (over rev) the motor. Even double the output won't bend conrods. These structures (columns = conrods) require huge factors of safety (margins) because of the potential catastrophic results of a failure. Therefore, these columns are designed with maximum loads (Pre-buckle) 20x, 30x, 40x, 50x, or greater than the maximum possible actual load just to make sure the possibility of initiating buckling is minimized. And that is in a static situation. This is a dynamic situation where there is a lateral load and a moment applied to the "hinged supports" that can uncouple the load line (line of action) from the column centerline causing instability that could lead to buckling. Design margins in cases like this are huge. I would have to believe that in order for a TDI engine to live the hundreds of thousands of miles it is expected to under all the insane conditions it will be exposed to over probably 15 years, the buckling margins the conrods are designed to are huge, 40x at least, which means that even if the .205 injectors delivered twice the fuel flow rate of the stock injector and the turbo over boosted and gave it all the air it needed, the pressures still would be far, far below those required to initiate buckling. Look at this way, if the engines that get these .205 injectors all smoke so much as you have said, the engines are saturated and are not producing all the power from the fuel being injected. The engines are not able to fully produce the power from the available fuel (Proven by the smoking) and the additional fuel flow rate is what percentage greater than stock? It can't even be 2X stock. So, in reality, the usable fuel flow isn't even twice stock, (The .205 injectors are only 22% greater flow area than stock.) the torque can't be doubled, the load on the pistons can't even be doubled, the buckling stability design margin on the conrod is probably at least 40x, there is no way over press & fat injector fuel flow rate could have come close to initiating a conrod buckling load.
Boundless: The only force opposing the wrist pin was the mass of the piston.
DavinATL: Aha. Are you sure? Where did that scuff mark come from on the "good" piston photo next to the crack?
Yes I'm sure. Well, add in a little friction. The only force to pull on the bottom of the piston like that is from the wrist pin from the conrod. The only thing that could oppose this (equal & opposite) is the mass of the piston (F=ma). Turbo pressure/intake pressure is obviously negligible. My understanding is that these scuff marks are from the piston contacting the counterweights on the crank as a result of the shortened conrods, which all were shorter, regardless of how "good" the pistons were. Look at where the scuff is on the piston wrist pin boss. The scuffs are on the very bottom of the boss, at least the piston I saw, which is below the wrist pin. This scuffed area would be between the wrist pin and the crank counter weight when the piston is at BDC. The loads from the scuffing and the failure loads that yanked the boss off the piston act on different areas. If the scuffing was severe enough, it could have caused weakening of the boss and allowed the boss to open or rupture on the intake stroke, allowing the wrist pin to be pulled out of the piston. If the crankshaft counter weights were severely hitting the piston skirts, I could accept that as the reason for yanking the wrist pin bosses off the piston like that. It looks like the skirts just chipped away. Also, I believe that the piston that was eviscerated was still up high in the cylinder, away from the counterweights. The wrist pin pulled off the bosses and the piston stayed behind.
Boundless: Hmmmm.... I still don't see it. The fact that the engine survived for quite some time with bent conrods is also further proof that turbo over pressure could not be a causal factor in this failure. If over press had the effect as presented in some of these postings, it would have destroyed all the deformed conrods instantly.
DavinATL: I disagree.
A bent conrod (such as those in the pics) has a maximum load capability that is far, far below (maybe 25%) that of a straight conrod. But the engine still ran without collapsing the rods. The severely weakened rods still worked for many, many miles. (There is a post that indicates the car was driven for 10 hours at 80 MPH @ 3500 RPM up & down hills after the BOOM) Doesn't that indicate to you how huge the margin is to avoid buckling in the first place? And how resistant to buckling the conrods really are? Your argument implies that the deformation we see in the pics occurred all at once - the BOOM you refer to. (BTW - from other postings, it is questionable as to how many times or when turbo spikes occurred.) Well, if you bend a steel/iron rod of that size and as fast as it must have been deformed to end up the way it did, the rods would have fractured because the strain rate would have been grossly exceeded. Have you ever bent a steel/iron rod and if you bent it too fast it broke, but if you went slowly, it didn't break. The same here. Those rods can't be bent like that in the time it took to go BOOM. Those bends took place over a long time, many days, weeks or even months, little by little. Also, when an engine hydrolocks and crushes conrods, the bearings (wrist pin & conrod) are also crushed in very obvious and apparent ways. Now if the hypothesis is that this engine buckled the conrods due to over pressure (similar to hydrolock), the bearings should also be crushed, but they weren't. Also, the wrist pins and wrist pins bosses of the pistons should show signs of severe damage, but they don't. Except for the scuffing & chipped skirts, the pistons are reported to be in excellent shape. In piston & rod assemblies, the piston is the weaker. In this alleged over press, the conrod is the only component that is damaged and all others of the conrod/piston assembly are unscathed. That further casts doubt on an over press. So if the over press occurred, the pressure was transferred through the piston - bushings/bearing - wrist pin - conrod - bearing -crank, and the conrod is the only component to experience directly attributable over press damage. In hydrolock situations, the piston is often the fuse, it fails first, then bearings. I find it very unusual that an aluminum piston was able to survive an alleged over press unscathed while transmitting a force that beat up a steel conrod.
DavinATL: Steve by his own admission (and that of others) was not in the habit of running this engine at high RPMs, and on that I believe him. I don't think that this engine saw RPMs over 3500 until that overrun.
You mean to tell me that this car that has a factory redline of 4500 RPM, and is heavily modified with just about every after market hop up item never went within 1000 RPM of the factory redline? Oh what was I thinking.... silly me!!!! There is also no evidence of lugging.
Boundless: More boost in a diesel could give nothing more than more excess air to more completely burn the injected fuel.
DavinATL: As I said above... there was a LOT of injected fuel from those fat hole injectors that wouldn't have burned at stock air masses. The spike allowed it all to burn.
How much more fuel than stock? 2x?.... Even if it all went off, it couldn't come close to the stability design margin load of the conrod.
Boundless: All this failure mode needs is too much power setting & RPMs way too often. These engines weren't meant for that kind of duty cycle. It's that simple.
DavinATL: But that's the problem. Steve did NOT abuse this engine (well, didn't abuse it by overreving it ). He was light on the foot and kept the RPMs low.
Let's see, ol' Silver had every performance mod known for no reason, and the engine would never be allowed get within 1000 RPM of the factory redline.
Of course, silly me for even thinking otherwise.
Tell me the truth, the whole truth, and nothing but the truth.
Davin, I need something more to believe the over press happened. There are too many holes in that hypothesis, that are supported by the engine condition that ultimately eliminates the over press as the cause of initiating the rod deformation. I'm looking forward to reading what you are thinking. Unfortunately, all we can do is punch holes in the various theories.
