Interesting day in the dyno.... (pic)

[RagingGrandpa]

http://www-personal.engin.umich.edu/~lovelljc/F4iGoBoom.jpg

Whoops

[Dan Deussen @ Weber Motor]

I'm no expert....but I dare to say JB-Weld won't patch that hole http://fsae.com/groupee_common/emoticons/icon_wink.gif .

This pic is not quite so spectacular. It shows how a badly timed cam can cause a valve to hit the piston -> a connection rod to bend -> the rod to come through the case.
http://cs.svsu.edu/~dsdeusse/BrokenEngine_small.JPG

[BeaverGuy]

Raging Grandpa,

Is that a ZX6R? It sure is some interesting carngage.

[BryanH]

If f-sae engines had feelings they would be friggin terrified of the sound of the dyno being turned on, with one of Charlies "new teamers" standing there grinning at them.....

[The guru of guru's]

Holey moley...thats a huge bitch!
You know what they say....thats racing!!! I suppose a few tubes of ultra copper is out of the question for that minor repair.

[Charlie]

It's an F4i, you can tell its not an F4 by the gold 8mm case bolts on the front; F4 has silver 6mm case bolts. Yes I am a sick man...

[-]

So who is the unlucky soul who gets to go find a replacement for that...

Can I ask what happened to get the motor to gernade like that?

[alfordda]

Charlie, you are not alone. I have seen to many F4/F4i's torn apart, as well.

[RagingGrandpa]

Suffice to say, be damn sure all the wrist pin clips are installed and fully seated before holding an engine at 10,000 rpm.

And I'm the poor bastard desperately seeking the privelage to shell out $1,200+ for a new engine.

[Denny Trimble]

Wrist pin clip failure... sounds familiar! (http://students.washington.edu/dennyt/fsae/thrown_rod_web/)

You weren't running wiseco pistons, were you?

[-]

So were you running a different set of pistons other then stock? Also are you using new engines? Usually used f4i motors can be picked up for much less than $1200.

[RagingGrandpa]

Used engines, always rebuilt before we use them.

Great call on the Wiseco pistons, do you have a reason to suspect there's a clip issue with them? We've been running them for many years without a problem.

Here's another beauty:
http://www-personal.engin.umich.edu/~lovelljc/FuktPiston.jpg

[Charlie]

What orientation were they installed?

[RagingGrandpa]

The piston or the clip? The pistons were in correctly, the do have an intake and exhaust side.

[Denny Trimble]

Suddenlee was telling us how the orientation of the clip is critical in preventing this type of failure, but my engine guys say the clips spun freely in their seats.

We had a clip fail (found it in two pieces) after a full year of dyno, pre-detroit, detroit, and post-detroit use (200 hours). I believe the pistons and clips were new before that.

It might be worth replacing the clips every now and then...

[Charlie]

If you can spin a Wiseco clip in its seat you are using the wrong clips! Or at least something besides what we've been using. Those bastards are stout and I catch hell whenever I've had to try and remove one. I call them 'semi-permanent.' The stock Honda clips are much smaller gage and if you put one in a Wiseco it might spin freely.

Regardless I think those teflon plugs might be worth a look next time...

[RagingGrandpa]

One more goodie:
http://www.personal.engin.umich.edu/~lovelljc/F4iSlot.jpg

[Jarrod]

any of you guys care to elaborate on what you are doing to these poor engines? after the second time we spun a bearing and trashed the crankshafts, throwing one rod through the block, the guys at honda (factory race team in aus) asked how the hell we were damaging these motors, since the only time they ever hurt them was by going way over the redline, and even then they usually lived. We know it was because of high lateral g's, shallow sumps and dodgy baffling on our part.

[BryanH]

D_________n
10 letter word starting with D would be my guess

Bryan Hester

[Charlie]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Jarrod:
any of you guys care to elaborate on what you are doing to these poor engines? after the second time we spun a bearing and trashed the crankshafts, throwing one rod through the block, the guys at honda (factory race team in aus) asked how the hell we were damaging these motors, since the only time they ever hurt them was by going way over the redline, and even then they usually lived. We know it was because of high lateral g's, shallow sumps and dodgy baffling on our part. <HR></BLOCKQUOTE>

In case you weren't reading closely http://fsae.com/groupee_common/emoticons/icon_wink.gif It was a wrist pin that came loose on an aftermarket piston on a student-assembled engine.

[Jarrod]

sorry, it was getting to that point in between plug sanding and bogging sessions where i could only manage looking at pictures, words were too much for me to handle. I think we need more ventilation.
As an aside, how high are you guys bumping the compression up if you don't mind me asking? I assume thats what you're doing with aftermarket pistons.

[Charlie]

The Wiseco pistons have a smaller dish. But you can still bump compression the old fashioned way, but valve clearance is your limiting factor. http://fsae.com/groupee_common/emoticons/icon_smile.gif

[RagingGrandpa]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Halfast:
D_________n
10 letter word starting with D would be my guess

Bryan Hester <HR></BLOCKQUOTE>
Huh?

[Dan Deussen @ Weber Motor]

RagingGrandpa-

I would guess he was going for Detonation!

[Charlie]

Or Damnnation! http://fsae.com/groupee_common/emoticons/icon_wink.gif

[RagingGrandpa]

Has anyone gotten these engines to knock on 100 octane though? We've advanced spark rediculous amounts on our dyno and can't get it to detonate.

Premature detonation seems to be a problem afflicting only our team members, not our engines. http://fsae.com/groupee_common/emoticons/icon_wink.gif

[BryanH]

I have tuned many engines that give no audio indication of detonation, usually very efficient 4 valve chambers but not always.
The rate of pressure rise becomes excessive,
physical indication is unexplained loss of power or if your unlucky rotating parts that just doesn't want to be there any more.

This happened to a mate who runs an engine dyno, for some reason he halved the timing light readings when setting the base timing (he had done a few waste spark setups just before doing this one and had a major brain fade) on a $65000 all alloy 7litre motec'd chev. He only got as far as the 3000rpm maps when the crank tried to escape. It never rattled. the short motor was trash.No photos but there are court transcripts.

With turbo race cars I use a bolt-on knock sensor to set up the airtemp/ign maps on the track. I have found that you only hear SAVAGE detonation at high BMEP's! But don't be girly with advance or it won't go FAST!

[drivetrainUW-Platt]

I would appreciate it if someone would do a little explaining about what detonation actually is. I know it has something to do with air/fuel ratio's and that engines don't like it. Also, what is it about it that causes engines to self destruct. Thanks.

[Storbeck]

I know very little about it myself, but:

http://www.motorcycle.com/mo/mcrob/rt-fuel2.html

[John Bucknell]

That article is pretty good. The only thing I'd like to clear up is that 'knock' is uncontrolled ignition of end-gasses, rather than 'detonation' which is supersonic combustion associated with explosives. Effectively knock is the fuel-air mixture at the edges of the combustion chamber (end-gasses) getting to ignition temperature without a spark. This can be because you sparked too soon and cylinder pressure peaked closer to TDC (pre-ignition from too hot a spark plug will do this), or your compression ratio is too high, or the charge was too hot due to too little fuel, or coolant or air temp, or your fuel had poorer knock resistance than you sparked for (octane too low). There is a time dependency that allows high speed engines (10k+) to essentially ignore knock, and similarly smaller bores also help if you only have a single plug.

Anyhow, the resulting huge pressure rise can be high enough to hammer nearby metal into little pieces or more simply melt it. When a piston loses integrity this way, mucho destruction can result. And to Halfast's point, if you can hear it at high engine speeds - you're probably already done for. Professional engine tuners will put a pressure transducer in the cylinder to watch the pressure trace and look for the characteristic 'ringing' of knock. It looks like high-frequency (~10-14kHz) 'hair' on the power stroke pressure trace. A knock sensor actually picks up this excitation, but unfortunately an engine block will resonate at a different frequency and it takes a bit of experimentation to find it (and may not actually excite the block at all at high speeds).

I have a picture of a knocking pressure trace somewhere. I'll see if I can find it.

[Charlie]

Here is a knock trace (ignore the water injection stuff)

http://www.aquamist.co.uk/dc/technic/technic.html

Detonation and knock seem to be used interchangably when referring to the same problem. In aircraft piston engine texts, detonation is the term of choice for explosive end-gas burning after the spark event. I think this is because in the 50s and 60s when aircraft piston engines were at thier peak, the theory that knock was caused by detonation (sonic travel of the flame front) was preferred. After all it does have very high pressure waves and they may be sonic. But Heywood's book (late 80s?) mentions the detonation theory but favors the auto-ignition theory. The autoignition theory basically says, its not a single flame front travelling sonically, but multiple ignition points in the end-gasses.

Anyway in laymans terms I doubt you will ever seperate detonation and knock, detonation just sounds smarter. http://fsae.com/groupee_common/emoticons/icon_wink.gif

These terms are not to be confused with pre-ignition... http://fsae.com/groupee_common/emoticons/icon_wink.gif

All this is fun and interesting but if you are running a FSAE engine with 100 octane you might as well just skip this post. http://fsae.com/groupee_common/emoticons/icon_razz.gif

[RagingGrandpa]

Update-

Blew up ANOTHER Wiseco F4i just recently. Again a wristpin came loose. Cylinder #3 this time. What the hell?

[Charlie]

I dunno, we ran Wisecos in our 2002 car and that car has been through hell, no problems. But sounds like some solid plugs are not a bad idea next time.

[Matt N]

Hi. I haven't posted here yet but here goes.

A little while back (before me) there was a Wiseco-CBR600 F4i explosion here, at University of Washington, too. The witness marks and damage that resulted were this: connecting rod modified into several pieces, crankcase sawed open by said pieces of rod, but more interestingly the piston had one whole side of pin boss ripped open (at the bottom), same side of piston having no pin retainer, and vertical scoring marks where the one end of the pin touched the bore for a few strokes (presumably after the retainer left).

The explanation I was told for the engine explosion was that the retainer escaped, the pin moved, and all hell broke loose from there.

A couple of weeks ago I encountered an old F4i head and I still can't confirm which car it was run on. However, there were two places in each chamber where the pistons were clearly impacting the head. Now I don't have pics, but there is a machined edge on the top of each of the dome sides where the dome stops rising and meets the (remotely) flat valve-sides of the dome top. At the top of the dome in this location this little edge (about 1/4" long) was remachining an accompanying groove in the head, in three of the four chambers. Admittedly, a very small groove. The failed Wiseco, which is still around, has mashed over edges in the same locations.

I do know that no one regularly, if ever, has measured (with clay, or a more elegant device) the piston-to-valve and piston-to-head clearance during the installation of the Wiseco pistons here at UW in past years.

The thing that gets me is that I thought from my research that round wire retainers with an angle ground pin end were the most secure retaining method, and are used on many other racing piston designs as well as (I think) some diesel pistons. The Wisecos use this design.

As it stands, it sounds to me like people are leaning towards a failure sequence that looks like this: 1. one retaining ring leaves its groove (for reasons unknown) 2. the pin then walks out the non-retained end towards the bore, scoring the bore and creating friction 3. and also coming out of its bore on the other side of the piston 4. when the pin comes out of the bore on one side, the uneven loads make it get all cock-eyed and it rips the piston boss apart because all this stuff is happening at 11,000 rpm 5. when the pin gets all cock-eyed, the rod doesn't like it much either, and buckles sideways from the lateral forces imparted from having no support on one side of the pin, and proceeds to void the engine case in a spectacular display.

I'm honestly wondering if the failure sequence couldn't look like this: 1. the pistons slightly touch the heads in the two locations I discussed earlier - not in all cylinders, but just a couple of them 2. after sustained running with bursts of high rpms (read - some dyno testing, some test driving, etc.), the one cylinder with the most severe piston-to-head 'disagreement' has weakened the rod such that the rod breaks at 11,000 rpm 3. during the failure of the rod, which would take a couple of cycles probably at least, the pin boss is ripped apart as the rod buckles laterally 4. and also, as the pin boss is ripped apart, the retaining clip makes a hasty exit, because its groove was just greatly increased in circumference 5. and last of all, everyone sees a oily smoky display of mechanical defeat as the con rod saws the case open.

Any takers on which failure sequence causes this? And Raging Grandpa, can you look at the failed piston and chambers and see if there are any marks there? Generally, anything thats shiny raw aluminum is a dead giveaway... (also, just for the fun of it, look for marks from valves tickling the edges of the piston dome valve reliefs - I found those too...)

Sorry it's so long.

Matt

[Charlie]

Matt-

Interesting theory. However I have done extensive research into slapping the cylinder head with the Wiseco dome. http://fsae.com/groupee_common/emoticons/icon_wink.gif And my research indicated that if there is sufficient pressure to 'shatter' a rod like you say, that the rod bearing would definitely fail first. This does cause even worse cylinder head contact, but soon enough the rod fuses to the crank as the engine trys to protect itself. http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

Seriously, I did 'clay' the head for valve clearance, but never checked for dome clearance. And with rod stretch the pistons banged the head. Eventually a rod bearing (actually 2) spun and one seized. I can't imagine that a rod would break before a bearing fails.

[BryanH]

I have to join into this one, In many years of building/racing 16 to 20,000 rpm kart engines I've found only 1 cause for jumping circlips (excluding mishandling, using S/H circlips etc)
and that is misalignment of crank centreline/ bore centreline which must be exactly 90deg.
and possibly more relevant piston pin bore to piston skirt must be exactly 90deg. Enough misalignment turns the pin into a jackhammer which will drive out any circlip no matter how tight. Which way it hammers tells you which way the misalignment goes. Bent rod doesn't do it, does the bearings in first

[BryanH]

And Charlies research is spot on

[Denny Trimble]

So Halfast, you're saying crooked cases (cylinder bores and crankshaft axis not perpendicular) or crookedly-machined pistons are the root cause? If so, I'd guess the cases are fairly straight, which leaves the pistons themselves as the culprit...

Can you translate "S/H circlips" into northern-hemispherese?

[Scott Harsila]

Remember slight piston to head contact is actually desired for optimum dynamic squish. http://www.mototuneusa.com/the_factory_superbike.htm
I doubt that slight(I do mean slight)contact is causing any harm to the wristpin clips.

Anyone talk to Wiseco yet? I bet this problem is a lot worse in an unrestricted built sportbike that is seeing 15k rpm.

[Charlie]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by Scott Harsila:
Remember slight piston to head contact is actually desired for optimum dynamic squish. http://www.mototuneusa.com/the_factory_superbike.htm
<HR></BLOCKQUOTE>

You say 'remember' like this is something we should have read in our engine texts. Where might I find this statement justified somewhere that doesn't feature animated .gifs? No offense, it is interesting but its just another website, with not much if any backup data to the theories.

Besides his 'proof' of the piston hitting the head is a clean area on the head. This doesn't mean contact, it just means close enough that carbon cannot build up there.

The only reason the Wisecos hit was because out CR went from 12.8:1 (Wiseco advertised) to 13.8:1. By means of getting the piston closer to the head. It would not happen in a sportbike unless they performed similar modifications.

[Scott Harsila]

Charlie:
Point well taken.

Did you mill the head as well? Or is 13.8:1 the resulting compression with the wisecos(as measured)? Just out of curiosity, how did you measure this?

[Charlie]

Stock head, milled deck. Stock Compression is 12.0:1, Wiseco advertised is 12.8:1. We brought that to 13.76:1 to be precise. Measured by chamber volume, head gasket thickness and diameter, bore diameter, piston OD and 1st ring depth, Piston dish volume, etc. Formulas can be found fairly easily. It was an actual measured CR, not estimated. Of course I did not re-measure the chamber volume after I put dents in it with the pistons http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

[Scott Harsila]

I was just curious how you measured because most people don't;-) Sounds like you did the calcs thoroughly.

Could you tell at which RPM it started to hit? Any idea what the static piston to head clearance was? How long did the motor last?

[Charlie]

I really don't know when they started to hit. I also don't know what the clearance was-- I measured valve clearance at .06" but didn't think about the chamber dome stuff. The Wisecos are meant for F4 but F4i has a slightly different head so that may have been part of it.

The engine ran for awhile, we got a power curve out of it on our crappy water brake (getting a curve takes forever and is excruciating on that thing). Then the muffler blew off and we shut it down, and the bearing seized. Power started dropping a bit before the muffler blew off so it was on its way out.

[BryanH]

S/H = second hand, pre loved,
Scott is kinda right about touching,the idea is to have clearance of 4-5 thou at max rpm. A 100cc kart engine with square bore/stroke has around 26thou rod streach at 19,000rpm so squish is set to 31+. A 125cc Rotax Max is rev limited to 14,200 and needs 37thou.
Actual touching is bad experience for piston. In a 2stroke jamed ring/seizure is next. Buzzing engine beyond 21,000 however brief requires new everything from crankpin up.
Squish is measured with bent solder wire down spark plug hole, held against bore in 4 places while engine is rotated over tdc. Checked occasionally to measure rod stretch.

[BryanH]

Raging G, I've reread the posts more carefully (as I should). In a 2stroke jumped circlips are fatal but in a 4stroke not so, as the wristpin can only shift across a few mm and the circlip/pin just slowly damages the bore.
I've seen many engines that have thrown rods or broken cranks and at least 1 wristpin boss split in half is present in most cases.
The change in piston crown shape/compression may be causing abnormal combustion, beating on the piston/rod assy. Another possibility is that the wiseco does not have the structural integrity to stay in 1 piece at high rpm and simply leaves the rod and meets up with the head, leaving the rod to find it's own way back down the bore....