peejay

I finally was able to get around to pulling apart the engine that I killed in September.

The engine that preceded it suffered from flattened rear corner seal springs on the rear face of the rear rotor, severe gouging on the rear end housing, and destroyed oil control rings. At the time, I thought this was due to low oil pressure and severely worn rotor bearings allowing the rotor to wobble and take everything out.

I put this engine together with good used rotors, a modified GSL-SE eccentric shaft, new apex and side seal springs, new corner seals, new O-rings in the best oil control rings I could find, painstakingly matched up used side seals, and used wire-type corner seal springs (couldn't get FD springs). All used springs were well within specs for seal pop-up, the only thing questionable were the apex seals, which had poor pop-up due to being worn on the back side where the springs ride, but they clearanced out perfectly in the new rotors' grooves. I ported some 6-port side housings and ran them with my old T2 rotor housings.

400 miles later, it was runnin' great. Not as powerful or torquey as the half bridge T2 engine, of course, but it was a happy engine with no leaks or smoke, And great oil pressure.

Went to a two day rallycross. On day 2, compression started going downhill. I thought it was from dust and my already-worn apex seals, at the time. It was extremely dusty... By the time competition was over, the car would barely start, and I turned idle way up to keep it from stalling. I filled up with fuel before the 150 mile drive home, which turned out to be a good idea because by the time I made it home, the engine had such poor compression that it would barely maintain highway speed and would not hold an idle. I felt like an a-hole holding the engine up around 2500 at stoplights but I did NOT want it to stall! When I got it home the engine did stall and it never restarted. Compression was extremely low, 50psi front and 30psi rear.

I pulled it apart today...

peejay

All three side housings are grooved badly. The grooves start right where the spark plugs are and continue about 1/4 or so the way around the housing. It looks like it's the side seals that were doing it.

On the previous engine I thought this was a symptom of a different failure, but I'm starting to think that it's not.

The front rotor's seals and springs don't look too bad, still lots of tension on the side and corner seal springs, but the rear rotor's parts are basically hosed. The seals all still move freely but none of them pop up higher than the rotor height. And of course the oil control rings are shiny all the way across instead of .016-.020 like when I'd put the engine together. And the kicker - the apex seal grooves are all V-ed out, so the rear rotor is junk.

Engine never lost good oil pressure. Good is about 45psi hot at ~1000rpm idle, quickly running up to 90-95psi. The bearings have some copper showing in spots but they look normal, not scored. Eccentric shaft still looks good and still has the crosshatching on it from when I polished it with wet 320 grit emery paper, no shiny spots like it'd been rubbing anywhere.

I'm really at a loss here. The only thing "bad" that I can think of is that I'd been accidentally running with total timing in the 30-32 degree range (L+T) because I had misread the marks on my pulley, but I never heard or felt that characteristic chirp/buck of detonation. I also was running 92 octane in competition (along with about 1:80 Royal Purple two-stroke oil).

What else am I missing? I would like to actually drive my car this year instead of work on it...

bumpstart

spec out the housing widths at the FSM listed points and then same for rotor

especially check rotor land protrusion

methinks rotor too wide for the housing due to warpage of housing ,, and or pushed out rotor land/s due to over rev

peejay

Before assembly, I narrowed the rotors a couple thousandths of an inch for a little extra safety. The sides of the rotors still have the machining/sanding marks on them, they're not burnished like they were rubbing. At any rate, I'll doublecheck that when I next get to the engine parts.

Max revs were about 9000-9300, normal shift point 7500-8000 due to the crappy porting. I'd seen more than that on the previous engine built to similar spec with no issue. (I had it open a couple times for other reasons)

j9fd3s

iTrader: (3)

just to sum up, the seals are very worn, very quickly?

what is the air filter situation?

peejay

No, the side seals are fine - the ones I sampled were .103-.113 in height (were used to begin with), the corner seals were all .267" thick (were new) - it's the springs that were flattened out so they don't protrude anymore.

Oh - and it DID have the FD corner seal springs. The ones that were flattened down seemed to actually be stuck down like the diameter had been increased. I thought this was the engine that I'd assembled with wire style corner seals, maybe it's the GSL-SE engine in the car right now. I did enough engines last year that I just plain forgot what engine got what parts.

More info: Rotor housings both measure at 3.148" (79.97) at any point I care to measure at. This is where they measured out when I first got them a couple years ago. Rotors were a little difficult to measure since I still had the oil rings installed, and I don't yet have a proper workspace to pull everything apart and clean and do a thorough measuring, but I was able to squeeze them down enough with one hand to measure 79.80mm with the other hand, at which point I feel that I confirmed that it's not a gear walk-out. Not that I could see how a gear walking could cause the side seals to gouge the housings, but nothing's known until you measure.

Two other interesting tidbits. I'll have to wait until my new apex seals get here before I write off the rotors. I looked a little more closely at the one rotor and noticed that the corner piece was thinner than the main length of the seal. I pulled that seal out and measured the seal thickness at 1.6mm in the middle, and the cross section is definitely wasp-waisted so it's probably even thinner in the middle. The corner piece was about 1.3mm thick. So on the bright side, it looks like Atkins seals are rotor friendly

And the other weird thing. I can't remove the upper dowel pins.

bumpstart

the rotor gear face forms the rotor land on that side of the rotor and thus prevents the rotor tilting and placing weight and wear on the corner and side seals

if your dowels are stuck i suggest there is a chance these housings are warped,, and while they spec for width,, are they indeed flat?

try them on a very flat bench with with a feeler gauge

peejay

If it was the rotors walking around, why are the marks only at one spot on the housings, and there's NO evidence of rotor to housing or e-shaft to bearing contact?

I also fail to see how possibly warping the rotor housings is even relevant - if it's the same thickness all around, then the side housings will hold them together. At any rate, the engine sits "flush" when loosely stacked and the chrome is still intact everywhere. I've found that the chrome going misshapen is the first sign of housing distortion, as it's how I noticed that my peripheral port's intake ports were starting to collapse. Difference in housing thickness was only a thou' or two, but it showed up in the chrome plain as day.

I'm fairly sure (now) that the engine has been detonating, because everyone seems to agree that flat corner seals is a clear sign of detonation. Too much timing plus 100F air temps, 210F oil temps and 220F coolant temps plus a lot of time spent at full throttle 3000-4000rpm (mechanical advance is all in well before 3000) would seem to support this, but the nagging voice in the back of my mind says that I never heard or felt it. And what was causing the side seals (repeat, not the rotors) to wreck the side housings like that?

Could low-grade detonation with sufficient heat cause that scoring? It is awfully suspicious that the grooves start and are deepest at about the TDC:spark mark.

GtoRx7.

iTrader: (12)

It sounds like a loss of lubrication on the rear rotor. Since you pre-mix this technically isnt possible however. So my next thought is that the rear rotor was not getting oil cooling on the inside, making it super heat which then all the hard seals get retarded hot. So hot that the oil film being sprayed in with fuel burns off? Have you checked the e-shaft rear checkball and spring for a clog? Is this the same e-shaft that was in the previous engine that had a rear rotor issue?

Oh yeah, I have detonated for years with bad fuels on my n/a 13bre. Even with repeated 9k rpm for a few years and countless times seeing pre-ignition from testing no damage anywhere inside the engine was found.

peejay

Different E-shaft, different rotors. Basically the only parts I re-used from the previous engine were the apex seals, rotor housings, and oil pan. (And that infernal crank pulley) E-shaft was pulled apart, cleaned, teardropped, polished, cleaned again, and reassembled.

Now that I think of it, the most heavily worn apex seals were on the rear rotor. Worn meaning the backsides were worn out where the spring was riding. The front seals still had that snapping sensation you get when you push the long spring over the short one and pop it down into the seal, but on the rear rotor, two of the apex seals were so worn that they wouldn't hold the short spring in by themselves and I had to figure out a way of putting the seals in with both springs, all at the same time.

At the time, I was thinking that it was mid-September, so the engine only had to last for another two months before I could pull it back out and do it correctly with new seals. But now, I'm thinking that the seals could have been bouncing (gee, really?) and even if it wasn't detonating, it could have been suffering the equivalent of REALLY early ignition. They were 2mm thick when I put the engine together, I KNOW this because I checked the grooves for proper clearance. No way would I have put the engine together with seals with the sides (groove faces) worn like that!

It's also possible that the distributor cap was crossfiring internally, but I threw the cap away over winter so I can't check it. At any rate, I'm going to be doing away with the distributor and the MSD when this goes back in.

What really gets me is that the rotor housings even now still look okay, just some waviness that has been there since three assemblies ago when I made it my first half-bridge. If this was a 12A, the rotor housings would be the first to go... hooray for progress?

peejay

There's no sense crying over every mistake
You just keep on trying 'til you run out of... housings

I'll just treat the problems symptomatically, moving backwards. I know that the corner seal flattening coincided with the switch to the Racing Beat pulley. I also know that the housing grooving coincided with the flattening of corner seals.

Here's a theory: The corner seals are broad and the springs help keep the rotors stable in the housings. If the springs have lose tension, then any stabilizing force remaining must come from the side seal springs, which are acting on a far smaller surface area, and the oil ring springs, which are also dragging across the ditch being created by the side seals.

That doesn't fully explain why the side seal grooving is only in the area of combustion pinch to about 90ATDC, though. Excessive combustion pressures *might*, but I have a hard time believing that 600hp 13Bs don't have this problem but 175hp 13Bs do (This 6-port was nowhere near 200hp, sadly)

I just got my new set of apex seals in the mail, now I can properly measure the rotors to see if the extreme clearance was entirely due to the apex seal wear, or if the rotors are junk.

The plan is to scare up a hunk of aluminum so I can whittle up a Turbo II to 4-port 13B adapter, then run my TII-ized 6-port intake manifold on a set of bridge ported 12A side housings on GSL-SE rotor housings, with Series 4 internals. And then run very conservative ignition timing, and actually running the ignition timing that I think I'm running, until such time as I can hit a dyno and tune it for a couple degrees less than diminishing returns.

So, er, I may be giving you a call in a month or two, Logan.

You oughta ride along at a rallycross sometime. OVR holds them at one of the overflow lots at National Trails. I haven't been able to see any in-car vids for a while and I don't pay as much attention to the tach as I should, but I know that I have a habit of running second gear down to 3000-2500rpm before downshifting. The problem is that 1st gear is just nothing but tire spin, even with a wimpy little 6-port, rear suspension that works, and WRC-sized Michelins, so bogging in 2nd gear is still faster. Thus why I've been sacrificing at the altar of mid-range torque. And why I freakin' LOVE the half bridge/ping-pong intake combination.

I'm hoping that the Miata gearset will help in this regard. It'd mean more shifting, but the 3.1 First would help traction, and the 1.9 2nd would maybe keep me from running the engine out to 9k on the long straights that the Lot C field tends to generate.

j9fd3s

iTrader: (3)

hmm you know i had an engine die like this too. the engine was a stock S4, in a gsl-se. my roomate at the time took it drifting one rainy night, and was telling me how it was bouncing off the rev limiter the whole time, which of course it didn't have....

so it runs fine for a couple days, then i hop in it to do a trackday, get like a block, and it takes a crap... the quickie compression test reveals zeros...

so i pull it apart, expecting to see carnage, but instead it looks great. all the seals are there, they all look ok, housings look great, rotors are clean.

the ONLY thing i could find, was that the corner seals didn't spring up, and it had actually pounded the corner seal springs INTO the rotor.

i was gonna say i didn't have the extreme wear, but that engine had 130k on it, with plenty of 9000+ rpm runs.

i ran stock timing, but it was running out of fuel, so it would ping and then stop running in 3rd.
with your incorrect pulley, maybe it was just detonating? and maybe in a 255hp FD motor detonation dents rotors, but in a 175hp engine maybe it gouges side irons?

TonyD89

iTrader: (10)

You would have to have a lot of rotor bearing clearance to to push the tip over and crush the seal. As you said, it's not 600 hp. But, you did say you revved the hell out of it. I don't know. What about the rotor land clearance and tip clearance?

peejay

The rotors never made contact with anything, they haven't been hammered. I didn't get to measure them tonight because something else came up, but the apex seal groove clearance opened up dramatically. At least some of this is because the seals themselves are heavily worn on their sides. I've never seen that happen before, usually the seals are the same thickness but the rotors are V-ed out. (The record was a 12A that was so worn that I could slip an apex seal spring in next to the seal, and that was on the rotor that got me home)

I wouldn't say that I "revved the hell out of it". When it was a street port, I saw five figures on several occasions and would run it to ~9200 or so at autocrosses to keep from having to momentarily shift to second gear. (It didn't make much power over about 8200, but it wasn't totally useless) When I pulled it apart to bridge port it, the internals looked fine. The bearings were trashed, but that was because I'd been running on about 30-40psi hot without knowing it. Even with .007 too much bearing clearance (that's what, .011" clearance?) the internals looked fine. Apex seals were still at the .002" slot clearance that they were when I first assembled the engine as a street port, as well.

All the problems seem to have started when I assumed that the RB pulley was 0-10-20 and not 0-12.5-25. I had been setting timing by running it up until the mechanical advance was all in (about 2000rpm) and then setting the timing to be halfway past the last mark. So I thought I was getting 25 degrees all-in, I was really getting 31-32 degrees, with no split, at 2000rpm where the engine had enough VE to make ~140ft-lb... assuming that there was no cross-firing to the wrong trailing plug, but I know from experience what that sounds and feels like, and it is UNMISTAKEABLE.

So I know that this was a problem, I am going to eliminate it as a problem and see what other problems come back. I'm going to do science to it

j9fd3s

iTrader: (3)

um 31-32 is too much @2000rpm. mine pings around 18 degrees, and the Rx8 does too, although i'm not sure what timing it runs (20? 25?)

and you're making me worry about MY RB pulley!

peejay

AH HA!

So while looking for Group B homologation paperwork, I found a copy of the Competition Prep manual. I had a copy but that computer is dead, so I re-downloaded it and was poking around in there.

Saw this on measuring stepped wear on the side housings, my current bane:

Note:
a) Pay particular attention to the spark plug side of the side housing, this is the combustion area and is usually the location of the greatest wear.
b) If excessive wear rates are measured over a relatively short period of time, for instance .1mm after 2000 miles or less, it may be an indication of dirt or other contaminants in the engine and the intake air filtration system should be checked.

D'oh. Should have just trusted the gut instinct.

TonyD89

iTrader: (10)

Do you think the corner seal could get hot enough, for any reason, to kill the spring temper in the corner seal spring?

peejay

Minor update. Cleaned the rotors - the apex slots are all exactly .080" except for where the corner piece was riding - on the open face side of the front rotor, and on the gear side of the rear rotor. .082-.083 there on the front rotor, and .084-.088 on the rear rotor.

Half tempted to try to "fix" them and run them rear to front, but only half.

Also there evidence of a TINY bit of rotor to side housing contact on the rear face of the rear rotor. By tiny I mean a roughly .5mm by 2mm shiny spot, just "ahead" of the corner seal. So it'd be just over the trailing side of the side seal. Given that this is the face where the seals weren't even proud of the rotor, this is understandable.

To answer your question Tony the corner seals look awesome and the sorner springs are all "springy" and not dead like they lost their temper. They're just flattened. The ones on the front rotor don't look that bad actually, but if I'm going to replace some, I'm going to replace them all. The corner seals all look awesome and still have as-new height.

The side seals on the rear face of the rear rotor, incidentally, had bizarre taper wear. The leading edges of them are worn significantly thinner. When I did my street port I made extra certain that the leading edge of the side seal wouldn't fall into the port, so I know that can't be it. There was, hoever, a significant "landing zone" at the closing edge of the ports where the trailing edge of the seal was slapping it. Guess I didn't taper the port enough.

Apex springs are shot. They still have adequate height, but the long springs are definitely asymmetrical now, and I wouldn't chance running them if I didn't have to. Fortunately, Atkins use OEM springs so I'll get those from the dealer when I order new corner seal springs.

Barry Bordes

Peejay, do you have any pictures of the side plate wear?

This is what wear from side-to-corner seal clearance of less than .002" looks like (and notice it is at the plug/compression area) .

Barry

peejay

I wasn't able to get a good photo because the camera kept picking up reflections off of the surface.

The housing you posted would be considered by me to be an extremely good used piece and would be assembled without worry.

peejay

Pictures en route. I figured out how to get the lighting set up and configure my camera to work right.

But here's some interesting trivia: The vertical groove on the spark plug side of the rear end housing was .025" deep. I didn't forget a naught. The side seals from that side of the rear rotor were worn down to .090" in the middle and .067-.070" at the leading ends.

Only three of the side seals measured okay for taper wear. I haven't found a spec (just for protrusion) but the three "good" ones were .117" in the middle and .114-115 at the ends. Probably won't reuse them, they look worn on the sides. All of the others had a minimum of .010" taper from the middle to the ends, and were at most .110" tall.

Friggin' cone air filters.

peejay

You can see that not only is there a vertical line, but there's one at about a 45 degree angle slashing across where the oil control rings ride. This is the same wear I had on the previous engine, but only on the rear end housing and not all three, and that engine did not have issues on the other side like this one did, as seen below...



These gouges aren't as deep but they're more visible because of how the light is. Plus they extend further around the rotor. You can clearly see where the leading edge of the side seal was riding (these ports open at 10.9mm from the coolant seal groove) and you can also see the landing zone where the trailing end of the side seal was smacking the port closing edges, even though I gave them a hearty bevel and (thought I'd) put a decent angle to the closing line. Guess it wasn't enough. Again, though, the vast majority of the side seal wear was on the leading side, not the trailing.

Not bad for 400 miles, eh? Oh wait, I mean not good. But on the bright side, I've got a GREAT tracing of where the side seals ride, now

j9fd3s

iTrader: (3)

we put a new trans in the honda a couple weeks ago, and i had the intake tube off, and there was just actual dirt in there, like the kind that's on the side of the car when you go off track.

AEM cone filter! its old but still, its not very good either

and wow are those grooves bad!

Barry Bordes

Peejay,
That is the worst wear I have ever seen.
On enduro bikes we found that oiled foam filters worked great for dusty conditions but if it ingested water too it would wash the dirt right through the filter.
You might need a snorkel.
Barry

peejay

More:

http://www.facebook.com/media/set/fb...6&l=8d70807ccf

No water to be seen anywhere...

Part of the reason I did so well on the second day (posted fastest 2wd times, with an already-sick engine!) was the competiton was having mechanical issues... one broken trans, and a couple people were getting so much dust in their airboxes that the cars wouldn't run. That, and I was able to take full advantage of the long straight, letting the engine sing out past 9000rpm without having to upshift momentarily or downshift in the braking zone. (Someday, people will figure out that automatics are kinda the best idea for this)