Thanks for all of the inputs gents. Definitely plenty of sound thoughts, some which we've also had.

Marcel Though I think you're right about an obstruction not being a point failure, were just checking for whatever sounds even remotely possible. It's sort of like when you've lost your keys or wallet and you start to look in places you know it can't possibly be (i.e the fridge or closets) but you're just perserving on hope at that point!

SCOTTY: Thanks for the kind words and very resonable questions. As for this challenge, it's all good. Been here several times (though to a lesser degree) in this rotary-masochistic fetish that I've developed over the last decade and a half. This too shall pass!

As is evident from Aaron's replies to your questions/comments, obviously great minds think alike. Unfortuantely I think we've already rejected most the possibilities posted. I have asked about rev limiters and confirmation that the MoTeC Traction Control was completely off, though I had low confidence that either was truly the problem source. If they were, it should have been a more dramatic cutout.

Calculon You had me thinking about the Map/EMAP efficiency mapping, till Aaron's reply. Unfortunately, I gave up the EGT ports to plumb the EMAP ports, so no EGT sensing in the setup. I think your right though, in that it might give critical indications especially between rotors.

Aaron THANK YOU for the contributions to this thread ! Your comments and replies are exceedingly welcomed. Best of all, you've saved me quite a bit of typing! ;-)

I've mentioned that the power drop correlated with low Lambda values and wondered if a malfunctioning LSU sensor might be faulting to a lean-bias. This may be no real correlation, but if there is, it might be sufficiently rich to rich-misfire even though it's not obvious in the dyno video or the log. In desperation, I wonder if E85 might have milder rich-misfire characterization than straight gasoline and therefore be asymptomic relative to the rich-mifire of gasoline. I'm sure the E85 tuners on here know how E85 rich-misfires... Enzo, Chris, Brian, Glen... a little help please!!

Have ya'll changed out the O2 snesor to reject the possibility.




I know there are one or two smart tuners on this list that have ideas. I sure would appreciate you thoughts. I've paid for my first track event in FIVE years and sure would love to drive the VIR Full Course at the end of August, but I could really use some help to pull it off at this point.

 

Carlos, I have a brand new bosch lsu 4.9 at the shop we could swap in as that sounds like a reasonable enough possibility.

I'm more than happy to contribute and try and get this thing going. It's just ridiculous that something like this happens to the most detailed person ever...

 

Carlos, nice to see you've got the right attitude about this. I had a professor who once said "How do you eat an elephant? One byte at a time." It was partly a cheezy programming joke but partly a reminder that even complicated things are just a lot of simple things stacked up or arranged in a very particular manner. I'm sure you guys will find it eventually. Didn't mean to imply that something simple like a rev limiter or traction control should catch you guys off-guard, but it's a nice simple thing to check and rule out quickly.


For what it's worth, I haven't personally done much work with E85 (just a couple of piston-engine cars), but what I've heard from more experienced tuners is that E85 seems less likely to misfire due to an overly-rich mixture. Both E85 cars I worked with made better power with a richer Lambda using E85 compared to pump gasoline (we tested on both fuels at one time or another). A friend's non-turbo Miata was happiest at an indicated 0.80-0.82L, and a turbocharged Nissan SR20 made better power simply by richening the mixture from 0.8-ish Lambda (E85) to 0.7-sh Lambda (E85).

Each car was using a different O2 sensor and tuned on a different dyno... neither was using a Motec O2 sensor or tuned on a Mustang dyno so don't get too hung up on the values. I've heard that most O2 sensors are generally designed and built for streetcars so they tend to be accurate near stoich and less accurate as the mixture gets further from stoich... then from there you've got to worry about the aftermarket circuitry that's talking to the sensor and interpreting the data. Long story short I'm not sure I'd trust the O2 sensor as the end-all be-all way to determine that the fueling is what the engine wants... presumably the tuner has already tried adding or removing 10% fuel to see if the misfire gets better or worse?

If exhaust pressure seems high (I've never measured and don't know what would be 'too high'), have you tried disconnecting exhaust pipes downstream of the turbo? Don't forget your earplugs!

 

Aaron: Thanks for the help, in what ever maner you chose!!

As for the O2 sensor, the hard part is actually going to be changing the sensor connector since I think there was an old LSU 4.0 in there. For that matter, Paul had asked me whether it's easier to change it from the top or the bottom, so he may have already done it.

Scotty Every input is welcomed. It's so easy for problems to often "hide in plain sight." Like you intimated, due diligence requires checking the obvious as well as the nuanced.

Your comments about low AFR (as read on a gasoline calibrated Wideband) make perfect sense. When you correct those values back to Lambda and compare to E85's stoich, it correlated well. As for the exhaust backpressure, it is the one sensor correlation that appears to be the most suspicious. Just have to keep my fingers crossed that there is something there, and if not, that the "Rotary Muses" look kindly on us! ;-)

Cheers, Carlos

 

Looks like we'll know unconditionally by next week if the problem is a electronic control system problem or not. A recently purchase 4-channel PicoScope will be put to good use logging all of the key trigger and ignition signals to unequivocally determine input/outputs and function. If the logs don't reveal any anamolies during the hi-RPM hesitation, then we're moving on to mechanical obstructions or deficiencies. Got my fingers AND toes cross!!

Aaron. Any chance I could bribe you into a picture or two of the new Bosch coils. Already thinking ahead to how I'll permanently wire and mount them before VIR at the end of next month. Keeping the faith!!

 

Coil itself:



Coils on bracket resting in engine bay during testing:

 

Amazing how much easier thing are with the right tool (PicoScope 4223 Automotive O-scope). Even though I had scoped the ignition trigger output before, without logging it, I didn't have enough resolution to see the problem. First log with the PicoScope, and Paul found a clear problem.

A test of an ignition coil (TF) ECU input trigger, primary coil amperage, and secondary votage clearly shows the misfire rate. The attached picture is just the primary coil winding amperage with the dashed cursors bracketing ~7300 RPM.

Now it's back to the CAS sensor, wheel and wires to hopefully eliminate what should be an erratic trigger input.

 

That PicoScope is an amazing tool! Would have loved to have it when I was experiencing ignition difficulties..

 

Much as "Nature Abhors a Vacuum," my Wrenchin Abhors a Rotary Vacuum. And so I fill it...

The fuel pump and injector planning tool I've been using to assess 044, Cosmo and Supra pump capacity is attached.

Spreadsheet - Fuel Pumps Specs

Hope someone else finds it handy.

 

Car is fixed! But, spreadsheet won't open unless I get proper crudentials???

 

Right on Aaron!! Got the good news in the mist of some crappy personal events so it was doubley welcomed.

Anyways, as the old saying goes, "For want of a nail (or wire)..." Guess I did a shitty crimp tying in the tach input to the Lead Coil trigger.

As for the spreadsheet, I've gotten rid of the credentials requirements, so it should be viewable. Please let me know.

 

Glad it's fixed. The OEM tacho wiring gets its signal directly from the stock ECU, does the M4 not have a spare output that can be used to send a tach signal to the dash?

 

Scotty, there is an MoTeC tach ignition adapter/expander. However, reading off the Leading coil trigger works fine as an input to our stock tach and saves from using a dedicated ECU output just for the tach.

Per Paul's (and many other professionals) best practices, I will now incorporate a pull-tests on every crimp as a SOP.

"Murphy's" resourcefulness and creativity never cease to amaze me!

 

Man..... all that head scratching b/c of a bad crimp... Sigh...

Glad you found it. Hopefully the car will be running and driving shortly.

 

Hy Carlos, I noticed the stright fit on your fuel pump exit line.


I have Bosch 044 (I think you do too) and i comes with a fitting for a Banjo fitting. (side exit) But the banjo fitting piece also has some sort of control valve in it. does it really need that? It also seems to restrict the fuel pump exit...

Iwas thinking about getting a straight fit as well but with thesame size as the fuel pipe (on the tank cover) and fuel line.

did you use a 8 or 10 mm fuel line towards the tank cover?

thanks

 

Sorry Speedwork, but I'll have to defer to Aaron for question about the in-tank plumbing since i didn't install that pumps. From the looks of it, the orginal Bosch checkvalve may have been replaced with a barbed hose end fitting.

BTW the banjo and checkvalve fittings have come standard on all of the -044 pumps I've ever purchased. I've always thrown away both fittings and replaced with an AN-08 outlet adapter.

 

Ok thanks for the info :-)

 

Well, my CYM be-otch won't give me a break. Looks like a lost a front apex seal while on the dyno about a month ago. Best part is I have the log, and there is nothing obvious that might have cause it,. I guess I'll hopefully find out later this winter when I crack it open.

So for now, the first question is what are the effect of this level of damage to a turbine:
- Is it resueable with a few points of efficiency loss and a shorter bearing life?
- Too much damage with >5% airflow degradation; requires replacement?

Sean, Enzo, sure would appreciate your opinions!

 

From a practical standpoint I'd run it until the engine management is sorted. No point loosing another turbo should the engine let go again trying to nail down your current issue.
Regards,
Crispy
- **NOT A TURBO EXPERT**

 

You can clean it up and use it for now. But it will be off in performance compared to a good one, you may also run into balancing issues if you remove too much material and that will cause premature failure as well, so I would consider it a temporary thing only.

~S~

 

I've had a turbo shop dress and re-balance a turbine wheel for fairly cheap. It just reduces the risk of the metal fatiguing and tearing. Obviously some performance might be lost, but it will sure save you a lot of money until you get your bugs sorted!

 

Crispy and Sean, thanks for the helpful, pragmatic replies. I think you're on the money and will in fact use this turbo as a stop-gap until I've sorted out the rest of my setup. Guess the only question is whether it's worth it to "dress and re-balance a turbine wheel" if the turbo will only see less than a dozen hours of use before replacement.

BTW, Sean I'd still like to discuss if I have any option in turbo selection with this short manifold. Will try to call again sometime this week.

Trots, thanks for the heads-up on the potential repairs options.

 

After cracking the keg open, the failure mode was glaring and embarassing... my shitty port exhaust work!!

Looks like when I did the exhaust ported back in 2002-ish, I did't sufficiently bevel the closing edge of the RB-templated exhaust port. The ceramic 3mm eventually must have ripped the chrome off a the port edge, cracking in the process. The associated rotor broke off 3-4mm long chip off the apex seal groove at the corresponding edge.

The only small consolation was that neither side housing suffered any damage.

So my only real question now is whether I want to go back to the 3mm Ianetti's or try OEM 2mm seals... Guess I'll have some time to mull it over waiting for Mazda Motorsports to reactivate my membership.

 

Damn Carlos, sorry to see those pictures

Good news is that we've started porting your S8 irons and you won't have to worry about any of this kind of stuff in the future

 

Wow Carlos. Sorry to see this but the silver lining is you know EXACTLY the root cause and the associated fix.

Regards,
crispy