The Lotus Cars Community banner
  • Hi there! Why not register as a user to enjoy all of the benefits of the site? You may register here. When you register, please pick a username that is non-commercial. If you use a name that appears on any search engine commercially, you must pick another name, whether it applies to you or not. Commercial usernames are for supporting vendor use only. If you want to become a supporting vendor and grow your business, please follow this link. Thanks!

Blew motor... rebuilt and still running lean

10K views 46 replies 13 participants last post by  johnnypolex 
#1 ·
Guys,

I installed the VFII supercharger and drove it spiritedly for 3 weeks (only 1 autoX day) till the motor one night decided to lose all power, blow oil and smoke out intake filter, and seize up. My friend and I disassembled the motor and found 3 pistons had seized rings and broken ring lands from detonation (see Picasa album below). All 3 cylinder walls were also obviously damaged.

A new Toyota short block from MWR and BOE surge tank were installed and set to 47psi at idle. A Larini de-cat pipe was heat wrapped and installed during the rebuild which Jermaine mentioned would not cause any additional problems. The dyno session today confirmed that the motor was still running extremely lean (>15:1 @ 7k rpm) :wallbang:. Any ideas would be greatly appreciated.

https://picasaweb.google.com/107666765286885491188/LotusBlownEngine#5755918049494121186

https://picasaweb.google.com/107666765286885491188/LotusBlownEngine
 
#37 ·
well measuring A/F on dynojet the sensor is placed in the exhaust tip. This sensor is like a microphone (i'm sure you've seen it) so it goes several inches inside the tip. At WOT not sure (especially a decat, with how fast the gases fly out of the tip) how outside air will mix in.

Dynapack's sensor gets placed inside one of the bungs (for decat).

Having said this, I'd prefer the bung sensor. Though is more work to prepare (if you have the grille on).
 
#38 ·
I'm just speaking from experience.

I've seen lots of widebands (and narrowbands) act up in the presence of free air... Many times it doesnt even appear to be free air at all.

Case in point about a month ago we were tuning an LS1 car. One bank just constantly showed long term fuel trims about 20% (!) higher than the other. After swapping sensors, then wiring bank to bank it just kept following that particular physical bank. We ended up pulling the cat back and found some pinholes at the top of the collector flange causing a very small exhaust leak... One we couldnt even hear. After welding it up and strapping it all back together, fired it back up and the trims were within 1%...

Have a 2jz 240 I'm helping a guy out with. At idle its reading 16ish AFRs if I add 20% fuel it hardly moves. That's with an open downpipe with the sensor located about 2 FEET from the outlet. I will admit the readings get much more reasonable at a cruise or wot, but the point still exists even 2 feet from the exit and it skews the results.

Having a sensor located at the tip on a twin exhaust 4 cylinder car is just not a good way to measure AFRs. It most certainly can change the reading. When trying to diagnose a problem like this you really have to be sure you are getting an accurate reading or else you are simply chasing a red herring.

Good afr data along with some OBD data will go a long way in finding out what is going on.
 
#39 ·
EGT might provide insight as well.

Until the A:F is rechecked, one should probably assume that the readings are correct and treat the engine accordingly, as there is no evidence that the A:F is correct and with one engine having been fried... that suggests that not all is correct.
 
#40 ·
Hint: the Lotus ECU learns and adjusts the long term fuel trims constantly. If the car runs too rich or lean at part throttle the ECU will trim for it. This takes a while. This DOES change how the car runs at WOT. The car might run great after the ECU is reset then too rich or lean at WOT after it learns for a while. Part throttle tuning is very important.
 
#41 ·
I haven't bought an iPhone OBD logger but I've some information to share. The fuel pressure gauge was extended to make it visible from the driver's seat. The pressure was solid (+/- 1 psi from vibrating needle) throughout the rev band @ 64psi. I've thus ruled out fuel pressure.


This past weekend, I had a 3rd O2 bung welded ~2 inches after the pre-cat O2 on the de-cat pipe and installed an AEM wideband sensor.

I experimented with increasing fuel pressure (tested at 64, 72, 78psi) which brought high-RPM WOT AFR's down to range of 11.0:1 to 12:1. This has not seemed to affect partial throttle response in any way yet :). In my humble opinion, my dyno graphs from today resemble Phil's graph from post #25. Here are the dyno runs from today:


I want to reiterate that the BOE surge tank was not installed until after the engine rebuild. This was my attempt at fixing the mystery problem and IMO has been worth every penny.

Here is a snapshot of my short-term and long-term fuel trims at idle (using OBD scanner):


Lastly, I paid an extra $50 today to confirm how inaccurate the tailpipe sniffer can be on a dual pipe stage 2 exhaust Elige. The tail-pipe sniffer read consistently 2.5pts higher/leaner than the correctly installed wideband O2! :crazyeyes:


I expect an iPhone OBD logger will tell me why I am having to compensate by setting very high fuel pressures.
 
#42 ·
I haven't bought an iPhone OBD logger but I've some information to share. The fuel pressure gauge was extended to make it visible from the driver's seat. The pressure was solid (+/- 1 psi from vibrating needle) throughout the rev band @ 64psi. I've thus ruled out fuel pressure.
Good one Johnny!
Nothing beat data to understand what is going on.

But it looks like leaner makes more power?
And it does not answer the question about why the first engine let go...

It is possible that there was a leak in the piping to cause tail O2 to be off, or is it really air getting drawn back in?, or are the sensors that much different?
 
#43 ·
Johnny very nice! Some real data.

Yup that test you did with the fuel pressure gauge pointed at the cabin is proof positive the surge tank is doing its job and you have solid fuel delivery (at least to the rail).

The fuel trims actually look very good there from the quick snapshot at idle... What fuel pressure were you running at the time? Once you get the BT adapater and app, you will want to log your LTFT (long term fuel trims), and STFT (short term fuel trims) as you drive. The ECU will learn across various rpm and load cells. So there is not single fuel trim value, but rather a map of them. It would be interesting to see how they run across under varying conditions.

And yea many of us figured the dyno WB readings were skewed. Very good to hear that you are actually running pretty darn good AFRs.

And holmz I actually want to point something out here... also pointing it out to other 2zz tuners. On the low cam there is basically no difference in torque between the rich reading and the lean reading. In the mid-range I know there are tunes out there that IMO run extremely lean. This particular car showed basically no difference in torque output between low 11:1 AFR and low 12:1 AFR on the low cam. Certainly for this car there is to push it and run 13s and 14:1 like I see some of the tunes run.

On the high cam no doubt about it more torque (and thus power) was found at high 11:1 as opposed to low 11:1.
 
#44 ·
Update! I finally received a wifi OBD logger ($50 from ebay) for my iphone and here are 7 logs (*.xlsx) of different style driving, FP set to 75psi, from yesterday evening:
Index of /~gtg928x/data_logs

AFR's dropped to high 11's last night for the datalog. However, 1 week ago on the track (Little Talladega) the AFR's were in the 11's and crept up to high 12's :wallbang: after four 20min sessions when I gave up and drove home. The ECU gradually learns itself lean no matter what I do. This is very similar to Phil's Katana smaller pulley escapade. Dropping the FP and bring it back up usually resets this cycle (rich, then gradually lean again).

From all my research and dyno runs, 12:1 definitely makes more power than 11.5:1 or lower. However, I'd rather run safe and lose 5 or 10 hp considering the money and time I've spent.

The first engine probably went kapooie because it gradually leaned itself out over the 3 weeks. I highly recommend an AFR gauge for any supercharged elise/exige.

The tail-pipe sniffer is definitely ghetto-rigged looking. Kind of like the sniffer during an OBD1 emissions test... but worse. Also the exposed 2nd muffler tip doesn't help.

WhatsADSM, the fuel trim photo I posted 1 week ago was w/ FP set to 75psi. The datalog from last night was also set to 75psi but gives much more information for different driving manners. Now that I have more data, any ideas?
 
#45 ·
...
From all my research and dyno runs, 12:1 definitely makes more power than 11.5:1 or lower.
...
That I would believe.
And it also would have a better BSFC.


...
And holmz I actually want to point something out here... also pointing it out to other 2zz tuners. On the low cam there is basically no difference in torque between the rich reading and the lean reading. In the mid-range I know there are tunes out there that IMO run extremely lean. This particular car showed basically no difference in torque output between low 11:1 AFR and low 12:1 AFR on the low cam. Certainly for this car there is to push it and run 13s and 14:1 like I see some of the tunes run.

On the high cam no doubt about it more torque (and thus power) was found at high 11:1 as opposed to low 11:1.
It gets complicated as the ignition advance needs also vary with A:F, so it is hard to just say that A:F should be X, without knowing the ignition advance.
It really needs to be a graph of torque in Z with A:F on X and advance on the Y-axis. That is a lot of data.
You could also have EGT in there too.

Once the WFO A:F and advance are set, then what happens at partial throttle is a function of whether you want efficiency, emissions, or other. So there are 100's of ways to skin the cat on that part of the map.
 
#47 ·
Jermaine reflashed my ECU with a custom map for hotter conditions. I've been driving spiritedly for a week now and everything has been great! AFR in the high 11's, partial throttle city driving is significantly smoother and responsive, and start-up and idle is perfectly stable! :nanner:

I dropped the fuel pressure back down to 53psi before removing the ECU to be sent out to minimize variables once the ECU was returned and reinstalled. Before this, when changing increasing/decreasing FP, the ECU would always learn itself lean again.

Seems like some power was picked up, but only a dyno in the near future will confirm it. Power definitely feels more consistent now. I'm a happy camper now. :D
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top