I was running Denso IXUH22 plugs. Two heat ranges colder with a .4mm center electrode as compared to the NGK's .6mm. They were gapped to .027 when I installed them but when I noticed the pre-ignition I pulled them to look at them. Come to find out they fire so hot with the smaller center electrode they had dished out the lower electrode and the gap had opened to .033. As an interim solution I put my old set of NGK ILZKBR7A8G plugs back in gapped to .022.

Right now I'm trying to decide between the Brisk QR12LS or NGK ILZKR8C8G plugs which are used in the Peugeot RCZ-R.

 

Interesting! But what is it about the Peugeot RCZ-R plugs that is much better than say the NGK 4771 or NGK 5992 plugs that we have easy access to? Those Peugeot RCZ-R plugs seem harder to come by.

 

On the NGK's plugs the number immediately following the R for Resistor identifies the heat range of the plug. So both the BR7A-G and BR7A-8G have a heat range of 7. The ones for the RCZ-R have a heat range of 8. The file attached is from NGK's website. I have always heard that the -8G plugs for the JCW were one heat range colder but this document from NGK appears to indicate otherwise. The Brisk -12 rating is comparable to NGK's -8 per Brisk's website so that why I'm looking at those two.

Typical rule of thumb is one heat range colder for each 75 HP increase as each increase in heat range allows the plug to be able to transmit 70°C to 100°C more heat to the cylinder head.

The Brisk's are much easier to get but I've heard some conflicting reports on them so I'm still researching and trying to decide.

 

I didn't see anywhere offering the R8 NGKs. I even looked on their website and it didn't appear to exist. When I looked the route to go for a colder NGK was to get DILKAR8A8 for the GTR.

 

Would love to find someone thats used them before. With the GTR plugs we lose the inductive type resistor for a straight resistor and lose the copper core so they might need more powerful coils. Okada coils are on my list but not for a while. The NGK's aren't available in the aftermarket yet so you have to call a Peugeot or Citroen dealer to get them at the moment.

 

Thats great info! I should really get some plugs then because im making close to 250whp on the stock jcw plugs. Maybe I should get a set of the 4771? I wouldnt mind getting the Peugeot RCZ-R plugs though instead.

Found them here

http://m.ebay.com/itm?itemId=271230025006

 

Tigger, good to know about he GT-R plugs. I was planing on using the Brisk. Before you get the Okada coils you might want to see what ra2fanatic's feed back on the
Ignition Projects that Battle sells. He's been running them with the Brisk combo and they're half of the price of the Okada. He's not one to shy away from spending money on quality parts.

 

Thanks for the tip. He seems pretty happy with the coils and the plugs so I went ahead and ordered the Brisk QR12LS from Sneed. Think I may go ahead and order the coils as well.

Did some more research and although I already knew about reduced dwell time at high RPM, I had never considered increased resistance across the spark gap at high boost. Higher boost = higher air density = higher resistance. I had always assumed the only reason we reduced gap was to reduce flame kernel blow out at high boost. It's the little details like these I'm still catching up on coming from a normally aspirated background.

 

I would think air would be a better conductor as it gets denser, I know that it will aid in combustion as you compress it.

I guess what really matters is what happens to its dielectric breakdown properties as it is compressed.

Maybe the decrease in gap is to also control the spark, i.e. just have it jump across the gap and not have it jumping around the combustion chamber.

 

(OCD Geek mode warning) Mode=on...lol

Nitrogen (N2) and oxygen (O2) the primary components of air are natural insulators as both are polar and made up of two atoms with a shared covalent bond. Even in liquid form they are not conductive. LOX can be conductive if placed under great enough pressure as it then assumes metallic properties.

If anybody's eyes are bleeding from reading that my apologies but I did warn you hehe.

From what I gather misfiring (not knock) in our application is primarily a product of three things. Reduced dwell time at high RPM (coil has less time to charge), flame kernel blow out (rapidly swirling air in the combustion chamber blows out the flame before it can spread), and increased resistance between the center electrodes due to air density.

Closing the gap should help with the last two and the IP coils should help with one and three.

Still digging around to find out if the apparent short term lean condition from misfiring can cause the ECU to advance spark. Seems like it would as the O2 sensor would report higher levels of oxygen that the ECU would advance timing to ensure a more complete burn. That's a WAG on my part.

 

Did some more digging around and discovered that our Bosch ECU's, both the MED 17.2 and MEVD17.2.2, contain a shift ignition retard table that is used in automatics only. This table contains a list of timing values in relation to engine load. From my understanding the values in this table are used to reduce the total timing of the engine during shifts effectively preventing pre-ignition and the subsequent pulled timing. What I do not know is whether this table contains the same values in the S and the JCW ECU's or if they are different.

I believe this may very well be the source of the surging problem common to some automatics when running high performance tunes. When reviewing my data logs it would appear that each time this has occurred it did so very close to shifting.

Once I get the new coils and plugs I plan on increasing the boost to record more logs for analysis. But my conjecture at this point is that only a re-tune and adjustment of this table will resolve the issue.

 

That is an interesting find!

 

Do you think that the ECU retards timing during a shift (for automatics) to make it smoother feeling?

 

Transmission preservation

 

The transmission is part of the reason. However, shift RPM and shift characteristics are controlled by the GSF21 ECU on the transmission itself. That's why RPM limits are not raised when tuning autos.

The primary reason timing is retarded during shifts is due to the operation of the turbo with an auto. With a stick you release the throttle while shifting, even if for a very short period of time. When this happens the turbine RPM decreases and must then spool back up after shifting. With an automatic there is zero reduction in turbine RPM as the transmission shifts. Suddenly the engine RPM drops while the turbine is still at full spool. So when the engine RPM drops and the turbine is still pumping the same CFM's you get a sudden spike in boost pressure of approximately .5 to 1 psi. During that window reducing timing is a very good thing indeed.

Second to third, and third to fourth gear.



BTW I ordered an OBDLink SX from Scan Tools. The first OBDLink I received was DOA and the SX is supposed to have a higher capture rate since there is no buffering circuitry for the wifi transceiver. Well see once it arrives.

 

The Brisk QR12LS plugs arrived from Sneeds Speedshop. Two plugs were gapped at .027 and the other two were gapped at .026 so set them all at .026.

Also received an email from Sneeds with a 10% discount code for my next purchase. Oh Happy Day! It's time to order some coils.

 

Any update on the obdlink sx? What kind of scan rates are you getting?

 

Unfortunately it is no faster than the wifi adapter. It appears to be a limitation of the standard OBDII query/response protocol and how many requests can be processed at once.

The good news is I'm working with a developer on something new. They currently provide logging software for BMW's that utilizes the K+DCAN cable we use with INPA/NCSExpert. This will provide much higher through put and give us access to data we've never been able to see before. Things like the knock sensor, vanos/cam position, fuel injector timing, etc.. I provided him with the MEVD17.2.2 DME code that's utilized in the N18's and he's working on it for me to beta. If it works half as well as I think it should be awesome. The next step will be to provide him the MED17.2 code for the N14's/JCW's and the MEVD17.2.7 code for the N18 JCW's so we can all benefit.

 

Wow, that is awesome.