slo

iTrader: (1)

I've seen lots of comments regarding people getting 13B's to idle much leaner then normal, at least into the mid 13's

whats the secret?

Negative split?

batter map?

end of injection map?

Zero throttle maps?

crispeed

Some details of your setup would help.

slo

iTrader: (1)

Halwin 1.81 beta, running a haltech E8

stock (550) primaries, 13b-RE irons, very mild porting, otherwise stock and in decent shape.

for plugs, 9's all around, got some one per plug old accel super coils running off stock stock ignitors on the leading, with a crane cams DI2 on order or should I say backorder.

Everything else stock 13b-rew parts (turbo's etc) nothing else that I can think of which would effect idle mixture

RETed

Negative split, although I dunno if it applies to idle also.
It's more like light load areas.
There's a good thread on this - try do a search on "negative split".


-Ted

crispeed

So what's the present idle afr?
Ignition timing at idle?

slo

iTrader: (1)

the highest afr I could get a steady idle was 12.5.

I fixed that and got a 13.2-13.7 idle.

The battery correction map was turned off, I turned it on and played with it until things improved some watching the haltech voltage.

I set about 18* advance around 1000 RPM with 6-7 degrees negative split, with this I was able to get it to idle leaner but still had some fluctuation.

Next I set the haltech rpm points to the low range (will never see 9K rpm anyways)

then I used the afr target table to slope the idle AFR's from about 12.6 to 13.8 between 500-1500 rpm and turned on closed loop idle off the wideband (which works very well by the way). With the engine warm, but most of the correction maps off or negigable, I used the idle control trim setting to move the idle up and down, and adjusted till the wideband correction was less than 1%.

Now around 1K rpm which is where I like the idle (I have a real quiet exhaust) I am seeing about 13.4-13.6.

Now with the air pump turned off (CA sucks, I do have a CAT) the exhaust doesn't smell at all in drive thru's and such. The cat I'm using is the metal type OBD 2 cat which don't normally use air injection.

Though more than likely the car wouldn't actually pass smog with the air pump turned off, I'd bet it would be close

The reason I like having the airpump just off is because I'm wanting to keep the cat as cool as possible to make it last as long as possible, turning the airpump on even at idle bumps the exhaust temp (cheapo pyrometer pointed at cat) from 900 degrees too beyond the pyrometers max of 1200 degrees. I know that the pyrometer is way off, and the cat is hotter than 900 degrees but cooler is better.

crispeed

You seem to have all the basis covered. Try retarding the timing at idle and see what happens. Port timing will effect how much retard you can have at idle. Since you got a pretty lean idle allready try and increase/decrease the 'end of injection' angle to make the mixture richer and then proceed to lean/lower the main injection pulse further at idle. A lean AFR does not necessary mean you're getting the best economy or using less fuel. AFR is just a ratio and injection pulse vs rpm is the actual indication of fuel volume being used. The lower the pulsewidth the less fuel being used. One of the benefits of sequential injection. Also on most rotaries this does not have a dramatic effect as when comparing to piston motors especially the closer they are to stock port timing but still give it a try and you'll see what happens. Don't get to carried away also because there's a happy range for the ecu to be able to put the fuel in on time.
You know the cat has an operating range. Too cold is just as bad for life if not even worse.

slo

iTrader: (1)

Thanks crispeed,

I was able to adjust the end of injection map (only played with it for about 10 minutes per your instructions), and pulled about 6 degrees of timing out at idle, also adjusting the negative split to 6-7 deg at 1K linearized to zero split at 500 rpm with less timing also at 500 rpm linearized up to 12-13 advance at idle.

This gave me a rock steady idle at 13.7-13.9 AFR on open loop, I also adjusted the idle sans BAC back to 600 rpm (it was at 900, with the bac bringing it up to 1K) this made the idle just rock steady. I am not using an input for the ac, so the computer doesn't know when I turn the ac on to raise idle, but there is only a momentary (un-noticeable) 100 rpm dip before coming back up to 1K rpm.

as a side effect, at 1K rpm, my vacuum has gone from 15-16 inhg to 18-19 in hg

When I say mild ports, what I mean is I took a jspec 13b-re apart and noticng the lip behind the port I just smoothed this out taking the port up and down about 2MM or less. I made the primary port have the same timing as a 13b-rew, but its out slightly wider (because it was wider to start with). I also used t2 style exhaust inserts, and just smoothed and beveled the exhaust port a tiny bit, little actual change in timing.

I did a minimal cost rebuild, the only thing I replaced where the apex seal springs (the seals had nearly new spec) and obviously the gaskets and such. The engine couldn't have had more than 15K on it since the last major rebuild the housings where all in great condition, as where all of the other parts. The irons looked like they had just barley been broken in.

slo

iTrader: (1)

one more thing, is there any benefit to adjusting the end of injection map at other than idle?

crispeed

Yes for the very same reasons not to mention some power gains!

slo

iTrader: (1)

I would think that once the IDC goes over 50% and the engine starts essentially no longer running sequential injection that there would be no gain with having the sequential injection have spot on timing.

I suppose it does make sense, if you inject at the right time you can minimize or eliminate fuel lost to overlap, which is clearly why a rotary idles so ******* rich.

I wonder if using really large primary's and or medium primaries with higher rail pressure if you could truly eliminate all fuel lost to overlap in the under 3000 rpm areas before the ve levels off and gets good.

There's a chance you could make an older 13B idle like a renesis at dead on stoic.

Really makes me wish mazda had gone to direct injection with the rx8 and kept the peripheral exhaust ports.