naishou

I have been very seriously considering getting a performance kit for my WRX from APS that uses the Unichip. It sounds like a very good piece of gear and has been used by many successful big name racers in Australia and South Africa.

To the poster who said they wouldn't trust anything from South Africa, what an ignorant comment! Go wash your mouth out. The South Africans are some of the smartest and most practical people on earth. There's no BS about the Unichip.

The only "problem" with the Unichip is that many stock ECUs including the WRX and S2K will change their maps according to temp or whatever, and also "learn" based on things like knock sensors. My take on this is that it's not a "problem" at all, but rather a good safety feature. If you're causing knock you've adjusted things too far. The main use of the Unichip is for modified engines that need revised maps to reveal their new power potential. It's a feature of the Unichip that the factory idle characteristics and safety mechanisms are retained. Sometimes you can get more (permanent) power out of a stock car using the Unichip because the manufacturer has been overly cautious, mostly with turbo cars. It allows individual tuning to your car's characteristics but must be done on a dyno.

I'd be very keen to see the results of a good Unichip tune on an S2K, particularly after a month or so installed.

shingles

But at WOT, the maps for WOT don't change or learn... isn't that what's more important? the WOT maps, because after all, you should be WOT most of the time when you are racing...

-Shing

naishou

You're probably right there, but you do maintain knock sensor operation, so if the car's already on the verge of detonation there's not a lot to gain. I've seen some dyno charts for VTEC Hondas that do show gains with the Unichip, and a smoothing of the VTEC transition. I'd love to see the results on the S2000.

ultimate lurker

The only place any _fuel_ tuning really needs to be done on any NA S2000 is at WOT. The ECU will adjust mixtures according to the O2 sensor while at part throttle. If you have really hairy cams or massive flow gains due to modifications, adjustments, even at part throttle may be out of range for the ECU and trigger a code, but by my estimates you'd have to be producing on the order of 280-290 crank hp before that will happen (here's to hoping I someday do :-). So whether you chose a Unichip or VAFC to do fuel correction, you wouldn't want to worry about the light throttle areas - from a fuel perspective.

The VAFC allows you to choose where Lo and Hi throttle transitions take place. If these two points are not the same, it interpolates between the two. I would set lo throttle to be relatively high (say 50-60% throttle opening) and hi throttle in the range of 70-80%.

I like the fact that the Unichip allows you to adjust ignition timing. Adding some advance at high rpm would probably net significant power for us. It appears that more advanced versions of the unichip can also control VTEC, although apparently not as easily or with the simple corrections provided by the VAFC (looks like you'd need one with solenoid control and then setup extra maps for the VTEC range).

The only real question is, how much does a decent Unichip system cost? That would be the determinant for me. Under $1000 for a full kit would be nice. Any more and I'd start thinking about a full replacement system.

UL

naishou

[QUOTE]Originally posted by ultimate lurker
[B]The only place any _fuel_ tuning really needs to be done on any NA S2000 is at WOT.

ultimate lurker

I can't offer a wit of comment on the WRX since I don't anything about their ECUs.

On the S2000 however, the ECU has a transition point between closed loop (reading the O2 sensor) and open loop (working solely off maps and long term fuel trim numbers). This point is largely throttle position sensitive, although from empircal observations, load appears to play a role too.

When you're cruising along at light load, the ECU reads the O2 sensor and proceeds to implement what is known as a 'bang-bang' adjustment regimin. That means that it will continuously adjust fuel up and and down till the mixture goes out of spec in one direction and then adjust back in the other direction. On an o2 sensor voltage readout, this can be observed as a rapidly swinging voltage within a confined range. If the O2 sensor is in good condition and nice and hot, the swings occur several times per second.

When you pass the WOT transition threshold, the ECU snaps over to the WOT maps. Thus, alterations can affect the performance here. However, you are altering these maps with the VAFC as well, so any tool that alters WOT maps should give you similar effects.

UL

tze

U lurker is correct, All honda ecus work on 2 loops.

The unichip will let you operate on both loops, and more importantly, correct multiple zones on the closed loop (non WOT) positions.

When you are racing - ie. not just in a bloody straight line, you would be using various throttle positions all the time. Hence it is important to be able to modifiy these.

However, U lurker, even though theoretically at closed loop the ecu is supposed to retain stoich proportions, it doesn't. It seems to work off a base map which is still slightly rich etc.

In the old integras with distributors, a manual adjustment of the distributor to increase the timing would do so at even closed loop. It seemed like the distributor position gave the ecu a base line.

I am very interested in turkey's comments that the motec s2k has been damaged?? Any further info??? Was it detonation?? As I said, a good ecu tune with an electornic retardation device - that "J whatever its called" argh someone help me out!!!!!!!!

ultimate lurker

tze,

We need to be careful we're not confusing the behvior of fuel and ignition timing on Hondas. The ignition timing is based solely on an index signal (on older cars it was the distributor, now based on crank position), load and rpm. The ECU selects the appropriate map point and using the index signal as a reference, triggers the discharge. The only adjustments made to the stock timing curve (by the ECU) occur for varying temps or when knock is detected. Thus, there is no ongoing closed loop feedback on timing. This is why its o.k. to modify timing - although with modern Honda ECUs, if the timing seems funky, it can try and correct (ran into this on OBDII Preludes) which is why you'd want to intercept the CKP input signal to the ECU IMO.

The fueling OTOH is constantly being adjusted from the base map according to O2 sensor feedback during closed loop operation. If watch the o2 sensor voltage, the mixture tends to oscillate between slightly lean to mildly rich. Its the only way to pass emissions and get good fuel mileage.

In terms of altering the closed loop sections of the fueling map, again, I say there is really no point because the ECU is going to strive to meet its stoich (or near stoich) goals usin hte O2 sensor. All you'll succeed in doing by altering the part throttle maps is altering the trim numbers.

Oh, and the tool you're thinking of is a J&S knock sensor - although the MOTEC has a knock sensor option as well.

hecash,

10 and 50 seems to be what everyone uses. However, I don't want the ECU to even get an inkling of the full throttle changes I want to make. Along with the fact that I wouldn't make any lo throttle fuel corrections, I'd set the numbers higher.

UL