Is it IMPOSSIBLE to get rid of tip in detonation?

[Holden]

AvanishV: If you want to see it, cruise at 3500-4000RPM in 4th gear. The A/F will be doing its' normal dance. Then punch the gas. You will see the A/F go so lean NO lights are on. This only last 1 second or so, but apparently if you are on the edge, whether by increased boost, cheap gas or advanced timing, you get detonation doing this.

Greg

[BlownAway]

Is it necessary to get rid of tip-in (pre)detonation? What harm will it do?

[Four Door GSR]

I don't get any tip in detonation on my 94 GSR. When I step in the gas, the A/F meter instantly goes in the rich region, like it did before I was FI.

Maybe its a OBDII problem.

Anyone with OBD1 suffering from tip in detonation?

[rjaffey]

It doesnt look like the Cartech is reacting any faster. I saw similar responsiveness with the JR regulator by both the A/F and fuel pressure gauges. I'm also using the supplied restrictor and check valve (a must for the Cartech to work right).

On sudden light throttle application, it goes rich first then goes back to maintaining stoich. On sudden full throttle application it goes rich and stays there.

At part or full throttle the J&S *occasionally* takes out 3 degrees at most in the midrange and that's with 12 degrees initial timing. Not all cylinders retard that much either and only on rare occasions do I see even 1 degree taken out past midrange. Perhaps this is due to higher cylinder pressures prior to VTEC cutover or more agressive ignition timing in that RPM range.

About the TPS sensor idea, check it out:



This may wind up using unused poles in the existing JR relay, or perhaps the JR relay gets replaced with one that has more poles if necessary.

This should tell the ECU that the throttle is wide open after 1psi (point where pressure switch opens) of boost is reached regardless of the real throttle position. In theory the result is fuel delivery based on MAP and IAT signals. The O2 sensor's out of the picture and the ECU will no longer be trying to maintain stoich mixture. A side effect may be that a J&S Safeguard will arm sooner than it would otherwise, but I'm not sure about this. It may arm from the TPS signal before there's enough boost to open the pressure switch, de-energizing the relay, in which case the J&S operation won't be affected in any way. I'm not sure it matters either way.



[This message has been edited by rjaffey (edited June 11, 2000).]

[Holden]

Isn't this going to cause me to run really rich at part throttle (1-5psi boost)?

Greg

[Turismo]

I read the post about Splicing into the TPS with input from the pressure switch and it made sense to me from the get go.
Especially with a boost dependent fuel controller (PMS) it seems you could completely tune out the tip in problem.
I seem to only see this detonation when a good amount of throttle is applied but not enough to kick it out of closed-loop.
Since I ditched the JR pressure switch-relay in my setup I could use it solely for this purpose now.
I want to try this setup but am a little confused on how exactly to run the wiring (i.e. which wire from the relay to which TPS wire and does it matter where the 12v signal comes from???). Rjaffey's diagrams seem to confuse me (as usual ) but if someone could put it laymans terms I will try it and report the results.

[zip]

To keep the J&S TPS signal unconditioned just tap the J&S blue wire prior to the relay. Therefore, the J&S still recieves the proper TPS voltage.

[zip]

How about this Rjaffey:



I'm a EE, I just graduated a couple of days ago. I would approach this project from a microcontroller standpoint but it may be too elaborate and costly. However, this schematic will work as prescribed by your theory.
Since I have an Si I had to generalized the schematic since the wire colors maybe different from the GSR.

When you have pressures below whatever pressure relay activation you have, the ECU will recieve the correct TPS voltage signal. As soon as the pressure sensor is activated (turns off the relay), the TPS voltage is then conditioned to a WOT voltage. This drawing may have mistakes because I drew it in haste, but you'll get the gist of it.

BTW, if some of you haven't noticed this schematic requires two different relays.

One correction, the normally closed relay should be connected to the 4.5V signal instead of the TPS signal when the relay is off.

[This message has been edited by zip (edited June 11, 2000).]

[rjaffey]

Good call Zip. This also has at least four other potential problems that I can think of:

1) If the pressure switch is disconnected, the ECU will always see WOT.

2) If the relay is unplugged, the TPS signal will be disconnected altogether.

3) I'm not sure how the ECU will react to the abrupt transitions in TPS signal voltage, especially the momentary breaks as the relay makes its transition.

4) For those installations in which the relay oscillates rapidly under some conditions, this might totally confuse the ECU and lead to error codes and/or driveability problems.

So, this trick might still be worth trying, but may require some more creative engineering to work well. I'll think on this one, any EE's on this board care to chime in?

As to the cofusion, I'll add OBD I (maybe same as OBD II) wire colors to the diagram. The 12V and pressure switch leads connecting to the relay coil are already in place as it's the same relay used for the temp sensor trick. As to which relay leads to use, I'm expecting to continue using the same relay, just spare poles. I haven't looked at it closely, so I can't answer that just now. It is possible that the relay would have to be replaced or a second relay would have to be added.

Holden, my understanding of the TPS signal is that it's *not* used to control fuel delivery as a function of throttle position.

[lundgren]

This is an issue I've been working on.

Mostly, what causes detionation at certain points. I have NO detionation for tipin, UNLESS it is right at or around the VTEC crossover. WHICH doesn't matter with the V-AFC I have, if I move it, detionation follows.

So I think the ECU is adding timing at that point. It maybe too there is noise there too, and it's reacting to noise, not real detionation.

With the V-AFC I've managed to tune most of it out, but there is some left. I'd like to try a timing controller, but I think I'll leave this up to the next version of the J&S

------------------
http://www.netbsi-kc.com/integra/
http://www.netbsi-kc.com/aa/
http://www.groupbuycenter.com/