Pulling timing during staging via Knock sensor possible?

[powermaxx]

Okay, a little back ground first. I need to get more power at lower rpms (Pat's converter is a 4200 stall but I can't get there easily with my G head combo, 3100rpm MAX!). Once I get to 3500 the thing is a wild animal! But it takes several seconds to get there, if it gets there at all and that's a lot of heat and strain I don't want!

So I'm thinking of ways to pull timing on the line (quickly hopefully). being at 5800ft altitude doesn't help either.

So I'm thinking the knock sensor, I may be able tap into the line and send a false signal and pull about 10 degrees or so (may need to adjust it so keep that in mind)? So what kind of voltage does it send? Is it straight DC or is it some sort of "Alternating" signal? How does the LM interpert this and will it do other things (I have no idea what that may be, just covering the bases..). How quick will the normal timing curve come back in? Etc... I run a manual boost control for right now and was also thinking of possibly using the MAP sensor as another way of fooling the LM.


Worst case I have an MSD 6 and a MSD 3 Stage Retard box I can throw in there.

I was going to pull the auto this weekend and go back to the 5sd for the rest of the season but this may solve that issue and save me from doing it again (trannys been pulled 3 times in the last month and I'm getting tired of that...)

I want my 12.2 runs back!!! The best I could do with this combo is 13.8's and that I ain't good enough!

[Frank]

It could be possible, but there are a few things to consider. What computer are you using? It would probably be easier and more consistant with recalibration. If going the knock sensor route, you need to find someone who has the knock vs rpm table. This will allow you to know how much voltage you need at what RPM. You can potentially just add a baseline voltage that is injected onto the knock sensor. Depending on the calibration in the car, it may pull off too much timing. I think if you didnt want to wait, you could inject 2 volts.

What turbo do you have on there?????

[powermaxx]

I'm using an 87 LM (T2) with a custom cal from mw6886 (Matt Wright). He was kind enough to give me cal (:thumb: :thumb: :thumb: ) but I haven't heard from him in a few weeks (I know he's busy ;) , maybe he'll see this...) but all I got was the bin file so I can't get into it (and don't want to! Not experienced enough yet to fool with it!).

I've seen the launch rev limiter built into the cals but don't trust our speed sensors to always work. So I'm thinking some sort of setup could be made to work within the cal that is either switchable (kinda like Paul V's 2 stage boost setup) or maybe use that speed sensor deal but change to timing vs rpm limiter. (You experts can tell me if I'm off base here.)

I run a T3/T4 hybrid from TU (works great!).

If I run the MSD setup I'll have to put a rev limiter in addition to my launch limiter (doable) but if the cal can take it I'd rather go that route for now.

[mw6886]

Hey, I finally saw this?! :) Having a kid can make you pretty busy. :thumb:

We can probably work out some details - maybe give you a switchable timing curve? That could be possible, but I think maybe we should try out the staging mod. That has worked flawlessly for me and a few others (Reeves for one.) Maybe we can put it at your stall? Dunno much about autos, so it will be trial and error. :confused:

Send me an email - I now have a little free time. :amen:

[ShelGame]

What I've been thinking about is an additional timing table that would reduce the advance based on how far away MAP is from the target boost. But, you'd pretty much have to use the Boost from TPS table to keep from getting a ton of retard at part throttle when you're not trying to spool up the turbo. Maybe I should just write an anti-lag routine...

[GLHNSLHT2]

Rob, Jason Richardson's 2 step routine jumps to a different timing table when it's activated. You can only go down to 0 but every bit helps. Maybe your staging cal can do the same thing?

[ShelGame]

That's a good idea. But, why can it only go down to 0?

Actually, I'm thinking about making it so that it will work both ways. It will pull timing when transitioning from P/T to WOT (the timing pulled will be proportional to how far away from target boost you are). I think what I can do is, when the staging limiter is active, I can ignore the transition condition (P/T to WOT) and just pull the timing anyway.

[contraption22]

That's a good idea. But, why can it only go down to 0?


Because the PCM is only capable of advancing timing beyond base timing. It cannot physically retard the distributor. So even when the scanner reads "0" electronic advance, it still has X degrees BTDC of advance, depending on where your base timing is set.

[mock_glh]

Because the PCM is only capable of advancing timing beyond base timing. It cannot physically retard the distributor. So even when the scanner reads "0" electronic advance, it still has X degrees BTDC of advance, depending on where your base timing is set.

Wait a minute. Your computer cannot advance the ignition to fire before the signal from the sensor. You can only delay the firing by controlled retard. Your distributor (or crank) are set to fire the spark at maximum advance. Your computer then delays the spark as necessary to give the proper timing. That 12 degrees inital advance we do is actually a programmed amount of retard from the true setting of the distributor.:nod:

[powermaxx]

This is very cool!!! Keep it going guys!

[ShelGame]

In the code, if the calculated advance is negative, it defaults it to 0. I just wasn't sure why. I've never really understood the reason for our "base" time to be 12*. I think I'm starting to, though.

BTW, I'm working on a fully relocatable source for SMEC's based on the '89 T1 code (similar to BlueBerry), but with some significant upgrades. I've made most of the code hacks (well, they're not hacks anymore) integral and each can be enabled by an option flag - much like the A/C can be enabled/disabled in the stock code. Coming soon!

[contraption22]

Wait a minute. Your computer cannot advance the ignition to fire before the signal from the sensor. You can only delay the firing by controlled retard. Your distributor (or crank) are set to fire the spark at maximum advance. Your computer then delays the spark as necessary to give the proper timing. That 12 degrees inital advance we do is actually a programmed amount of retard from the true setting of the distributor.:nod:

You're right. I'm dumb. :thumb:

[Dave]

I'm not positive as I don't have my notes and google all of a sudden sucks ;), but I remember through electrical class that it is an active sensor generating AC voltage anywhere from 1-4 V depending on the application. I do not know our voltages. Wish I did...

The idea that concerns me with you applying voltage to the knock sensor to attempt at triggering it may not work (although I like where you are going!). You need not only voltage but Hz. I'm 99% sure our PCM picks up on Hz not voltage. So you may need find out how many Hz the crystal generates and at what magnitude does it vary Hz to correspond with a reference signal to the computer so it understands it is pinging.

Also when it pings, I believe our PCM only pulls timing and nothing else. It will retard a maximum of 2* per every 2 rotations of the crank.

I hope that helps. Please, correct me if I'm wrong. I'm young and still learning. This is also interesting to me because I would really like to parallel a digital voltage gauge in my cabin to read the knock voltage while driving. Just to be more informed.

-Bryan