Getting wet feet....junkyard TBI

[Bob Valentine]

JW wrote:
"But again, I have found that for a daily driver where MPG is more
important than MPH, the closed loop logic will take care of everything. "

Ok, I'll put the stock 350 Chevy TPI binary back in the 403 Olds tomorrow.
 Will also get rid of those pesky tack welds under my dist cap.  Yep,
shoulda left a stock bin with loose dist. for a constantly variable timing
curve for best MPG.

I guess the tuning work I did tuning to get 23-25mpg on the highway was a
waste of time.    

-> Bob Valentine
-> bob at tecmark.com



At 02:26 AM 5/30/2001 EDT, you wrote:
>For starters, get the ignition module mentioned recently and wire it to 
>trigger from the pickup coil in your distributor. Then disable the
mechanical 
>advance in your Jeep distributor by removing the springs and weights, but 
>reinstall the springs to hold the timing steady. Others on this list
advocate 
>tack welding the rotor mount to the distributor shaft so that there is no 
>slop, but I have just put the springs back in and have had no problems. The 
>only time I can think you would have any slop is when decelerating hard in 
>gear, and in this case the rotor mount would have to outrun the distributor 
>shaft with the spring tension fighting this motion. Again, I think its not a 
>problem. Disabling the vacuum advance is done by simply leaving the rubber 
>vacuum line off and plugging the line.
>
>For spark tuning on a non-GM vehicle, I start by getting a copy of the 
>factory manual showing the vacuum and mechanical advance curves. I then 
>program Excel spreadsheets with the vacuum portion alone, the RPM portion 
>alone, the static timing alone, and the fourth sheet being the sum of the 
>first three. This sum sheet will be a table with MAP (same as inverse of 
>vacuum) as the x axis, RPM as the Y axis, and the spark value as the entry
in 
>the table. This is a 3 dimensional table since you have two independent 
>variables. This table is then programmed into the chip as a starting point 
>for say a Jeep. If you want to tweak it after that, its up to you, but the 
>spark table fine tuning is where the hidden power lies, that and getting the 
>WOT fueling right. By the way, there is no closed loop spark tuning. This 
>part of tuning is feed forward only. What you program stays that way even if 
>the engine runs crappy. The exception to this is the knock sensor system,
but 
>this only retards timing. It doesnt add timing when a timing increase could 
>probably help. The only feedback for spark tuning is in the seat of your 
>pants (or an accelerometer for performance tuning).
>
>As for fuel adjustments, start by getting a scantool and seeing where your 
>BLMs are at hiway cruising speeds on a flat road. Then adjust (ie reprogram) 
>the Base Pulse Width constant to get your BLMs around 124-128. Then check
the 
>BPW constant to make sure its not too far off from what it should be. What I 
>mean by that is the BPW constant is calculated from the injector flowrate, 
>the engine cubic inches, and a units conversion constant (see Tunercats help 
>file for the 1227747 ECM at www.Tunercat.com). Since the cubic inches ain't 
>gonna change (258 ci right) and the conversion constant is a constant, the 
>only variable in the BPW constant is the injector flowrate. The 4.3L 
>injectors were rated for 45 #/hr based on what I've learned and found with 
>actual injectors. However, some flow more and some less. So if the BPW that 
>gets you a good BLM is back calculated and the injector constant is more
than 
>10% away from the 45 #/hr number, then you either don't have 45#/hr
injectors 
>or your engine has a problem (in my case, my engine was worn out causing a 
>very high injector size to be calculated; what I think was wrong was all the 
>blowby I had required more gas than a fresh 360 engine would require; 
>therefore I needed a very high BPW to make my Jeep run; 139 should have done 
>it whereas my Jeep with AMC 360 needed a 170 BPW constant to run well with
61 
>#/hr injectors).
>
>Now assuming that the BPW that gets you a decent BLM range also yields a 
>reasonable flowrate for the injectors you are supposed to have (the color 
>bands on the top of the injectors tell you the flowrate), you now have two 
>choices. You can tune the hell out of it like most on this list insist has
to 
>be done, or you can drive it awhile and let the GM closed loop logic tune
the 
>system for you. I have found that on a stock engine for a daily driver, that 
>I don't have to do much of anything at this point except watch my BLMs
with a 
>scantool for a few weeks just to make sure the values aren't too far off. 
>What the closed loop logic does is maintain a 14.7 AFR at all times by
nature 
>of the O2 sensor design, and it adjusts the fuel tables via the BLM (block 
>learn mode). So no matter what the fuel table says, the BLM will steer the 
>system towards an AFR of 14.7. The exception to this rule is at WOT when the 
>closed loop logic is turned off. In this case, the logic looks at the
highest 
>BLM cell, which is assumed to yield a 14.7 AFR at WOT, and recalculates the 
>AFR based on a WOT RPM vs. AFR table. For performance gurus, the WOT RPM vs. 
>AFR table and the spark table are where all their efforts go. It takes a
real 
>good tuner and probably a dyno ( or good spark plug reader) to fine tune 
>these tables for performance. But again, I have found that for a daily
driver 
>where MPG is more important than MPH, the closed loop logic will take care
of 
>everything. I don't plan on racing my Jeep Grand Wagoneer. I just let the GM 
>stock WOT settings govern here.
>
>In summary, this is the basic start for reprogramming a GM EPROM just to get 
>the system running well. To get it running perfect, you have to go much 
>further. Also, keep in mind that this is a simplified explanation and not
all 
>exceptions to the rules are covered here. JW
>----------------------------------------------------------------------------
>To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
>in the body of a message (not the subject) to majordomo at lists.diy-efi.org
>
>
----------------------------------------------------------------------------
To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org