It's ironic that there are lots of posts when we're trying to figure something out, but not so much once the deed is done. The EBL-ized '7747 on the 331ci SBC is still running fine.  :)

Technorati Claim post...

Ignore me.

Technorati Profile

TH400 electronic kickdown

My TH400 uses a solenoid to kickdown instead of the linkage method. Since this is controlled by the ECM (instead of TCC control) it needed new settings. RBob quickly responded with the following config which is working great. Posted here just to keep my notes together.
=========================================
What is interesting is that GM sets up the TH400 kick down calibration to use a set TPS%. Not based on MPH or RPM, just set to 85% in the case of ABHC. The thresholds are set up as the coast or decel TPS% unlock. Unlock (no kickdown) when the TPS% is less then the coast value. Then lock (do kickdown) when the TPS% is above the coast value.
So with that, this is what I found works for the TH400 kickdown (all TCC parameters). This is only bench tested, but looks OK:
Set Lock enable CTS to -40C
Set TPS% opening/closing deltas for unlock to 100% Set Relock delays to 0 (all) Set Forced Lock to 255 MPH
Set the 4 TCC TPS% unlock/lock tables to 255:
LoGr TPS% to Stay Unlocked
LoGr TPS% to Unlock
HiGr TPS% to Stay Unlocked
HiGr TPS% to Unlock
Set the 4 MPH thresholds to 0:
LoGr MPH OK to Lock
LoGr MPH to Unlock
HiGr MPH OK to Lock
HiGr MPH to Unlock
Set these 2 coast values to 85%:
LoGr Max TPS% coast unlock
HiGr Max TPS% coast unlock
Set these 2 coast values lower for hysterisis (probably mispelled), at 75%:
LoGr Min TPS% coast unlock
HiGr Min TPS% coast unlock
With these settings the TCC ECM output will go low when the TPS% is greater then 85%. The Relay is set up to provide +12 volts to the downshift solenoid whenever the TCC ECM output goes low (grounded).
RBob.

EPROM Emulator -- Ostrich

Since I've got the EBL, and a notebook pc to use with it, it became way too inconvenient to go to a desktop pc for editing of eproms. The Ostrich EPROM emulator by Craig Moates solves all of that. It replaces the EPROM and is fully programmable via USB port. Now I can monitor with the EBL interface and immediately update the tables in the controller. To top off the deal, Moates offers a discount towards the purchase of TunerPro editing software. I had been using "promedit" software to edit bins, but it had a limitation on the number of items it would recognize from a single .ecu config file. TunerPro doesn't have the same limitation and offers other advantages too. Very nice.

Engine hardware update

Been a while...cracked head interrupted EBL tuning. Those swirl port heads may have done something to create swirl (or maybe not) but they're weak castings. After 2 (!) cracked heads I've replaced them with aftermarket Dart Iron Eagle castings.

331ci Engine specs:
TBI unit (stock 1989 Suburban 2500 4x4)
trans-dapt TBI to 4bbl intake adapter
GM 4bbl intake
Hyd cam (.443I/E lift and 214d dur @.050)
Dart Iron Eagle heads (straight plug, 180cc I, 64cc chambers)
Speed-Pro Hypereutectic pistons (2 valve relief flat top - 327 + .030 over)
Stage II rods from Competition Products
350ci 2 bolt block
327 crank
OEM exhaust manifolds
Gutted cat
Dynoflow 3" turbo muffler

More ecu file updates

Ooops. My stripped down ecu file didn't include the "Choke" function.

Truck is running too rich during warmup, so I'll add those. Should be 62F,640,or 654 I think.

Note also that the shift light function at 8C7 has 2 ecu file entries so hand editing of the file will be required when that function is included. (Actually....I think I'll pull the 6cyl entry from the large file and keep it for inclusion if/when I ever do a 6cyl).

Working files

I'm using a folder on my website as I work on these files. See http://www.aces.edu/~gparmer/efi/ebl

Legend:
ebl.ecu is the corrected (no duplicate entry) version of ebl_v14.ecu.
ebl.awk is the awk script used to pull entries from the ebl.ecu file
060331a.ecu is the file created 3/31/2006 to edit a bin with promedit

On the Road Again

I think I'll drive this one a little bit and watch TGO for more info about the TCC logic and other descriptions. The WUD showed 14.7mpg average on my last trip. I expect that's for the single trip, but it is encouraging any way I study it...this is a big vehicle and the best it's previously gotten on a tank of gas is 11mpg.

FWIW--I was out of blank '128 chips when I created this bin, so I used type sub003.bin sub003.bin > sub003-dbl.bin to "stack" the chip with a double image. Seems to have worked fine. You'll notice from the filename that I've made very few chips. So far I've only noticed it being a little too rich at a spot during the warmup, and a little knock off-idle. The sputter at mid-throttle wasn't noticable this morning. It may learn itself away.

New bin

Used the limited VE learn info from today's trip to work to build a new bin. Unfortunately winbin didn't like the ecu file, so I didn't get to even out the maps. Went with the learn only. Also increased the idle speed to a min of 600 rpm, since that's about as low as I expect it'll run smoothly anyway.

sub003.bin Promid=3 Doubled the image to use in a 27256 eprom.

Road trip

Used my potentiometer/IAT and set the temp to 40dC. Let's pretend it's a bit warm.

Hit the key and she comes to life. The stinky flooded hydrocarbon smell is gone and she's running pretty well. Idle is at about 600rpm. A bit rumpy at that speed, but running OK for a first start.

Drive to work with datalogging on. Notice a misfire at 2000rpm and mid-throttle I'll have to tune out. Otherwise, she's running better all the time.

Will use this trip to create a new "learned" bin, then start on the TCC/shiftlight/kickdown logic.

Aha! ECU file revelation

Address 7DB is in there twice, with an extra blank line to boot. That messes up the syntax.

With that corrected, my awk script works pretty well at pulling out the proper variables and numbering the Items so that the file looks nice.

Used my custom ecu file to edit the bin.
331ci,55#/hr injectors
Burned the chip. Will test drive tomorrow.

Night 5,6,7

Spring Break. Worked at a site close to the beach one day and added a little beach vacation the next day. As nice as the Gulf Coast Beach (Gulf Shores, AL) is, I wish I were tuning.

Night 4

The IAT pin doesn't even have a wire in my connector. I put some solder on a stripped back wire and stuck it in the hole. Connected to a potentiometer and grounded the other side. Instant settable IAT. :)

Truck runs now without SES light.

Test drive shows proves that she'll run. The VE cell in use is highlighted as the engine runs. Very nice. The software will create a new bin with edited VE tables as learned. Very nice.

Start stripping settings out of the ecu file. There are bunches (322 as shipped) and promedit will only read 245 or so. Created a little "awk" program to strip addresses for a new ecu file. It's still having problems though.

Post to TGO message board so BobR will know I fixed the SES and his work is paying off...and so others can have some encouragement.

Night 3

Heck with editing the bin. Tried burning the shipped one. Truck cranked but SES light was on. The USB-to-serial cable I bought uses COM5. Once that connected and I found that the EBL's WUD interface shows error codes, I saw that IAT was low. Duh, it was not ever installed.

EBL Self Install

Second night:

Got the non-chlorinated brake cleaner and cleaned the goo as described. Used pencil torch and "played" the flame across the connectors to remove the daughtercard. Cleaned up a couple of pins which had excess solder using a small soldering iron. Installed EBL. Used dremel to cut out the case for access to the new connectors. The calpack has a line marking pin 1, and the eprom has a divit in the top. The EBL board has a divit marking pin 1 on each as well.

Tried loading the included bin file in promedit with the included ecu file. Bzzzt. No joy.