It was the socket booster.
Moates was kind of enough to desolder the old one for me, solder in a proper socket, and include a new booster that allows data tracing at no charge other than the RMS shipping. I've had some trouble sorting out their products, but they are complicated and they've tolerated my failure to follow instructions.
The ECU only hits the first fuel table.
The new socket booster worked well enough, but read the tables differently somehow, causing a lean condition during cruising situations when a closed loop system would trim back to stoich. This is what was going on all along and it wasn't good. However, it was possible to trace through the booster and use the Ostrich to tune away the lean spots pretty quickly while my wife drove around our rural neighborhood. So that's reason enough to use one if your ECU won't talk to an Ostrich any other way, but otherwise it's easier to leave it out if you can get away with it. Apparently YMMV is an understatement, as there are many that apparently need one that are based on the same processors as the Swift.
Once the booster was gone, however, the 210/340 chip maps worked exactly as I had expected them to in the first place. I still found a few cells where I still decided to change things, but there were only minor changes and I didn't see any AFR's to get too worried about.
Here’s what else I’ve learned:
I still don’t know where the closed loop switch is.
I found that by keeping only the values that were different from both skba103 and skba 105 and keeping the latter when they weren’t, I now have a tune that acts like Sandro’s but has closed loop operation just like stock when I plug in the simulated narrowband, or runs in open loop when I unplug it .
Here’s a big one: The fuel table is limited to just over 6,000 rpm in any tune, so I expect that’s one reason you’d want to go standalone if you went turbo. There is some way to monkey with the way the map is “stretched” over the power band though, because the trace on the 210/340 maps appears to jump to 2000 more quickly and therefore devotes more space for tuning the rest of the powerband up to 6k. After that, load may fluctuate across the bottom of the map but you’re stuck with last row. Load increases from back to front and left to right in the graphs, which is not was I was thinking.
The ECU won’t trace ignition properly, with or without socket booster L It skips all over table 2. I don’t know that I’m qualified to mess with it, anyway, wish I had a knock sensor. Sandro’s map seems to work fine on premium, so I’ll keep it but still wish I had a better idea of what going on. I don’t hear what experienced tuners say they hear.
_________________
Quote:
No one expects the Spanish Inquisition!
Spanish Inquisition Racing chip burning service--build yourself a custom chip!
http://www.teamswift.net/viewtopic.php?f=17&t=57216
Login
Register
Gallery
FAQ
Search
.
