I will have to do a bit more coding to free up the non-cat base addition map, as it is mixed up with my big WOT map, but it is doable.
Sport Mode should work fine as it just adds to the stock map.

I bought a couple of sets of these injectors way back when because the top color matched the stock tan S3 (and S4) injectors.
I haven't found any more detailed specs for them: 0280150991, F5TE-C1A, tan top (24lb. 250cc/min 14.6 ohm high impedance)

I set a middling injector open time for idle and cruise. For WOT, I switch it to a more advanced time (note inj. adv. richens).
It's possible some other injectors could have idle quality issues, but I expect that most will work fine when the ROW CO pot is adjusted for mid-14's at idle.


Octane Adaptation should work regardless. At the moment I am deciding if the long term adaptation should be kept in SRAM.
I'm thinking of keeping it until the coolant temp drops below some threshold, so that it doesn't need to relearn if the engine is turned off for a short time.

 

Both the LH and EZ use 8052 processors (technically 8032 w/external ROM ), so they have an additional 128 bytes (8-bit) RAM vs. a 8051.
They both have an additional 256 bytes of external SRAM (which is kept as long as there is battery voltage).

The LH doesn't use any of the 8052 extended RAM and not quite all of the regular RAM.
It uses almost all of the SRAM for CO/idle adaptation and diagnostic variables.

The EZ uses a lot of vars and flags! It uses all of the regular and about half of the extended.
It uses about half of the SRAM for diagnostic functions only.


There's enough unused RAM and SRAM on the EZ to make both short and long term, 64 column, WOT adjustment (retard) tables...
Meaning it could essentially generate it's own WOT maps for individual engines over time.
I'm not sure if this is necessary, but it would be cool.

 

Cool, close enough to not worry then..

Your matching ones: 0-280-150-991 24lb, 252.2cc/min, 181.4g/min
Ford Racing: 0-280-150-947 24.25lb, 254.9cc/min, 183.3g/min

Just about 1% difference, which is what a flowmatched set is considered to be in terms of variation. So I"m good.

You just wanna be able to measure it on the dyno

Seriously tho - how fast would it un-adapt if during a club drive, you were forced to put in lower octane due to availability?

 

OK, maybe I'd like it to remember the last level for some interval...for dyno runs.

Currently, it starts advanced (every key on), then retards by the knock count, so going to a lower octane is fine.
I'm trying to figure out a procedure to reset it if you put in higher octane - besides unplugging the EZ.

You'll appreciate, perhaps, that I'm using the Australian (89 AKI/95RON) cruise map as a base (regardless of the coding), but adding a bunch of advance to it.


Note you'll be able to wire a 5v LED to the knock output pin on the diagnostic plug to watch the knocks fade out.
One of the last things I want to do is keep that light on all the time if there is a knock or hall sensor fault.

 

The knock sensing is perhaps even essential with your chip as some of the cars may have very bad carbon build up. Some injector-, valve cleaner and other
trics may be good idea before switching the chips..

Regarding MAF signal stability, I'm quite a sure it comes from he fact that airbox floor is too flat and turn to MAF input is too tight.
Perhaps the best solution is to modify the airbox floor the way that it starts to turn down eariler and turn is smoother.
Wonder how much space there is to lift up the airbox that it still dosen't hit the bonnet..
I'm having yet another stupid idea to make the turn smoother, but perhaps switching to Alpha-N at same time and throw MAF away.
This will give more space to make smooth turn for the floor and remove highly restrictive component from intake tract.

 

I think the lower airbox profile may add some turbulence noise to the MAF output which is amplified by the A/D process.
Even with an apparently clean MAF voltage log, though, the LH A/D frequency converter seems to make some random adjustments.

One of those lucky folks with a 3D printer should crank out some toilet-bowl-wax-rings to fit (silicone) onto the lower airbox opening...


For persistent Octane Adaptation, I made the coolant temp cutoff at the upper edge of the first white block (ca. 130f/54c).
Or, if the pedal is pushed a little bit when the key is turned to position '2', any stored retard will be cleared. (Release pedal before cranking.)
If the engine has cooled or the pedal was pushed, the knock light (if installed) will light up to show that OA has been reset (and that the LED works).

 

I bought a medium format printer (12x12x12) for April deliver and have a deposit on a big one (24x24x24) with a June delivery. Airboxes will be on the menu. I can print something now, but I can only do ABS.

The medium size unit is a very high heat model, hottest out there I know about, and will allow polycarbonate, nylon and other polyamide filaments suitable for underhood temps.

If you want to test something now with an ABS part, send me a file or sketch, and will pop one out for you.

 

I do not think that is correct. The wild-card is injector latency (dead time, opening-time, whatever you want to call it). Flow specs are always static (wide-open) flow but injectors in real life are pulsed.

Typical injector pulse-width (Tinj) at cruise is around 2.6ms, total. Stock 19# injectors have 0.95ms latency, so around 1.6ms fuel. The '947 Ford injectors are a little quicker, other Bosch injectors are slower- up to 1.6ms for some.

An error of 0.2ms in opening-time is a couple of AFR units. That's not close enough.

Short of testing, or a Sharktuner, I don't see any choice other than running the same injectors that the maps were tuned for. Adaptation won't fix it.

Cheers, Jim

 

This ^^^ ...

... plus this ^^^

There is no room to jack up the air box, but without the MAF there is room for a very nice transition into the throttle body. I don't think Alpha-N itself adds any horsepower, the turbulence certainly goes away but I am not convinced that it is anything other than an optics problem. But picking up 5" of intake length could be very welcome.

Hans, email inbound.

Cheers, Jim

 

There appears to be plenty of room under the airbox around the MAF to lower the floor around the MAF opening to create a better bellmouth.

FWIW, the airbox opening at the MAF screen is 85mm, the edge of the radius is 4"/100mm. A fat 4" id o-ring, cut in half wouldn't hurt?

My code has effectively made turbulence a non-issue, but more flow would be great.

 

Here's a quick concept:

 

I think you'd probably want about two pipe diameters of straight pipe before the MAF to get a very stable reading. If one doesn't have room for that, I think a honeycomb cell at the MAF sensor inlet should work, right?

 

Surely just replacing the airbox with something in the style of DR's Blackbird intake would be easier than screwing around with the airbox?

 

PorKen --

Can you reprogram the LH to use different fuel table for cylinders 1, 4, 5, and 8 vs. cylinders 2, 3, 6, and 7? If you can, then there are some interesting new options. The runners 1, 4, 5, and 8 are slightly above 20cm long and the runners 2, 3, 6, and 7 are slightly under 30cm long. The two sets would need different fuel tables and, possibly, different valve event timing.

If you can figure out how to trick the LH to run in such a "schizophrenia" mode, then it's going to be something like "...you take the red pill – you stay in Wonderland and I show you how deep the rabbit-hole goes."

 

That is true sequential unless he can make the batches line up with those cylinders - which would probably involve the ignition side.

Todd did this, but with true sequential.