some pics of stuff.
The FC Tweak is picky about data QUALITY, which is kind of a double edged sword. you get better results, but sometimes its hard to get an old car to have better signal...
on mine i got to skip the voltage being erratic, but the O2 sensor was noisy. after some screwing around, i mounted the wideband unit right on the Datalogger, and i twisted the wires, wrapped them in foil (matches my hat!), and used a shielded ground cable, which worked.



it lived up under the dash on two unused tabs, not sure why my thumb is so shiny

and looked awful, although one of the goals was to have this so i could move from car to car easily, and it worked, ~10 minutes to pull the thing out

 

the previous, and future setups have the Wideband thing on a bracket i made that bolts to the top center of the dash.
you'd think this would be a great place to have it, and it would be easy to see while driving, but its not

 

do i sense personal trauma here? maybe there's a support group???

 

lol, there is a google group, which is actually really good.
its just hard to get windows to see a cable, and this box, and then that all needs to talk to the PFC, and then to do tuning there is another piece of software....

and then once its tuned you need to reverse all that.
when it works it works, and you don't want to touch it!

 

i had all day Saturday to mount the DL-340XB box, so i enjoyed myself and took some pics.
First is where to mount the thing.

there was a FEED built FC that came up for sale a long time ago, and it had an F-Con on the kick panel,
Awesome

not awesome

and yes mini m&m's fit in the JDM flare holder

 

i looked at the glove box too, which i think would be best,
but with the cables, its kind of long and would require lots of cutting, and then where would i keep my gloves?


under the dash would work too, like the old box, but i'd need to make a new bracket


i think i decided on the other side,

 

i swear that the thing isn't as light colored in person,
its also not flat

so i took some scrap and made one,



and Viola, that tumor in the harness is covering the WB to DL340 connector

 

I apologize as I've been absent from this thread for a bit, but why stay with the antiquated electronics?

 

the PFC? there are a few reasons, none of which are really all that logical, lol
the tuning software, FC-Tweak is really awesome, its really unlocked features that the PFC had the whole time and nobody really knew about.

also i need to smog this thing still, and so i need to keep the Air pump, ACV, injectors, and ECU, although i can put the stock ECU lid on the PFC

 

Ah, gotcha. Well, to each their own. Good progress so far!

 

thank you! i could change my answer to, i don't know, its not like i planned it

 

here is a screen shot of the playback screen in the new DL-340 box, so far this is the really big leap.
this is a 1st and second gear run from a stop, the AFR syncs up really well, but you can see that peak boost (green) is highest after i let off the throttle (yellow), its just interesting to see the time lags

its small, but you can also see the turbo come on boost, the wastegate makes it flat for a bit, and then boost creep takes over, lol.
not a great boost curve, its a little lazy and then comes on hard at 5500. i could fix some of it, but hooking up the boost solenoid (i need a vacuum tee, preferably not plastic)

and then you can also see a little of the weird stuff, something hasn't been quite right for a while, but you can actually SEE that the timing is doing weird stuff at low throttle openings.
i'm really not sure WTF would cause that, i can look at any of the stuff on the left side, and none really correlate, so possibly a bad connection.

 

Does the tweak-it datalog offer chart unit points? It would be helpful to see what's actually going on in the data log, in terms of actual numbers.

 

the data on the side is what is feeding the graph, so sort of. i can move the cursor, and it'll display the values on the side.
i can also change it to two or one graph, so i could have like TPS and Boost on the same chart
so far i'm happy with that, although i'm going suggest that i could full screen the chart, just to post pics

 

i was going to say i found my problem! but really its just A problem.
its been doing this thing where it bucks on decel, and its sort of inconsistent (and the metering pump is kind of flakey)

the only thing showing up in the logs was that on decel the timing was doing weird things, and the PFC is pretty dumb, the timing is just rpm vs load, there aren't really any corrections for it.
i got the (bright) idea to just turn the log on and wiggle some connectors (hoping to see if the metering pump is my fault at the ecu adaptor, too, its not)

and i found that the MAP sensor connector isn't happy. since the FD map sensor uses a different connector i have a little jumper harness, which apparently isn't working.

Red arrow is just me shaking the connector, not good!
Blue is where i disconnect it
and the Green is plugging it in again.

i cleaned it, and ran another log and it seemed ok, but i took it for coffee this morning and its not happy.
so i get to try again, and then pack the connector with something so its sealed.

can be seen here, the FD connector is sealed, and the FC connector isn't, and the FC usually has the MAP sensor as a roof...

 

The connector issue must definitely be addressed, but note that the timing on this ECU is not only dictated by the RPM - boost maps. The PFC performs various corrections in real time based on:
(a) water temperature
(b) air temperature
(c) TPS signal
(d) low RPM stability algorithm (turned on by Idle-IG ctrl)

 

after some putzing around, i went and got the shop manual out, and
supply voltage is a solid 5.11v, but the ground is all over the place


so today i get to pull the plenum off and have a look. actually if i'm lucky, and stupid, i forgot to hook up #9 last time

 

thanks for that. i have some screen shots and i was wanting to ask in the PFC group, but i wasn't sure what the question was.
the symptom is a buck on decel, basically. i only found it because of the log playback feature, which didn't exist before.

 

again i had a free day to just take my time and work on the car, which was nice, its been a long time since i had to actually work on it.
the offending ground. there was a little oil around, but the bolt was tight, terminal looked ok to me.
obviously it wasn't, so it got cleaned up. the other ground on the right was attached, i had it on the coil bracket (FC harness, FD vacuum rack FD ignition)

it tests ok now, although i would have liked to see something more bad. like a big obvious, oh thats the problem problem
i leave the throttle body hooked up, so cooling system stays unopened



the other "while i was in there" thing is that on hot starts sometimes it just runs badly for a while, and then something switches and its fine.
the only thing i could think of that kind of met the symptom (and i hadn't checked/replaced) was the ACV check valve.
we will see if that is it, the old one looked ok, so i kind of doubt it

i had a new one, i was just waiting for, i guess now to put it in.


ACV worked fine, i changed a check valve cause it wasn't check enough,

added a couple clamps too.
it wouldn't surprise me if the problem i have had is one of these check valves hanging open, but we'll see

i also changed the knock sensor, i was using a pretty scabby looking FD one, and since the FD one has a different connector i had a little shielded adaptor harness
but i noticed that the FC one was actually long enough to just plug in, and people seem to have used the FC sensor on FC engines without any trouble, so i switched em.
the FC sensor i had was really nice too

 

i decided to add a grounding kit, it HAS to be fixed now!

 

i decided to test the ground before i put it all together and had a hard time getting a good one, and then i looked at the battery terminal, which is pretty bad, then it broke.


i replaced it but i think i'm battery shopping. this battery was $40, in 2017, and i've treated it really really badly, and its hanging in there, totally got my moneys worth, but
the terminals just don't work well with the factory cables, or cable ends. not sure if the grounding kit is going to stay like this either

 

so i got it together, and tried to fire it up and it was all warnings and check engine lights,
one of them was the fuel temp, which is on a little expansion harness i built. i had noticed that the JDM tuners usually run the FC harness and then just add whatever they need, they usually run an F Con V, which lets you do this easily
i had built one for the stock ECU, which basically added the purge valve, Double Throttle control, and Fuel temp, along with all of the turbo control solenoids, to keep the check engine light off.
for the PFC, i got rid of the turbo stuff, so it was just the three wires, and i was puzzled because i played with the other two (Purge and Double Throttle) and then rewrapped it, but the problem was the fuel temp.

i pulled it out and unwrapped it, and everything looked fine and i was confused, but then i looked at the ECU side and it was unplugged
plug it back in, and everything is fine....

i did this so long ago, i forgot i did something clever. the FD has an extra connector, so instead of just doing a straight FC harness to FD ecu adaptor, this extra plug is the expansion harness, so its FC harness to FD ecu AND then this extra plug goes to the appropriate FD stuff.

it is nice that without the turbo control stuff i can pull the harness without pulling the intake

 

so i asked the PFC group, and Xavier noticed this


he said something about it being a PWM switching on, and then suggested trying without the O2 sensor hooked up.
and no more noise


he also suggested that i should ground the O2 sensor anywhere else, so i did and it worked.
i went for a drive, and its back! it got to 90% tuned in a drive and a half, it runs awesome, very happy!

i do wonder how widespread this noisy O2 sensor thing too, if we're tuning to an AFR, an accurate signal is really important.

 

Wonderous Bread and i have talked about how hard it is to get a MAP type ecu car to hot start and then they run lean when the engine is warm, but not hot
i love this playback feature so much, i was delighted to find that both things were really easy to find and figure out.
here is a hot start,


i had thought i was all in the idle cell, N2, P5, but its not so i went to the editor and N2 P6, and hit + once (i'm not sure what the increment is, lol)
and here is the next morning

way better. i think i hit + again

the warm (between 50c and 70c), was similar, although it was using different cells. and i also added a little to the 10c air temp map, as it was just generally lean
i've been wondering what Mazda had done, and well i guess tune the car, lol

 

since it was running so well i decided to screw it up.
i bought the fast air temp sensor like all the cool kids have, and put it in the elbow, and called it good

but i noticed that if the secondary's weren't open, air temps would go up. so like if i start it hot, and just drive home, like i do, air temps don't change.

the Cosmo elbow has the sensor in front of the primary throttle,



after thinking about it for a while, i decided that i had a choice of the place in the elbow or the stock location under the intake.



so i switched it. and its different.
temps are higher, but also its more responsive, it moves around more.
it changed the tune too, which is kind of the dumb part, i should have tuned it 100%, and then moved the sensor.
and who knows which one is right? and how would you know?

i guess the problem i have is that the idle mixture varies, the cell is tuned, but when its colder its leaner ~13.2 and when its warm is more like ~12.6
why not adjust the air temp map? i guess you can't be a mad scientist without the mad part....