RomRaider

IF YOU DON'T KNOW WHAT CARBERRY IS, AND/OR DO NOT HAVE THE DEFINITIONS REQUIRED TO WORK WITH THIS ROM, THEN PLEASE SEE THIS POST: https://www.romraider.com/forum/viewtopic.php?f=36&t=7594

As it would seem, between the various forums and groups I am a part of, there is a still a need for a decent base. A decent set of the majority of compensations that would stay the same between various mods.

This base is set up to fit the largest audience, which means it should run decently from vehicle to vehicle if the conditions of parts required is met.

Requires 02/03 USDM ECU or any 16-bit Drive by Cable EDM/ADM/JDM ecu.

This base is intended for standard compression 8.2:1 to 8.5:1 EJ205s.

This base is set to run speed density off the stock MAF IAT. Think of this as an off-the-shelf SD ROM. type of intake does not matter, the fact that the IAT is still being used from the MAF body, that is what is important.

It doesn't matter if you have the EGT in the up-pipe or not, codes are removed
It doesn't matter if you have the Rear O2 sensor or not, codes are removed

It is intended that you have removed the TGVs. Codes are removed, and TGV control by ECU is disabled.

It is intended that you have a 3.9 Final Drive 02-05 WRX transmission or other WRX transmission with same gears. Calculated gear thresholds are dependent on trans gear ratios and tire size.

It is intended that you have STi pink injectors, but the next post will include instructions for all the table changes regarding different injectors.

It is intended that you have a stock TD04L at max wastegate pre-load.

It is intended that you have the upstream oxygen sensor in the stock location, before the up-pipe/turbo.

It is intended that you're running wastegate pressure, there is no wastegate duty.
* But if you should have a MAC 3 port or similar (Cobb, GS, etc.) then TD and BCS PWM hz are set up, IAT comps, pear gear comps, etc. You just need to adjust target boost and wastegate duty min/max.

91/92 AKI octane, or 97/98 RON

you will probably remove a couple of degrees in some areas if you're on 91 ACN or 95 RON
you might be able to get a couple degrees more timing in some areas if you have decent 92/93 octane or decent 98 RON

enjoy

ITEMS TO CHANGE WITH INJECTOR CHANGES:

(where old scalar / new scalar is your factor against IPW values)

Injector Scalar
Injector Latency
*enable/disable low pulse width
*low pulsewidth compensation
*base pulse width activation threshold
*IF your injectors require / or have this. If you're using large injectors that don't come with this, good luck, that's not advised.

Fueling - Injectors:

Minimum injector pulse width

Fueling - Cranking:

Tables A (normal) and C (hot restart)

CL/OL Transition:

CL to OL (with delay) base pulse width

Fueling Tip-in Enrichment:

Throttle Tip-in Enrichment (main table)
Minimum Tip-in Enrichment Activation

Fueling - Decel Fuel Cut:

Fuel Resume Initial Injector Pulse

After being in and out of the scene with the same 02 wrx, I finally created a new account after being in email-activation/account verification purgatory for 6 years.

I wanted to thank Jon for his hard work on this and helping others like myself understand the carberry map and helping with creating basemaps. I replaced my engine just to flash a carberry rom that ran too lean to blow that engine up and replace it again and this time build up a much better understanding of how the map reacts to different changes and adjustments. I am really excited to see something like the VE map get its first basemap and development continue to happen. I have built up my map and literally read every post about injector timing and how avcs affects it, how to get by with maf scaling, ignition dwell tables and how those rotational idle tables don’t really seem to do anything. Finally my car is running well, but its stored for winter!
And this was all in the last 2 months!!

I am really curious now with the VE map and hope that I’ll get to try it some time.. any pointers I should look at for sti dark blues, vf48 and no cats? JDM ej205 with avcs 9:1cr!

I'm almost done with a base for that setup, I just need to run it through my spreadsheets to get the output tables so I can post images and set up a new thread for it - such as this thread.

Should be later today.

i am going to try and post a pic of my map.

its setup for 10hz pwm (if i remember correctly) on a 3port bcs,

94oct
some decent timing, waiting on jon to see if i could try more
23psi


want to go fmic, but don't want to cut my burn-up wrc bumper...

I have 440cc injectors, should I modify the following values? Because comparing this ROM with a WRX stock, it has differences ...
Then, I modified the values, and the gasoline pump relay was clicking intermittently and fast all the time, and this made it fail when it accelerated.

What am I doing wrong? Thank you

Forester s-turbo/GT (A4SE700x) doesn't have a fuel pump controller, you need the fuel pump duty cycle to be 100%.

Thanks mate! Any other suggestions? The table is IDC?

EDIT: Mine is EJ205 with individual coils, equal to wrx 02.

Enviado desde mi Moto Z2 Play mediante Tapatalk

Hey Jon just curious, why is the displacement set as 2085cc? Should I not set this to the normal 1994cc so it doesn't mess with what you've made or is that something not necessary? I figure it would lead to different fueling so I thought I'd ask.

In Carberry, you don't have access to SD Multiplier / Offset and displacement as you would in A4TJX100

I use a custom definition that changes the representation of the engine displacement.

Because you don't have access to the SD Multiplier, it's a static value. The way to change this value is to change the displacement.

The purpose of this is that at the same centered IAT temp, MAP x RPM VE tables can be directly interpreted from or to Cobb. Because I know Cobb's SD multiplier, I can convert from MERP, to Cobb, Carberry, or anywhere between.

In fact, I can actually convert any of the above VE tables to a IPW based table, or similar. All the math is there.

It's actually a higher SD multiplier * 1994cc, but you don't see how it works with just the DISP.

If you had the definition I have, you could change the displacement to 2457cc and Cobb VE table / Carberry would be directly transferable (given table size change, interpolation)

TL;DR - yes. table value is correct for EJ20.

Phew, that makes a lot more sense and it actually popped up in my mind that there's probably some behind the scenes work on this rom that I can't see. That was actually going to be my next question, is this whole project derived due to Cobb also having a SD setup for stock MAF/IAT placement? It would make sense, considering if ever the need to convert you have all the data.

I honestly appreciate all your effort into this basemap. Compared to the stock rom my new wagon had I did some tweaking and it just runs smooth like butter now. It's fun getting used to SD while my sedan is finishing up paint, that way I'll have a bit more knowledge on the ins and outs of it. Obviously that'll be running a mafless speed density setup with a gm iat and such, so this basemap will obviously work differently, but more experience and knowledge never hurts.

One last question I do have is tuning VE. I use vgi's maf VE rescaling tool and switch my car over to running full time open loop so afr corrections play no part in my rescaling. Even when I do a correction of say, 5% VE my corrections tend to swing wildly that way which leaves me confused. Do you recommend any other spreadsheet to deal with tuning VE on Carberry?

Megalogviewer HD, hands down

also, with the advances in being able to look at the logged data in complicated ways, I actually do not recommend running vehicles in open loop entirely, without any help from closed loop, for scaling things, there is not reason you can't look at LTFT + STFT over time as an accurate correction.

as for IAT usage for SD, I absolutely ******* hate those GM 3/8 NPT sensors. they are slow, and they are huge, and they soak up heat and retain it with that huge brass body. Get a 1/8 NPT sensor and 3/8 NPT nylon 6/6 TEFEN bushings. They are 3/8 NPT threaded on the outside, and 1/8 NPT threaded on the inside, and they work great as an insulator, and cost like $12 for 10 of them on amazon.

As for where I like to get my IAT sensors, order through Haltech and they get shipped out of California for like $37 plus shipping or something.

Oh wow, Megalog looks amazing. Being able to visualize data like this easily when we don't have realtime tuning is actually a godsend.

Noted, I'll refrain from using full open loop then.

I found those bushings on amazon for $4.45 for a pack of 10 on amazon for the blue ones! I'll definitely grab those and convert to the Haltech sensor then. I've generally heard things about the GM sensor being slow and while it being extremely quick doesn't bother me, I do need an accurate reading especially when I do track days in the other car. Should I just use their PDF from their site and interpolate the data into our IAT scaling or do you have any data scaled for the Carberry ecu? Honestly, the other car will be running a Link ecu so their PDF info should be plenty but you never know!

Thanks again, Jon!

Hi Jon,
Great work on this rom and on your guidance and contribution to the forum.I have a few questions ,request,could you please share your formula for wideband/target afr as used when using megalog viewer hd?Is it okay to apply the full corrections as generated or just to apply a 50% reduction on each cell?i.e if a particular cell requires .86 change,do i apply the full correction?Does this also apply on the stft/ltft trims?
regards.

First you have to figure out the delay between the lined log and the wideband.

This delay will be greater at low exhaust velocity (low rpms, low load) and less at high velocity.

Depending on the sensor's placement between the downpipe and midpipe and logging speed, so for me to say it's about 3 lines difference at WOT would apply to some logs/vehicles and not to others.

I have a couple different calculated fields for the same equation, because of wideband delay differences because of placement, and also number of logging items and overall logging speed.

I'll add to my MLV walkthrough sometime this week.