T-leg Piggyback ECU R&D

[NICO]

ok one minute hear, THAWA what are you going to use to controle the ecu. what companys are you guys useing to make this work.

should i get a greddy e-manage or is there a alternet root

[THAWA]

I'm not using anything but an evo 8 patch harness to go between my stock harness and the stock turbo ecu so i can change the wires and add more without having to cut my own. No piggyback for me right now.

free5ty1e is creating a piggyback from scratch. Not modifying some other one.

[NICO]

ok i get whats going on now i was lost in all this thanks thawa, so where can i get one of those conecters

[THAWA]

http://bbs.legacycentral.org/viewtopic. ... 278#106278

[franticfour]

How about just bypassing the maf 20% or routing an adjustable flow tube around the maf to open in increments until correct.
Kinda getto, but cheap

rally on !
eric

[free5ty1e]

Whats up everyone?

V - I think we're in agreement on how to handle the MAF I/O. Achieving 500 HZ MAF sample rates should be no problem, even if I were to implement a complete transfer function. Still don't think it's necessary, however. The trims in the table will allow precise altering of the MAF signal with either method - I will probably end up implementing the original, but keep the PGA scaling idea as a backup should there be any unforseen difficulties. Might even provide a setting to switch between the two. It'll make more sense once you can mess with it.

Anywho, I haven't been able to devote any time to the still-currently-unnamed piggyback system this week, but I forsee some free time soon. I'm open to naming suggestions for this product, BTW, you guys should have a say in what it's called since you're who it's for.

Just so everyone knows, at the beginning of May I will be making a huge cross-country move from Orlando, FL to Elko, NV in my Legacy, my '92 Saturn SL2 that cost me $50, and my big-ass 24' decomissioned U-haul '88 GMC C-6000 box truck. As one might imagine, a trip of this size (2700 miles) could take a bit of time to complete. Add to this the fact that my box truck runs on gas, not diesel fuel, and can barely hit 60mph - and this trip is gonna take even longer. Umm... I am really hoping to get a prototype ready to send to Vikash before I have to devote my time to packing and moving. At that point I will also include a few extra burned chips so that while I am busy, he can have some to send to others who'd like to help the beta testing. Then once I get there and am back up online, I will be able to make any changes that need to be made and should be able to offer them for sale at that point. So we're looking at a realistic release month of June. May for some beta testers. The beta chips will be free but you will be responsible for breadboarding the circuit according to the schematic in order to test the unit, as I won't have time to do a circuitboard before I have to leave. You will also be able to exchange your beta chips for updated chips at no charge.

So if you feel comfortable breadboarding circuitry, then you'd probably be a good beta tester. If you also dabble in programming your input would probably be even better. I hereby open the discussion of beta testing and who would like to be involved - I won't have many chips to give out, maybe 5...

[free5ty1e]

By the by... do we now know where to get male and female connectors for the four sections of our ECU? I'll need a pair soon, if we know where to get them I'll order some....

Is someone sending me an unused ECU and possibly a connector? I forget where we were on that.

[legacy92ej22t]

Dibs, I wanna be a tester! That is if Vikash will help me with the breadboarding.

[THAWA]

yeah I was supposed to send you two ecu's. Just havent got around to it yet, sorry.

[mikec]

I'd be up for testing... Breadboarding is no biggie. Only considerations are I'm north of the border, and have no mods aside from a turboback.

[evolutionmovement]

How about the name having something to do with a rabbit as in the STI logo and, of course, rabbits are fast. Just hope Matt's wolf won't eat it. I'd sign up for beta testing if my car were at that stage. Dammit, if I hadn't been unemployed for a year ...

Steve

[free5ty1e]

Thawa - no worries lol, I couldn't work on it last week anyway. We'll probably just be using direct connections at the sensors for the initial tests, so the connectors arent needed just yet.

My list of mods also does not require a piggyback tuning system yet. I could possibly lean it out a bit so I'm not running as rich as my Subaru wants me to, but the optimum test situation would be someone with oversize injectors....

It's not a requirement for testing, though. It'd be good to see how the system allows for slight tweaks to the stock setup.

Steve - I hear that... if I hadn't gotten involved with this microcontroller, written app notes, and gotten noticed by an iPod developer who wanted to make use of it I'd still be unemployed.... go figure. All those interview techniques they taught me in school kinda sorta did jack for me

For the name, hmm.... as long as we don't piss off STi I suppose thats a good starting point. Although, if this thing gets big enough to piss off STi, I suppose I'd be pretty proud lol

What about something involving "Legacy" - the Rabbit is a VW It kinda turns into "Legend" allright, but then again that's an Acura. Damn car manufacturers, they've taken all the nouns...

[vrg3]

Chris - I don't think we completely agree regarding the MAF signal processing.

I think if we're going to the trouble of building this whole system to manipulate these signals, we may as well manipulate them meaningfully. Applying the correction factor to the voltage means that a single correction factor can have hugely different effects depending on the actual airflow. Why not make the result of the signal mangling as comprehensible and predictable as possible?

As for a name, I'm pretty uncreative. How about the "free5ty1e chip?"

Your trip sounds like it'll be quite an adventure. Are you planning on towing the Saturn? They work well as toads. Supposedly Subaru's current position on 5MT AWD cars is that they can be towed with all wheels on the ground, in neutral, with the ignition switch on as well.

I'll be glad to try and help keep things going while you're in the moving process, but I can't promise that I'll be able to give the project the time it deserves. I'm about to graduate and for all I know I'll be moving too. I'll do what I can, though.

I'm not sure breadboarding is a sensible way to do this. It'd be a little more of a pain, but soldering to proto board (with a socket for the microcontroller, of course) would be much more reliable in a car.

But also, I don't know if having us amateurs try it out is the best way to start testing... I think an ideal beta tester would also have experience tuning engine management and the resources to do it appropriately. What are your thoughts on that?

The male and female connectors for the ECU are available in the form of an extension harness that Hardy has found for us. The receptacle is a single 76-pin unit that connects to all four plugs. It's meant for the Lancer Evolution VIII, but apparently the only difference between our ECU's connectors and the Lancer's are the color. Hardy provided the link earlier in the thread, but here it is again: http://z1auto.com/prodmore.asp?model=ev ... prodid=241

At $110, that's a deal. It costs more than an ECU and some connectors might, but the amount of labor required to put it all together is easily worth the difference.

Matt - Sure, I'll help with wiring and stuff if you want to try it.

[Nomake Wan]

LegacECU Tuning Chip, how about that for a name?

[Kelly]

How bout something simple like this:

Legacy Sport Tuning.




(all must bow to my paintshop skillz)

Freestyle Technologies
LegTek
???

[THAWA]

LegacECU Tuning Chip, how about that for a name?

To expand on that. How about "LegaCU"

[Nomake Wan]

Ooh, I like that!

[azn2nr]

i think "the fre5tyle chip" would be sufficent. it gives credit to the maker just as the vrg3 fcd or the morgie fcd or the vos chip does. then you will have made your dent in history.

[NICO]

about you guys having to soldering the chips in, have you guys ever open a legacy turbo ecu? well let me tell you that there is 2 types of ecus out there on legacy turbos, 1 type has a single chip you can pull it right out the other has a double stack chip 2 chips on each other. when i got mine chiped he siad that you cant chip the double staked one cuz there is something wrong with it.

if it helps anyone out

[free5ty1e]

I'm liking the names so far. I'm sure one will stick eventually and all will be well. I kinda likes the "LegaCU" name, meself.

As far as the MAF signal handling goes, well, let's see -

The first thing I'd like to see the system do with the MAF signal is duplicate it exactly, from an ADC to a DAC. The car shouldn't know the difference, and drive like normal.

Once that's in place, then I'd have the DAC output an altered value based on the trim table, which could contain:

1) An offset, how far to shift the DAC output voltage from the base ADC input voltage

2) A scale factor

or

3) An absolute DAC output value

Any of these three possibilities is no problem to implement, in fact it could be an operational setting if there would be use for all of the above methods. Now if we come to the conclusion that #2 would be the most logical choice to utilize, then a PGA would work in pretty much the same fashion, as there are many gains above and below unity. By this, I mean:

The MAF voltage would be input to a PGA. Based on the current RPM and MAP readings (as above), a gain factor is pulled from the trim table and written to the PGA, which then alters the MAF voltage by that amount. All would be with respect to the car's reference voltage, of course.

So what are your thoughts on that? Do I have the wrong concept of what I should be doing here?

Nico - no worries, we don't have to go into our ECUs. I was just trying to establish that certain individuals could test pre-releases of the chip if they're handy enough with circuitry to breadboard (or otherwise wire up) the proper components, go for rides, and play with the settings.

It's only because I'll be moving and unable to work on it or communicate with you guys, I figured some of you'd like something to mess with during that down-time. And then you'd probably have all kinds of suggestions for me that would make the product better.

Besides its no skin off my back to burn some extra chips I have lying around with the latest version of the firmware and make them available to a few people.

[vrg3]

I don't think any of the three options is great.

The MAF voltage would be input to a PGA. Based on the current RPM and MAP readings (as above), a gain factor is pulled from the trim table and written to the PGA, which then alters the MAF voltage by that amount. All would be with respect to the car's reference voltage, of course.

So what are your thoughts on that? Do I have the wrong concept of what I should be doing here?

I think that'll kinda-sorta work, but it's not ideal. I don't mean to repeat myself, but I just want to make sure I was being clear about it:

Because the MAF sensor's transfer function is nonlinear, a multiplicative correction factor applied to the MAF sensor voltage will not yield easily predictable adjustments to measured airflow. Depending on the actual airflow, a 5% voltage trim could mean a 10% airflow trim or a 20% airflow trim.

In a sense, that means the airflow correction table's entries will have to contain information about both the engine's volumetric efficiency and the MAF sensor's transfer function in order to accurately adjust for things like different injectors.

A priori knowledge of VE -- and end-user knowledge of the transfer function -- would not be so important for this purpose if we did it The Right Way: applying trims to airflow measurements (i.e. read MAF through ADC, pump it through transfer function, apply correction factor, pump it backwards through transfer function, send it out through DAC).

I realize that in the end it might not be a big deal since people would ideally be tuning based on concrete measurements at the dyno rather than the theory of what their tuning settings will do. But if all of us could afford dyno pulls whenever they were needed we'd probably just be throwing our money at expensive standalone engine management anyway.

I just want to make sure we understand exactly what we're doing. If you understand this issue and still think it's better to go for the simplicity of applying the correction to the voltage, go for it and we can see how it turns out.

[NICO]

i wanted to let you guys no that my car runs 100% perfect and faster with the sti injectors in at first it was ruff, now i can rip doors off. what i had to do was change the tps boost and the other sensor near the tps and the car runs perfect.

about this piggy back system what can it do, if it sounds ok i will try it out. is it like plug a lap top in it, and tune it. i was thinking it was like a greddy emange system right.

[THAWA]

After more looking and researching I've found more compatible ECU's. At least the connectors are compatible.

Code: Select all

Model     Chassis        Engine       Special notes
Supra     JZA70          1JZ-GTE
MR2       SW 20III IV	 3S-GE
Celica	 ST202/203      3S-GE
Corolla	NZE121/124     1NZ-FE

NSX       NA1/2		    C30A/C32B
Legend    KA7-9		    C32A/C35A
Prelude   BB1		      H22A
Integra   DB8/DC2		  B18C		   AT
CR-V      RD1		      B20B

Diamante  F36A		     6G72		   With MIVEC
Galant	 EC5A		     6A13		
Eclipse	D32A		     4G63
Mirage	 CJ4A/CM5A	   4G92/4G93
Evo VII	CT9A		     4G63		   2/01-3/02

RX-7	   FC3S		     13B-REW		12/91-11/95


Legacy	 BD5/BG5		  EJ20H		  10/93-5/96
Legacy	 BC5/BF5		  EJ20G
Impreza	GC8/GF8		  EJ20G		  11/92-8/96

These are only for USDM cars with the exception of the Evo VII and the Subarus. What I do find interesting is that as much crap FHI used from Nissan, they didn't use the ECU connectors.

Don't ask why I'm posting this, but it could come in handy somewhere.

[azn2nr]

you u saying that we could possibly use plug and play standalones meant for the cars listed above???

[THAWA]

Well I dunno about that. There will almost assuredly be differences in the wiring for the engines, not to mention not all the engines listed are for 4 cylinders. plus you have to take into account the difference in sensors and crap