pgmfi.org

Hey guys. This forum has a lot of great info and has enabled me to tune my car on my own; something I have always wanted to do. Thanks a ton to all those out there who helped make that possible. I'm sure many of you understand the satisfaction of building your own car and tuning it to do some pretty amazing things.

Also, I'd like to thank IntegraKarim for his thread "Progress of a Simple Minded Noob". Having a project's info in one spot and seeing it from start to finish was really helpful and helped me with the confidence to tackle things myself.

BTW, I asked him if I could piggy back on his thread and he was good enough to agree. Check out his thread at http://forum.pgmfi.org/viewtopic.php?t=10435&highlight=.

It's my hope that this thread can be a similar resource to noobs. I hope to make this thread a place to post details about my project from start to finish including obstacles and their solutions rather than a spot to ask questions and get off on tangents. With any luck, it will be a source of info for others.

The Beginning

I have always liked "machines", whether its a big block Chevelle, a tractor, a semi truck, or even a Civic. I also like speed, but who doesn't. I wanted to get a project car to work on, but didn't want to break the bank. A Civic seemed like one of the cheapest ways to have a lot of fun without spending all my hard earned dollars on a project car. I have a family to feed, too.

The thing that drew me to the Civic was that one could make a pretty quick car by using "technology". I have always liked the idea of a sleeper car. Although a Civic can draw a lot of "ricer" attention, I thought I might be able to get away with having one so long as I didn't add any external bling to it.

I bought a used '94 Civic EX 2 door coupe a few years ago. It had about 145,000 miles on it, but was pretty clean on the outside and seemed to have sound internals. I did a compression test and checked a few other things before purchasing the car.

I loved the lines of that little commuter car. I added some new wheels and tires that were on the lighter side, rebuilt the suspension with Eibach springs and Tokico shocks, an ST sway bar kit, poly-urethane bushings, and added a 2.5" stainless catback. Incidentally, I love my TRD (Thermal Design and Research) catback. It sounds great.

I bought a G-Tech accelerometer to measure HP gains and saw that my catback added about 6HP. The car handled great thanks to the new suspension parts. It also hooks up better and is a fair bit bumpier. Those were things I was expecting. In summary, I kind of wanted a street legal go-kart.

All this was fun for a while, but the engine was still lacking what I yearned. I knew I needed a turbo when a buddy bought a twin turbo 300Z. I've never felt acceleration like that and fell in love with turbos after the first ride in his car. Even faster NA cars just don't register on the butt dyno like a turbo car.

I installed a RevHard turbo kit on my car after a few years of driving around NA. The kit consisted of a T3/T4 turbo (not sure what trim or AR), a good size intercooler, an external Tial wastegate, a Vortech blow-off valve, a high-pressure fuel pump, a Missing Link, and Vortech FMU. I installed the kit myself in my garage. It required some cutting, welding, etc. etc. and was a bit of a chore, but I eventually got it installed.

I had a little hiccup in that I didn't get my waste-gate plumbed quite right. I overboosted for a fraction of a second and blew a head gasket when testing one day. I didn't have a boost gauge installed or a wide-band O2. The lack of instrumentation and a heavy right foot cost me about $80.00 for a new gasket and 8hrs of my own labor. I changed it out and the car's been running strong since. Although it was a stupid mistake, I consider myself pretty lucky not to have done more damage.

I have been driving my car for a couple of years boosted to about 5 PSIG using an FMU. It's pretty fun, but of course I needed more power. I have some big plans for the future, but in the mean time I have decided to tune my existing motor using the ECU instead of that crude FMU. I should be able to build more boost on the stock internals and get some practice tuning my car on the existing motor before putting a built one in. This motor is practice for something better. I've got much more instrumentation this time, so I hope not to have any blow-ups this time.

Interesting story!

Hope this thread helps a lot of newbs/ pros alike.

Good luck with your build(s)!

Education

I'm a big believer in understanding what you're doing. I have a degree in Electronics and Computer programming and have been designing and programming real-time embedded hardware and software solutions for a few years. I'm not trying to brag, but give some background in case some are discouraged if it seems they are struggling more with some of the stuff I picked up quickly. It's easier to learn something you've already mostly figured out once.

I'm not as educated as I would like to be on thermodynamics and engine theory. I've been trying to remedy that by reading books like "How to Tune and Modify Engine Management Systems", "Maximum Boost", "Honda/Acura Performance", "Honda/Acura Engine Performance", and various Internet resources. I still consider to be lacking in this area, but haven't stopped looking for more info.

It's also important to have a good people network for getting information. I work with a guy in a local performance shop for "real-world" info. He's a great resource to bounce ideas off and ask questions you might not trust to Honda-Tech or somewhere else.

Anyhow, the main point of this post is to do your homework before you try tuning a car. I see a lot of questions on this forum that make me wonder how so many people manage to safely tune their cars. I'm sure there are a few blown motors here and there. When people say they've read for 2 hours and still don't understand what they need to do, I would always like to reply, "READ MORE". 2 hours is a drop in the bucket compared to what your overall education will require. You don't need a College degree, just some work and resourcefulness.

There are a lot of smart people on this forum. They weren't born with the knowledge they now possess. You'll have to put in the time, too, if you want to successfully tune your car and do the things they do.

Thanks IntegraKarim. Hope to summarize lots of the good info from this site here. I might be long-winded, but it will come...

The Plan

I got sick of not spending any money on my project one day. I hadn't done anything to the car except install a B&M short-shifter and change the oil for quite some time.

I had been saving my pennies and decided to buy another D16Z6 like the one I have and build it for lots of boost. My target is 300WHP, which I think should be plenty for my little Civic to handle. If I still need more power I think I'll find another platform to do it on.

I managed to pick up a used motor, tranny, and accessories for $600. I bought it and then it sat in my garage for a year. I recently tore it down and sent it off to get Darton sleeves. It's back now and looks great; can't wait to stuff some oversized pistons in it. I ordered J&E pistons and Eagle rods. Things are coming together slowly due to the schedule of some of the shops I'm working with, but it will get done. I'm not in a hurry for the new motor yet.

In the mean time, I thought I would practice tuning by using the hardware I have. My plan is probably more cautious and longer than some of you would make, but that's my style. I want to learn and have fun while I do this. Being a stress-case is easy for me and will kill the fun factor, so I take things in baby steps.

My high-level plan is outlined below.
1. Gather data from O2 sensors, EGT, fuel pressure, and any other data I could from my working car before making any modifications. This is my baseline.
2. Chip my ECU.
3. Get the stock ECU program running in the external chip and make the engine run just as well in this configuration as without the "chip".
3. Add bigger injectors and modify the ECU program to work as well as it did with smaller injectors. Keep the FMU and Missing Link for now.
4. Remove the FMU and Missing Link and tune the ECU to control fuel and timing at the 6PSI of boost I have.
5. Add a boost controller.
6. Tune for more boost (maybe to about 200WHP?).
7. Install new "built" motor and retune.
8. Eventually turn the boost up on the new motor to give me ~300WHP.

I've just completed step #4 and have a car running without FMU and Missing Link now! I still need to tune more, but it's already running much smoother and is a little quicker than before.

I have broken each high-level step above into smaller steps when needed. I ran into some issues that I never really expected after doing a bunch of reading and searching on the forum. I'll give details on each one as I go. Stay tuned...

Shopping List For a Good Tune

Here're the things I bought to make tuning easier.


Make sure you take your time and be organized. I ended up making two or three orders to moates.net because I was so over-excited about everything. I could have saved some shipping!

I've done lots of soldering, even on SMT boards, so installing the chip kit was no problem. I did have an issue initially that jumper J1 wasn't working. It's been fine ever since; the issue was that the jumper had a clear cote of something on it that prevented a good connection. Check my thread here http://forum.pgmfi.org/viewtopic.php?t=12302&highlight= for more info on that.

If you don't have lots of soldering experience I highly recommend finding someone else to install the chip kit for you. I cringe when I hear about some ECUs that have been mangled almost beyond repair. You need special equipment to install the chip kit without exposing yourself to undue risk. Using a $5.00 Radio Shack special is a gamble, IMHO. I use a Hakko 936 ESD Safe, temperature controlled soldering iron and a few precautions. Lots of people gamble on this one; with poor odds I might add.

I have used ROM emulators in other systems and they are a *big* help. I have a ROM emulator already that would probably work, but it doesn't have a battery backup, so I sprung for the Ostrich off Moates.net. Using the same equipment as everyone else is an advantage because it will make trouble-shooting easier if I ever have problems. It also ensures that I'm using an emulator that's compatible with Crome, which will make life easier. For $175, it's really a pretty good deal. Using an emulator will speed things up tremendously and will allow you to do things like Auto-tuning with Crome Pro, which can't be done if you're just burning chips.

I already have a chip "burner" so I didn't need to spring for that. I'm definitely going to burn a chip once I get my tune complete so I don't have wires hanging all over the place. It's also fewer things to break down in a daily-driven car.

A wbo2 (wide-band o2 sensor) is *essential* for tuning and the LC1 seems to work well. The Logworks software is really nice, too. Since I'm well versed in wiring and electronics, getting it wired up so noise wasn't an issue wasn't too challenging. There are a number of theads out there if you need help in this arena.

As far as injector selection was concerned, I wanted to have enough injector to reach the 300WHP number and Saturated RC550 injectors seemed like they fit the bill. Having an injector that's too big isn't good, though, since they're harder to control in some cases like idle and cold start. I worried even the 550s might be too big for an easy tune in my climate in winter, but everything's working now. Thank-you Fuel tools! As a backup I was going to borrow some DSM 440s until I could learn how to tune better.

I'll expound in later posts for each of these items since I had problems to solve with many of them. I'll try to post links that helped me. I have been keeping track of many of them as I read and search.

Getting a Baseline

Before changing things on my car, I wanted to get a good picture of how things were currently running so I would have something to compare it to later. My car had been running fine for a couple of years, so I wanted to take a snapshot of things in case I ran into problems later. Having a baseline and only changing things one at a time would allow me to diagnose problems more quickly and revert to a working configuration.

I already have a DIY electronic gauge with boost and fuel pressure. The values are displayed to an LCD, but more importantly readings can be logged to a PC. Although these data points are somewhat interesting, a much more important metric on a car's tune is the AFR (air-fuel ratio). I purchased an LC1 from Innovate Motorsports because it is relatively inexpensive and got good reviews on different sites I checked. The Innovate Motorsports website can be found at http://innovatemotorsports.com/.

Innovate also has a forum for help with their products, so be sure to use that if you get in a jam. There appears to be quite a number of people who have had issues with early LC1s, although it seems as though Innovate has worked those things out for the most part. Handle your wbo2 with care! They are somewhat fragile. It seems that the LC1 is more immune to noise than the LM1 due to the separate analog grounds.

Wide vs. Narrow Band

If you don't understand the difference between nbo2 and wbo2 sensors, you should do some reading. Here's a link to give you a high-level overview http://auto.howstuffworks.com/question257.htm. In short, an nbo2 can really only sense lean or rich conditions; its output is highly non-linear and typically changes from 0 - 1 volt. A wbo2 and associated controller typically outputs a signal from 0 - 5 volts and can accurately read and represent AFR values from .5 to 1.5 lambda. An nbo2 cannot be reliably used for tuning, but a wbo2 can. Your Honda ECU only understands nbo2 signals, though.


O2 Sensor Installation

The Innovate Motorsports website contains documentation and even some video on the LC1 and associated software. Check /pgmfi/forum/media/a3/a3376fb36dd323f9db9a61a6ae437db0685a6c7e4043a612c3e9056746108d4e.pdf for details on the sensor wiring, mounting location, and a lot of other good information. I've seen lots of posts on this forum asking how to hook up the LC1. RTFM (read the fine manuals) and you will find what you're looking for.

I mounted my wbo2 (wide-band o2 sensor) in front of the CAT, which is probably close to 3' from the turbo outlet on my car. I was able to mount my wbo2 so that it points straight up, which is the recommended orientation. It should not be mounted below the 3 or 9 oclock positions and should *not* be placed close to the turbo since the heat from the exhaust could damage it there. Placing any o2 sensor too close to a turbo outlet can also cause inaccurate readings due to turbulence and pressure in that location.

When I initially installed my turbo kit, I was told to put the stock nbo2 (narrow-band O2 sensor) just ahead of the CAT. I thought it might be better closer to the turbo, but did what I was told anyhow. I decided while I was moving and welding things up, I would relocate my nbo2 closer to the turbo since an nbo2 needs to run a little hotter than a wbo2 and I think my closed loop operation is smoother with it there. I used to log AFR readings from my nbo2 with my DIY gauges and noticed that the AFR swung more widely with the nbo2 further down the exhaust. I ended up relocating it about 10" after the turbo in the bend of my DP (downpipe). The stock configuration for my car is to have the nbo2 sensor in the exhaust manifold, not far from the exhaust ports so I'm sure it can handle some heat.

It might be worth mentioning that the LC1 comes with a bung and threaded plug for said bung. I used it for the wbo2 sensor that comes with the kit. Since I chopped my previous bung up removing it from the old location, I had to purchase another bung for my nbo2. I checked NAPA, Autozone, and a few other places, but found that an exhaust shop was the only place that had one on hand. The standard price seems to be $10.00 here. I used a 110V MIG welder with gas to weld things up. I have two o2 bungs in my exhaust so I can use both types of sensors simultaneously, which can be very handy.

Make sure you don't have any exhaust leaks! Leaks will affect your AFR readings and will create a lean reading or lean spikes. An exhaust leak will *not* register as a rich value, which seems more intuitive to most people at first. While I was re-drilling a hole in my exhaust manifold for an EGT (Exhaust Gas Temperature) sensor I noticed that the readings did indeed get leaner until the hole was plugged.

The Innovate Motorsports website has lots of good information about tuning and tuning theory. I read everything I could find there. The link is found above.


Wiring

You can find other posts with details on LC1 wiring from http://www.innovatemotorsports.com/forums/showthread.php?t=3410, http://forum.pgmfi.org/viewtopic.php?t=9645&highlight=. The link also points to a document found http://www.m2motorsports.com/files/lc-1_write_up.pdf that is pretty good. Also, http://forum.pgmfi.org/viewtopic.php?p=63467&highlight=lc1#63467 will give you an excellent explanation of how to connect things.

I wanted to mount my LC1 semi-permanently, so decided to add connectors to the end of the LC1 cables. I used a polarized two-position connector for the red and blue wires and a mini-DIN connector for the others. This allows me to move my LC1 to different vehicles and also connect it in different configurations in my Civic for different tuning needs.

Initially when I was obtaining baseline data, I just connected my LC1 to my laptop and used the Logworks software. I later connected it to my ECU and logged things through Crome Pro (the free version does not support logging), but I'll get to that later. Having the connectors allows me the freedom to do those kinds of things without messy rewiring. I obtained the two position power connector from Autozone and the mini-DIN male and female connectors from http://www.digikey.com. The part numbers are listed below. I also put an inline fuse in the LC1 heater circuit for an added measure of safety. The Innovate docs say to use a 5A fuse at minimum.

My initial wiring for the LC1 consisted of two small wiring harnesses. The first wiring harness was for main power and the heater ground, which are the red and thick blue wires respectively. I call this the main power harness. The second wiring harness connects the white, yellow, brown, green, and black wires to a Mini-DIN connector so they can be attached to different places. I call this the data harness.

To make the main power harness, I connected a black wire to the two position connector and then grounded it under a bolt that holds the ECU bracket to the inside of the car. The other wire from the two position connector is routed through a fuse holder to the cigarette lighter wire, which is a switched 12V source.

The data harness was created by soldering a Mini-DIN plug to the LC1 wires. I color coordinated the wires on the plug so they would come out on the same colors when the receptacle was connected for the most part. I then soldered a Mini-DIN receptacle into the car. The white, and green wires from the LC1 connect through the Mini-DIN connectors to a wire that is grounded under another bolt that holds the ECU bracket to the car. The black wire connects to the red wire of the LED and one side of the push-button switch that comes with the LC1 kit. The black wire of the LED connects to the white and green wires, which are ground. The remaining wire from the push-button switch is also connected to ground (white and green wires). I left the yellow and brown wires unconnected in this harness. I use them in a different version of the data harness later for a connection to the ECU.


Main Power Harness



Data Harness




Digikey Part Numbers

Mini-DIN receptacle - 275-1105-ND
Mini-DIN plug - CP-2080-ND


Replacement O2 Sensors

One of the links above will list this info, but replacement wbo2 sensors are available for about $40, which is cheaper than the nbo2 sensors I have been buying for my car. Not bad. I've taken the liberty to cut-n-paste part of the text here.




LogWorks Software

I think it's worth mentioning that the Logworks software provides a great datalogging interface compared to Crome Pro, for example. The only way to get graphs from Crome Pro datalogging is to export the data as a CSV and import it into another program (like Excel). Logworks has run-time graphs that can be manipulated and zoomed, etc. I like the software so much I may build a bridging device to convert my ECU output to a format that the Logworks software can understand. Innovate sent me details on their protocol. I'll let you know if I manage to build something.


Summary

I used the wbo2, EGT sensor, fuel pressure, and boost gauge to take a picture of how my car was functioning before starting to change everything. I would have liked to tap into the ECU to obtain readings from the myriad of sensors already on my car, but datalogging from the ECU cannot be done with the stock ECU configuration and Crome Pro. Other steps must be completed before that data would be available. Read on if you wish to find out how.

Chipping an ECU


Soldering

Here's a list of the tools you'll need for the soldering portion of your ECU modifications. Be sure to read the remainder of this post before you start.



I purchased a chip kit from http://www.moates.net because I'm too lazy to buy the parts individually from Digikey. If you're already placing an order with Digikey for the tools or supplies, it probably makes sense to get everything there to save on shipping costs. Chipping the ECU is as simple as removing the solder from the PCB (Printed Circuit Board) for the handful of new parts and then soldering them in. Check /pgmfi/forum/media/d2/d22424dbaf03c3bcade03f7a7d368228e8a74128a6b7931ab7f3538e5207388c.pdf for a good document on chipping the ECU. Moates.net has a few good docs you can check out at http://www.moates.net/documentation.php.

The ECU PCB (printed circuit board) is covered in some pretty goopy stuff that I removed with IPA (Isopropyl Alcohol), but I've seen others recommend carb cleaner. You should ensure that you have good equipment and decent soldering skills before you attempt to do this. You can buy chipped P28 ECUs for pretty cheap, too. Do a search on this forum, ebay, or wherever else. You can find good deals at http://xenocron.com/. A chipped ECU can be had for around $150.

Some things to watch out for when soldering are too much heat on your PCB which will ruin traces, ESD (electro-static discharge), and shorts created by sloppy soldering, desoldering or large iron wielding. These requirements pretty much rule out the $5.00 Radio Shack special, since it's not ESD safe, and will not regulate its temperature well. Please do not use one of the 110V gun-style soldering irons from Radio Shack. That's like using an industrial acetylene torch to make a wedding ring. I'm sure this has been attempted.

Having good light, a loop (magnifying lens), clean solder in an appropriate diameter, some solder braid, flux, IPA and a desoldering tool is also a big help. Do *not* use acid core solder. It is not for use on electronics. You should also take care not to use a spool of solder the size of small rope since it will be very difficult to use.

Another thing that soldering noobs usually do is create cold solder joints. A cold joint may look okay, but will create intermittent connections that can drive you crazy debugging. A single cold solder joint cost me tens of hours on a senior project. I thought the code in my embedded system was bad, but it turned out to be a cold solder joint on one of the power rails. Your joints should be nice and shiny in appropriate sizes. Too much or too little solder is no good, either. Jagged edges are usually a sign of problems, too. Just like welding, making a good solder joint takes some practice; dispensing enough heat in the right manner without overdoing it is key.

Enough soap-box on soldering. I wish you well soldering or finding someone who can do it for you. Getting a reliable ECU is paramount to the success of the rest of your project. A good link for soldering howto can be found at /pgmfi/forum/media/32/32eec3b28d98b7c99377c615b88d7d8e1895cbfe1c4f1bd3cc264dcba1d84161.pdf.



Problems

I soldered my ECU up and thought maybe I had a cold solder joint on jumper J1. I just couldn't seem to get the ECU to read code out of the external socket. It turns out that the jumper I was using was covered in something that prevented a good connection. Many of you will probably not have this problem. I guess I got lucky. For a more complete description of the problem, check http://forum.pgmfi.org/viewtopic.php?t=12302&highlight=. Things are working great now.


What is J1, anyhow?

Your ECU is a computer that is connected to a number of sensors and actuators. It runs code to translate sensor input into changes in actuators that make your motor run appropriately. The ECU comes with its code or program burned inside the MCU (microcontroller). Many MCUs are designed with a way to configure the location they read their code from. The OKI processor in my P28 ECU is no exception. The J1 jumper is a way to instruct the MCU where it should read its code from.

The J1 jumper pulls the EA# pin to ground instead of letting it pull-up to 5V via resistor R49. This configuration of a pull-up resistor and an optional short to ground is a very common technique used in digital electronics. The end result is that you can instruct the ECU to use an external socket for a new program by grounding the EA# pin and sending the MCU a logic "0" (ground) instead of letting it default to a logic "1" (5volts).

When J1 is in place, the MCU will read its code from the external socket. It *must* have a program there when J1 is in place or it will not run properly. If the MCU does not run properly, the ECU will go into limp home mode. Limp home mode uses comparatively crude circuitry to set fuel and spark settings. It ignores almost all sensor input and will result in very poor engine performance. You'll notice very poor idle, poor gas mileage, a low rev-limit, and of course a CEL (Check Engine Light). Changing the program or trying to use a damaged ECU can cause a CEL.


Chip Kit Contents Explained

The chip kit contains:

SST 27SF512 Chip*
28-Pin Collet Socket
28-Pin Low-Profile ZIF Socket
74HC373 Logic Chip
Jumper Header for J1
1k Resistor
0.1 uF Capacitors
4-Pin Datalogging header

The 27SF512 chip is the ROM that you can "burn" with a programmer. It can hold the new code or program for your ECU. A ROM emulator can be used *in* *place* of the 27SF512 chip. The ROM emulator and chip are *not* used together at the same time.

The 28-pin collet socket is what you solder into the PCB. It is a socket that makes a firm connection with whatever you plug into it. You could plug your ROM chip or emulator into this socket, but it's hard to push pins into and pull pins out of so won't be as handy for quick changes.

The ZIF socket (Zero Insertion Force) is useful because it is easy to add and remove parts from. Insert it into the 28-pin collet socket after it's soldered in. There is a small lever on the side of the ZIF socket that opens and closes its contacts. Open it and easily drop in or remove a ROM chip or emulator cable, then close the lever so the devices are tightly coupled together. Incidentally, I had a difficult time getting my Ostrich inserted into the ZIF socket due to physical constraints and due to the larger, round pins of the Ostrich. I added another 28-pin socket to the bottom of my Ostrich cable and simply plug it into the ZIF socket to make things easier. You can find extra sockets at Radio Shack, Digikey, or many other electronics distributors.

The 74HC373 is a specialized IC (Integrated Circuit) that is required by the OKI MCU when it reads code from an external device. Its function is to latch signals from the MCU during the first part of a read-cycle and hold them so the second phase can be correctly executed. Basically, it reduces the number of pins on the MCU that need to be dedicated to reading code from an external device. Pins are a scarce resource on an MCU, since they are otherwise used to read sensors and interact with other important external devices.

The 1K Ohm resistor is to tie the CE# (chip enable) line to ground so that it can be read when PSEN# cycles on the OKI MCU.

The 0.1uF capacitors are "decoupling" caps. Decoupling is in reference to the capacitor's function, not its type, which means a decoupling cap can be ceramic, tantalum, or a number of other chemical compositions. Decoupling caps are usually placed close to ICs and other electrical devices on a PCB to make sure the power is free from glitches and to source a small amount of electricity in the event that the nearby device needs a little more than can be supplied instantaneously through the PCB traces. You might be able to get away with not using decoupling caps, but I don't suggest you try. They are so easy to solder in and so cheap it just makes good sense to use them.

The 4 pin datalogging header is a connection the serial port on the MCU. It's what you connect your HULOG or other datalogging cable to in the event you wish to read sensor and other data from your ECU. I ordered a HULOG from moates.net, but you have other options here, too. Keep in mind that the signals on the 4-pin header are 5V signals and *not* RS-232. Do not directly connect a serial port from your computer to these pins. It will not work and may even cause damage to your ECU.

LC1 Data Harness for ECU

In order to tune a car, one must be able to correlate AFRs to vehicle load and speed in real time. One way to do this is to route your wbo2 sensor output into the stock ECU's nbo2 sensor input. There are a few important things to note when doing this.

1. The code in your ECU does *not* understand the signal from your wbo2. You'll need to disable the o2 sensor in your bin so the ECU doesn't get confused. Do this in the Options for your bin in Crome.
2. The stock ECU clamps o2 sensor voltage to 3.75v. You cannot send it a 5v signal and expect it to see anything more than 3.75v. This means you need to configure your LC1 *and* datalogging software (I'm using Crome Pro) to use a voltage level less than 3.75v to see the whole range of data.


Why Connect the WBO2 to My ECU?

If you drove your car around for a while and datalogged AFR alone you would probably see lean, stoich, and rich conditions throughout the log as you did different things during your trip. As you reviewed your log, how would you know if you needed to change the fuel map in your ECU? Even if you knew something as general as needing to add more fuel, how would you know *where* in the map to add it?

By connecting the wbo2 sensor to the ECU input, you can log your AFR values *relative* to RPM and load, which will enable you to properly change your fuel maps. Having enough data for any given point in time is the only way you can reliably tune your car. Any other method is just guessing.


Another Mini-DIN receptacle

In order to send the LC1 data to the ECU, I wired up a second LC1 Data Harness that connects to the ECU. When I got ready to datalog AFR along with RPM and load I disconnected the LC1 from the first Data Harness I created and then connected it to the Data Harness of the ECU. Now I can connect the LC1 Data Harness to the car Chassis or ECU depending upon what I want to do.

If you connect a wbo2 sensor to your ECU *be* *sure* to disconnect your stock nbo2. Having both the nbo2 and the wbo2 sensor trying to drive the same ECU pin will cause damage to one or both of the sensors and possibly the ECU. Different people on the forum have different strategies for managing which sensor is electrically connected to the ECU at one time, but I simply disconnect the nbo2 in the engine bay and connect the LC1 to the ECU via my secondary Data Harness or vice versa.

The wiring diagram for the secondary Data Harness is shown below. I opted against installing a calibration switch on this harness, but did install an LED so I could visually confirm that it was working.


LC1 Data Harness 2




Heater Ground

I also highly recommend that when connecting the LC1 to your ECU that you connect your heater ground to the engine block. It's a pain, but is the only way to ensure you don't operate your LC1 outside of its design constraints. See http://www.innovatemotorsports.com/forums/showthread.php?p=28897&posted=1#post28897 for more information.


I'll try to post the ECU pinout for reference along with some pics. I've seen it somewhere else in the forum.


Configuration

It is very important that you configure both the LC1 *and* the datalogging software to use a voltage range that the ECU can read. Since I connected the yellow wire of the LC1 to the ECU, I configured the LC1's analog 1 output to send 0v at 0.5 lambda and 3.5v at 1.5 lambda. You could probably use 3.75v for 1.5 lambda and maybe get a bit more resolution, but I'm playing it safer by leaving some margin. You configure the LC1 analog outputs using the "LM Programmer" application that comes with the kit.

I also configured Crome Pro to use 0 and 3.5v for 0.5 and 1.5 lambda. This setting can be found in the "Tuner Logging" tab of the File->Settings dialog. Once datalogging is working, you can compare the AFR values from Logworks and Crome Pro. I just eye-ball them when they're not jumping around much, but if you want you can follow this post http://forum.pgmfi.org/viewtopic.php?t=6085&highlight=lc1+calibrate and set the LC1 output to something steady long enough to calibrate Crome Pro and then revert back to the previous settings.


Software

You can use Crome Pro for datalogging. The free version of Crome does *not* support datalogging. Crome Pro also has something called Autotune, which can make your job much easier. It logs AFR values as you drive and makes adjustments to your map based on the difference between the measured and target AFR values that you set. Although I have tried it a bit, I plan to do a manual tune first to learn more and then try Autotune out more later.

You can also use Freelog to log your ECU data. This will allow you to tune your car by analyzing the Freelog data separately and then making manual adjustments in Crome. Not quite as flexible as Crome Pro, but cheaper. You can find Freelog at https://sourceforge.net/projects/freelog.


Plug for Crome Pro

While you can use the free version of Crome, I recommend buying a license if you get any real use out of it. Writing software is *very* time consuming. Maybe John can spend more time developing and supporting Crome Pro if he gets something in return for all his hard work.


Caveat Emptor

Completing the steps outlined in this post creates the electrical facilities to log wbo2 in conjunction with other engine sensors, but you *cannot* datalog using the stock ECU and ROM using Crome Pro or Freelog. You will need to be running the "QuickDatalogger + RTP" pluggin to get anything from your ECU's CN2 connector. More on that later; that can be accomplished only after you're running code from an external socket.

Good stuff keep it up =)

Tell me if I can fit an S2 Aries on my ecu!

STICKY..

Misc Pinouts

I'm attaching an annotated pic of the ECU connector for my P28-A02 below. I believe all OBD1 ECUs are almost identical. I didn't take the time to label each pin on the ECU, but did try to make the pattern pretty easy to decipher.


ECU Silkscreen

The ECU PCB silkscreen does show letters "A", "B", and "D" that correspond to each section of the connector. The numbers on the PCB silkscreen will *not* match what I have shown here or the manuals. The silkscreen numbers are useless to me. I got the pin numbering for this diagram from the official Honda manuals. They also match my Chilton manual.


CN2

I have seen pinout requests for the datalogging port all over this forum, but never actually found an answer; probably just need to look harder. I've included it here for easy access. I've annotated a pic I took of my ECU's header with numbering that matches the PCB silkscreen and moates.net socket labeling. You can see the cable from my HULOG plugged into it.

My Current Setup - FWIW

Here are some pics of my current setup so you can see what I'm talking about. This setup isn't as nice as some others out there like http://forum.pgmfi.org/viewtopic.php?t=11791 and http://forum.pgmfi.org/viewtopic.php?t=12356, but the wiring is reliable and allows me to do the things I want. I have some ideas that will clean things up a bit in the future. I'll get all the wires completely tucked away eventually.

I created a quick cover for my ECU out of thick Plexiglas to keep things from getting damaged by passengers. My intention was not to make it look cool. I chose Plexiglas because it was strong, non-conductive, relatively easy to work with and in stock in my garage. It's just temporary until I get my tune done, so I'll probably leave it a little on the rough side.

You can see the Mini-DIN receptacle with LED soldered into the ECU, the ribbon cable for my Ostrich connected into the ZIF, and my HULOG hanging out the side, too.

Very nice.

P.S: Wish this post was around when i was a n00b lol