lc-1 serial vs analog output

jerseydevil · Post by **jerseydevil** » Wed Aug 18, 2010 7:19 pm

I searched but didnt' find a direct answer to this problem. I'm just learning ectune here and the whole tuning process. I went out to day, switched my ecu over to my demon, connected all my usb cables up to the laptop. I have my LC-1 wired up, so that the ground is on the thermostat where the ecu grounds, power is from an option connector in the fuse box, analog out 1 goes the d14 on the ecu (currently programmed to simulate narrowband), analog out 2 goes to an innovate wide band guage in the dash on the factory programed output (0v=7afr,5v=22.5 afr). So, I turn on the key and let the lc-1 warm. I start up ectune, connect to emulator and then datalog. Go into settings, and select wideband over serial lc-1 option. I close the menu and look at the 02 sensor reading and it says 10afr, 0.1v. My guage in the dash cluster is showign as expected.. 18 afr, as lean as it's able to display. The engine was off during all this so 18afr was what I was expecting to see. Why is the serial into the datalog showing the full opposite? I had used logworks to access the lc-1 the day before on the same usb, same laptop, so I believe the device is working.

jerseydevil · Post by **jerseydevil** » Wed Aug 18, 2010 8:19 pm

I figured it out. I Need to disconnect and then reconnect datalogging and then they inputs read the same.

fr33w0rld · Post by **fr33w0rld** » Thu Aug 19, 2010 11:17 pm

why don't u tap into the wire and send the wide band signal to the ecu also on d14?? and data log the afr via the d14 input...an old local tuner told me that is the best way and minimizes any delay of data..so u actually are seeing what the ecu is seeing... just my two cents...

jerseydevil · Post by **jerseydevil** » Fri Aug 20, 2010 6:59 am

I already have d14 setup like that, but my lc-1 is currently putting out a narrow band signal on that wire. Which i'm leaving that way for the moment. I'm running the car of th stock chipset inbetween tuning \"sessions\" which are few and far between. While I learn the process I'm leaving the car on the stock ecu chips to avoid damage to the engine until I can get a nice map worked out that does what it's supposed to over the entire rpm range/gear range. I installed a switch on the side of the ecu case just for this so i can switch back and forth with ease. My lc-1 can be a pain in the ass at times. When I switch the curvfes sometimes it like to do a full recal, which really sucks if it does it when it's not in free air. Using the serial input also eliminates any chance of a voltage offset and you get the true value of the afr from the lc-1. I also noticed that my output is more stable on the serial. The analog output has a very slight (~.1 afr up/down variation).

fr33w0rld · Post by **fr33w0rld** » Sat Aug 21, 2010 12:13 am

really dumb question here...if your driving around on a stock ecu with this engine...and feel just fine doing that...why even bother have ectune?? almost seems pointless...unless u take it to a skilled tuner..i have my doubts u could do better then the engineers at honda to make a map for your car... does it seem kinda odd?? most of the people on here using this are running boost...i have always been told.. NA is an expensive way to run slow...how about a super charger now?? your serial input is going to be different from the d14 wideband input due to lag and such...why don't u connect the wide band on d14 have a switch to switch it to narrow band? and if u really wanna watch ur afr u could use log works from innovative...

jerseydevil · Post by **jerseydevil** » Sat Aug 21, 2010 9:25 am

I bought a demon and ectune cause I plan on doing boost. There's a learning curve here and a pretty decent one at that. I wanted to learn tuning maps and how to work the program on a relatively \"safer\" build instead of try to learn how to both setup a turbo and the fuel maps at the same time. Learn to tune > boost + blowup. Once I get a handle on this process, and get the maps close to where I think they should be, I'll switch over the lc-1 to output 0-5v on d14 all the time and the ecu will rely on that input for closed loop. For tuning purposes, serial is the way to go, (according to the more knowledgeable and professional tuners on this board). Quite frankly, the serial input is easier to use as well. All you need to do is plug in the usb cable and start ectune and your done. Logging over d14, first requires me to use lm programmer to change the output curve on the analog from narrow band to wideband, when doing so, the stupid thing resets it's calibration and I need to do a free air + heater cal over again. Then I need to check the voltage offset for accuracy in ectune vs logworks. Then I can finally start logging data. And when I'm done I need to reverse the whole process so I can run the ecu off of the narrow band again. That's just a huge hassle. Plug in the serial and go is soooo much simplier. As far as delay? I\"m not sure there is any delay. The lc-1 is damn fast, faster then the ecu is capable of sampling anyway. The only area there might be delay is in ectune's processig the signal between the two usb inputs. Right now i'm not using closed loop, I'm just datalogging and tunign the maps. And the reason why I switch back to the stock ecu, is cause I don't trust my tuning abilities yet to run on my own tune continuously. I haven't had a chance to tune the high cam at all, and it does go lean over 6k. The stock ecu runs my engine @12:1 afr diipping down to 11:5afr in the high cam profile around 6.5krpm. That's crazy rich for a mostly stock engine. There's power to be made by tailoring the curve to my specific engine instead of a broad parameter honda uses to suit every engine that comes off the line. PLUS, I learn something while doing it. IMO learning the process is more important that slappin shit on and takign it to a tuner and being done with it. I'm into engine building/tuning to learn as well as gain performance out of the process.

JaredKaragen · Post by **JaredKaragen** » Sat Aug 21, 2010 4:28 pm

I only have one thig to say when commenting on sideband offsets: install only in the manner innovate's instructions give... This applies to all makes of wideband. Ground placement is key. 'Nuff said.

jerseydevil · Post by **jerseydevil** » Sat Aug 21, 2010 8:40 pm

I grounded my lc-1 on the thermostat. Which should be the ecu ground as well. I got power from a fuse box option connector. Still have an offset. Thinking about rewiring it to the ground and power wires right at the ecu. Supposedly that helps elimnate offset. THough it should only be the power wire causing anything.. since my ground is pretty much already on the same wire as the ecu ground

fr33w0rld · Post by **fr33w0rld** » Sat Aug 21, 2010 11:41 pm

i personally just grounded it to the chassis of the car in mine...all soldered together to one point..with no extension of the innovative wires...u don't really want to be chasing the lean edge of things to make power...timing is were the power is made..my machinist / old school tuner once said to me...don't lean it out to make power that's problem one with blowing engines up..he did go on to say in higher boost levels me makes more power running 11.90's to 11.5 then most ppl do at 12.5 ...how do u really feel about boosting a high compression motor?? ur dabbling pretty close to the knifes edge the higher the compression u run with boost...it takes a skilled tuner a very good ear..and a dyno to tred those waters...i have had my experiences...and talked to a lot of the local honda tuners be for i went on my venture...and really...if u like building engines and such...whats wrong with blowing a few up to learn something?? thats how the tuners have learned how far u can push an engine...and what \"safe\" timing and afr's are...good way to learn...blow something up...

JaredKaragen · Post by **JaredKaragen** » Sat Aug 21, 2010 11:52 pm

Think about it; your grounding closer the coil (emi source), the ground is traveling all the way through the engine harness and subsequent connectors along side powered electrical wires then it reaches the ecu. Look at the OBD1 OEM ecu schematic; you can clearly see many sensors having a ground go from the ecu to the sensor; even when there are plentiful grounds all around the sensor. Especially the inductive pickups. They even have an extra shielding. If you actually read most (if not all) manufacturers directions they state to ground to the ecu's ground within x inches of the ecu... And other places. This topic has been beat to death for a long time on many forums; even reputable tuners/retainers sites. My main reason for purchasing the LC-1 was for the serial connection. Can't beat the accuracy, from there it's just sensor-to-head timing that changes from setup to setup. All things aside; there is a way to _minimize_ offset, but many things factor into offset; variations in wire resistances in cars over time, corrosion and other factors as well. Anyways, for example I had a .25 offset with my lc1 at first; and with my 'JAW', I could never get offset below 3.50!! That is quite an issue. Hope it helps.

jerseydevil · Post by **jerseydevil** » Sun Aug 22, 2010 9:44 am

Well.. that what I did. It's grounded on the same terminal as the ecu ground, so you can't get any better. I think I might have even soldered the ring terminal to the ecu's ring terminal, but I dont' remember for sure. The lc-1 has been in the car for nearly 2 years now. I did just about everything I could to make sure it was done right. Soldered every connection, clean wiring job. Really the only thing that could be changed now is moving the power wire from the fuse box to the output from the main relay so that the ecu and lc-1 are powered from the same source as well. When I first setup the demon and ectune, I check the output in logworks via serial compared to the input from the lc-1 via d14 into ectune. I had a 1.3afr offset.

JaredKaragen · Post by **JaredKaragen** » Sun Aug 22, 2010 5:55 pm

If your meaning that grounded to 'the same terminal' means 'grounded to the thermostat housing'. I give up trying... I hope that's not what you meant; because I tried to explain the basics.

jerseydevil · Post by **jerseydevil** » Sun Aug 22, 2010 6:56 pm

THe ecu ground is on the thermostat housing. I'm sorry but i'm not understanding what you're getting at. The all the sensors ground to the ecu via pin d22. The ecu then has THREE pined grounds on a26 a23 and a24. All three of these are tie into Ground G101, including the shields for the shielded wires like the cyp, tdc and ckp sensors etc. G101 is the ground on the thermostat. This is all straight out of the offical 1993 electrical troubleshooiting manual for a civic del sol. As far as this books states, G101 is the one and only ecu ground location. The ecu case, and where it mounts to the car, is not the ground. So yes, I ran my LC-1 ground wire to the thermostat where the rest of the ecu and it's various sensors ground themselves. I checked today when I was out trying ot learn more of this stuff and I had no offset. Now if only I could get my damn LC-1 to program to run dual 0-5v outputs I'd be happy. Damn analog one keeps resetting to narrow band output and it's pissing me off.

jerseydevil · Post by **jerseydevil** » Sun Aug 22, 2010 6:57 pm

Wait.. I reread your post a few mores times. Are you saying grounding the lc-1 output onto d22, where all the other sensors ground is the best solution instead of grounding where the entire ecu grounds?

JaredKaragen · Post by **JaredKaragen** » Mon Aug 23, 2010 2:05 am

jerseydevil wrote:Wait.. I reread your post a few mores times. Are you saying grounding the lc-1 output onto d22, where all the other sensors ground is the best solution instead of grounding where the entire ecu grounds?

Correct, make a 'least path of resistance' situation optimal by connecting within an inch or two of the ecu plug. Or run all shielded wiring direct to the battery, but even then I have a short distance between the lc1 and ecu pin as well (custom harness for tuning cars)