A community of enthusiasts dedicated to Nissan's SR20DE/SR20VE/SR20DET engines.
Start here: forum search. Be sure to search on what you're looking for before posting a new thread.
You are currently browsing the forum as a guest. In order to access special features, the image gallery, and post you will need to be a registered member. Registration is fast, simple and absolutely free so please, join our community today!
If you have any problems with the registration process or your account login, please contact the administrator.
Sorry if this has been addressed already but I was considering purchasing a wideband O2 sensor that has a programmable analog output (0-5V) that i can configure to the sr maps. That way i can just run one 02 sensor, the wideband one, get accurate readings, use the stock bung, and not have to spend a fortune going to the dyno over and over again. Does this sound like a reasonable idea to you? Have you or any of your customers done this? Any suggestions on how to calibrate the analog output? I have searched and the stock o2 sensor seems to oscillate betwween 0 - .9V. thanks
This is the model i was looking at. costs the same as 2 hours of dyno tuning.
AEM Gauge-Type Wideband UEGO Controller $279
nate
__________________
Everything must go... im retiring for a while.
Last edited by sr20sleeper : 07-24-2006 at 03:10 PM.
It is a very reasonable idea. Choaderboy2 also suggestes that you should use YOUR WB/O2 when on a dyno instead of the tail-entry WB of the dyno.
For it to work on the OEM/Calum ECU you do not have to do/calibrate anything. Most Widebands (including the AEM AFAIK) do have a "narrowband" output option which you use to "feed" the ECU so it works OK.
What you do need to do however if you are going to use it for tuning, is somehow correlate the datalogs from the UEGO to the actual RPM/load points of the ECU. How you can do that is beyond me!
I guess i was thinking that the stock narrow band just goes 0-1V right? So if you can scale the narrow band output on the aem to just register 0-1V wouldnt that tell you the A/F ratio accurately using the ECU consult? When you datalog the 02 sensor voltage is recorded right?
I guess i was thinking that the stock narrow band just goes 0-1V right? So if you can scale the narrow band output on the aem to just register 0-1V wouldnt that tell you the A/F ratio accurately using the ECU consult? When you datalog the 02 sensor voltage is recorded right?
Yes, this will work. If your using an LM-1 you can set the analog output to 0-1V in a linear curve (not simulated narrowband), with 14.7:1 at .5V. Then when you datalog the O2 sensor via the consult port (with the wideband feeding the ecu) you can back-out the afr in your datalogs. I modified a version of CalumSult to do just that on the fly, but never posted it. I need to post a more recent version of the source files so you guys can do this stuff on your own.
Innovate posted the specs of their communications protocol, what would be better is to just use that and datalog directly rather than piping through the ecu. I've never had the time to write software to do this.
__________________
All posts by Calum in this thread are copyrighted by Calum, and cannot be edited, deleted, or copied without his express permission. calum@calumsult.com
Yes, this will work. If your using an LM-1 you can set the analog output to 0-1V in a linear curve (not simulated narrowband), with 14.7:1 at .5V. Then when you datalog the O2 sensor via the consult port (with the wideband feeding the ecu) you can back-out the afr in your datalogs. I modified a version of CalumSult to do just that on the fly, but never posted it. I need to post a more recent version of the source files so you guys can do this stuff on your own.
Innovate posted the specs of their communications protocol, what would be better is to just use that and datalog directly rather than piping through the ecu. I've never had the time to write software to do this.
the reason i ask about the aem specifically is because i can get a really good deal on the uego right now. and it is on island so i dont have to pay for shipping. So is the stock O2 not a linear curve then?
Let me clarify by what I meant by 'linear'. The stock is a normal narrow band, so the output voltage curve looks like a 'S' (with the 'switch' point around 14.7:1). You want a nice straight line (in both cases plotting voltage vs afr).
If the AEM can scale the output like this then you can do the same trick, I just refer to an LM-1 because thats what I've got. If not, its not really a big deal, you can work around it.
I see, so the stock sensor uses 14.7x as a reference and the voltage on either side of that point will swing quite a bit. Will the linear curve throw off the fuel consumpion of the car very much? Im assuming it will be similar in some places but off in others. WOT doesnt use the O2 anyways right? Sorry for all of the questions, im just trying to get the whole picture here.
The AEM does not have a linear output from 0-1V. The analog output that goes from 0-1V looks like a narrowband but should be a lot more accurate right? I plotted the values from the instructions, the curve looks like this. So i would want values of .8-.84 to be between 12.75 and 13.5 a/f
Last edited by sr20sleeper : 08-09-2006 at 06:39 PM.
I am interested in the answer to some of sr20sleepers questions.
1) Would setting the voltage graph of the simulated narrowband to 'linear' effect gas consumption?
2) Do you have the version of CalumSult you created sitting on a repository somehwere? Or, whatever you windows guys call stuff. ftp server.
That might be the best option.. it has RPM input for the controller, so you don't need to run it through the ECU. You can datalog AFR vs RPM right out of the box.
That might be the best option.. it has RPM input for the controller, so you don't need to run it through the ECU. You can datalog AFR vs RPM right out of the box.
^^^^that sounds pretty sweet. I know a lot of guys on the forum recommend that one.
I am using the AEM UEGO because i basically got it for free. Hooked it up last night, its really easy to do. 3 wires is all it took.
First, let me say that the calum program looks awesome. I played with it on my ride to work today, 14.0-15.3 at partial throttle cruising around. Once i got on the gas the thing pegged at 13.2-13.5 all the way up to the redline.
The uego has a serial output to datalog but it does not have a rpm input like that other one you posted. I chose to replace my stock 02 sensor with the AEM one, i didnt want to deal with welding bungs and what not. So i am running it in the 0-1v mode that simulates the narrow band (nernst i think its called) and it goes through the stock 02 sensor wiring harness. Since it is taking a very accurate input value and converting it to approximate this nernst curve... I am confident in just transforming the output using the calibration curve and datalogging through calumsult. Since 13.5:1 is what im shooting for, i will tune until the values from the 02 sensor are .8 consistently at WOT since .8v corresponds to 13.5:1 in the calibration table. I will datalog some tonight after i clean up the install a little. I will try to post some screen captures of what i get and maybe calum can chime in if he has time (i know he is super busy).
^^^^that sounds pretty sweet. I know a lot of guys on the forum recommend that one.
I am using the AEM UEGO because i basically got it for free. Hooked it up last night, its really easy to do. 3 wires is all it took.
First, let me say that the calum program looks awesome. I played with it on my ride to work today, 14.0-15.3 at partial throttle cruising around. Once i got on the gas the thing pegged at 13.2-13.5 all the way up to the redline.
The uego has a serial output to datalog but it does not have a rpm input like that other one you posted. I chose to replace my stock 02 sensor with the AEM one, i didnt want to deal with welding bungs and what not. So i am running it in the 0-1v mode that simulates the narrow band (nernst i think its called) and it goes through the stock 02 sensor wiring harness. Since it is taking a very accurate input value and converting it to approximate this nernst curve... I am confident in just transforming the output using the calibration curve and datalogging through calumsult. Since 13.5:1 is what im shooting for, i will tune until the values from the 02 sensor are .8 consistently at WOT since .8v corresponds to 13.5:1 in the calibration table. I will datalog some tonight after i clean up the install a little. I will try to post some screen captures of what i get and maybe calum can chime in if he has time (i know he is super busy).
oh nice, good stuff. is this the same tune I have? DE MAF, VE inj, sr20ve tune?
Datalogged some runs with the new O2 sensor. getting some interesting results. perhaps Calum will have some idea whats going on. I watched the a/f gauge as i accelerated, it stayed in the 12.5-13.1 range most of the way up to redline. however, when i looked at the datalog... the O2 voltage is registering .9x most of the way. Im not sure why it is doing this, according to the specs for the aem the lowest a/f it will read is 11:1 and that corresponds to .90V. So im really not sure how i am getting .96v and showing 12.5:1 on the gauge. here are a few of the charts.
Last edited by sr20sleeper : 08-11-2006 at 05:46 AM.
I am interested in the answer to some of sr20sleepers questions.
1) Would setting the voltage graph of the simulated narrowband to 'linear' effect gas consumption?
2) Do you have the version of CalumSult you created sitting on a repository somehwere? Or, whatever you windows guys call stuff. ftp server.
Sorry, really tired.
Thanks in advance Calum.
Sorry for taking so long to answer, I was away from the forum last week.
1 - I don't think so, but I've never measured. I wouldn't leave it like that all the time.
2 - No, but I need to do that. I started work on a newer version, but kinda broke it. I need to fix it, or maybe just post it up and let someone else fix it.
(Falmouth)
