:::RHMT::: Real Home Made Turbo
General Category => Engine Management => Topic started by: ryan89crx on March 11, 2009, 06:20:33 PM
-
is it normal for the gauge and the datalog to read different values? if so, how much difference is acceptable?
in vac, the gauge reads mid 15s, but the datalog reads mid 14s. in boost the gauge will read mid 11s, but the datalog mid 12s.
which one should i trust? is there a way to calibrate either?
-
Id listen to my MAP sensor
-
Id listen to my MAP sensor
im talking AFRs, not vac/boost values
like for instance, sitting at an idle the gauge will read 15.5afr, but the active datalog will read mid-high 14s. the datalog gets its AFR readings from the o2 input, which comes directly from the PLX box
-
you will never get it perfect with crome. Especially when you have to scale a 5v signal to 3.8. There is a offset you can use to get it right, but dont expect it to be perfect.
-
you will never get it perfect with crome. Especially when you have to scale a 5v signal to 3.8. There is a offset you can use to get it right, but dont expect it to be perfect.
so basically trust the gauge and then adjust the scalars to get the datalog as close as possible
-
Ajust offset here
(https://realhomemadeturbo.com/forum/proxy.php?request=http%3A%2F%2Fi188.photobucket.com%2Falbums%2Fz204%2Flilpooh21186%2FCapture.jpg&hash=d6ace68c0529919f5803133ad69d45bf28750de0)
-
Ajust offset here
(https://realhomemadeturbo.com/forum/proxy.php?request=http%3A%2F%2Fi188.photobucket.com%2Falbums%2Fz204%2Flilpooh21186%2FCapture.jpg&hash=d6ace68c0529919f5803133ad69d45bf28750de0)
ya, i got that. just wasnt sure which number to believe in the beginning.
thanks
-
How did you connect your white wire into d14? Did you use those capacitors? I've noticed that if you don't have a real solid connection, your shit will vary A LOT.
-
i cut the wire in half and connected the WB to D14
-
How did you connect your white wire into d14? Did you use those capacitors? I've noticed that if you don't have a real solid connection, your shit will vary A LOT.
no resistors. i searched around before i installed it 2 years ago and it was pretty much a 50/50 split on people using the resistors or not. everyone that didnt use them said they had no problems. and since im a lazy fuck, i didnt install them.
do you use them? recommend them? ill have to go to Radio Shack and find some since im sure i tossed mine
-
i dont use them
-
i dont use them
datalog and gauge pretty close after adjustment in Crome?
-
Crome is about .2afr leaner with a setting of -0.05 but thats as close as i can get it.
-
They're not resistors, but capacitors. No, I've never used them. I was just wondering.
My AFR's are usually pretty damn close. Something I've noticed is that it varies from car to car. Might be something with the electrical noise going to and from the ecu? I don't know. I may be just talking out of my ass.
-
They're not resistors, but capacitors. No, I've never used them. I was just wondering.
My AFR's are usually pretty damn close. Something I've noticed is that it varies from car to car. Might be something with the electrical noise going to and from the ecu? I don't know. I may be just talking out of my ass.
ehh, ill just mess with the corrections and see what happens
i run my power from an open spade on the fusebox, and its grounded on the chassis. the output wire on the PLX is wired to te o2 input for the ecu right before the obd1 conversion harness
-
ground the ecu case to the chassis
-
you will never get it perfect with crome. Especially when you have to scale a 5v signal to 3.8. There is a offset you can use to get it right, but dont expect it to be perfect.
WTF???
Get your power and ground from the battery. No where else. After that, use the offset to get the numbers close. Grounding the wideband's wires properly and setting the offset you should be able to get consistent results.
-
you will never get it perfect with crome. Especially when you have to scale a 5v signal to 3.8. There is a offset you can use to get it right, but dont expect it to be perfect.
WTF???
Get your power and ground from the battery. No where else. After that, use the offset to get the numbers close. Grounding the wideband's wires properly and setting the offset you should be able to get consistent results.
oh shit! bignig returns!!
thanks for the input
-
you will never get it perfect with crome. Especially when you have to scale a 5v signal to 3.8. There is a offset you can use to get it right, but dont expect it to be perfect.
WTF???
Get your power and ground from the battery. No where else. After that, use the offset to get the numbers close. Grounding the wideband's wires properly and setting the offset you should be able to get consistent results.
Yep. I always connect my wideband directly to the battery. But on MY car, I get some of the weirdest readings than on any other car. I guess I need to add some more grounds or something.
-
Most Civics have the same 15yr old groundstrap-ons that Honda installed on the car at the factory. That shit corrodes over time - very bad.
Before someone swiped my M-300, I used it in Crome and Uberdata. The offset was minor - I used the caps 0.25" from the case, and got the power right from the battery. Remember that connectors can create resistance and capacitance, and long unsheilded wires can mess with the signal.
FYI the stock O2 circuitry DOES NOT scale a 5v signal to 3.8v -- it just won't read beyond 3.8v, which is ~17:1 and higher. Its useless anyways. Who the hell tunes past 17:1 anyways?? The only time a full 5v helps is if you have a programable output like an LC-1 and change it so it reads from 10:1 to 17:1, instead of up to 22:1.
-
Most Civics have the same 15yr old groundstrap-ons that Honda installed on the car at the factory. That shit corrodes over time - very bad.
Before someone swiped my M-300, I used it in Crome and Uberdata. The offset was minor - I used the caps 0.25" from the case, and got the power right from the battery. Remember that connectors can create resistance and capacitance, and long unsheilded wires can mess with the signal.
FYI the stock O2 circuitry DOES NOT scale a 5v signal to 3.8v -- it just won't read beyond 3.8v, which is ~17:1 and higher. Its useless anyways. Who the hell tunes past 17:1 anyways?? The only time a full 5v helps is if you have a programable output like an LC-1 and change it so it reads from 10:1 to 17:1, instead of up to 22:1.
Yeah, I've never fully understood why people want to read all the way to 22:1.
-
Your wideband controller is sending an analog signal in reference to ground, and displaying that same signal scaled as a readout.
Your ecu is taking that analog signal, referencing it's ground, and that is what you are datalogging.
If there is any resistance/corrosion/float between the two grounds, the readings will be different. People think ground means 0 volts, but grounds from different places in an electrical circuit may have 'voltage potential' between them. ie measure the tps output from it's ground pin may read 0.45V, but measure it to the battery - it may read 0.5V, measure it from the door panel bolt it may read 0.4V it all depends on how well and where things are grounded.
The best thing to make a wideband datalog through the ecu accurately in relation to a gage or pc, is to make sure they are grounded at the same point. A note the ecu is not grounded at the battery.
In my LC-1 that has separate grounds, I grounded the heater circuit to the battery directly to keep the noisy circuit isolated. Then I grounded the analog ground to the ecu ground. Through freelog, and then ectune both of my readings are very close to being the same, because they are both in reference to the same 'zero' measuring point.
-
Your wideband controller is sending an analog signal in reference to ground, and displaying that same signal scaled as a readout.
Your ecu is taking that analog signal, referencing it's ground, and that is what you are datalogging.
If there is any resistance/corrosion/float between the two grounds, the readings will be different. People think ground means 0 volts, but grounds from different places in an electrical circuit may have 'voltage potential' between them. ie measure the tps output from it's ground pin may read 0.45V, but measure it to the battery - it may read 0.5V, measure it from the door panel bolt it may read 0.4V it all depends on how well and where things are grounded.
The best thing to make a wideband datalog through the ecu accurately in relation to a gage or pc, is to make sure they are grounded at the same point. A note the ecu is not grounded at the battery.
In my LC-1 that has separate grounds, I grounded the heater circuit to the battery directly to keep the noisy circuit isolated. Then I grounded the analog ground to the ecu ground. Through freelog, and then ectune both of my readings are very close to being the same, because they are both in reference to the same 'zero' measuring point.
hmm Good to know
-
Your wideband controller is sending an analog signal in reference to ground, and displaying that same signal scaled as a readout.
Your ecu is taking that analog signal, referencing it's ground, and that is what you are datalogging.
If there is any resistance/corrosion/float between the two grounds, the readings will be different. People think ground means 0 volts, but grounds from different places in an electrical circuit may have 'voltage potential' between them. ie measure the tps output from it's ground pin may read 0.45V, but measure it to the battery - it may read 0.5V, measure it from the door panel bolt it may read 0.4V it all depends on how well and where things are grounded.
The best thing to make a wideband datalog through the ecu accurately in relation to a gage or pc, is to make sure they are grounded at the same point. A note the ecu is not grounded at the battery.
In my LC-1 that has separate grounds, I grounded the heater circuit to the battery directly to keep the noisy circuit isolated. Then I grounded the analog ground to the ecu ground. Through freelog, and then ectune both of my readings are very close to being the same, because they are both in reference to the same 'zero' measuring point.
For a long time I grounded the wideband at the ECU. Grounding at the ECU was accurate at some spots and then the difference became more noticeable at other voltages. Grounded at the battery, it needed some scalar correction but once the scalar was set, the results were linear. The ECU is well grounded because it has multiple grounding points. Isolating any singular wire can be problematic.
In theory, what you are saying is 100% true. In reality, you need to test and find out what works best.
-
I tested multiple grounds vs battery (and a few to each other), and there was almost no difference at idle or 5k rpm. Unfortunately I don't have a dyno to test during load, and don't have issues to warrant testing it on the road. I can always hook up my SSI-4 if I need...I love that thing.