V3 fuel sliders and DA target AFRs

[DSMtuned]

I'm new to tuning with ECMLink V3, so go easy on me...

I have my idle and cruise Mafcomp settings in line and I'm now working on WOT. I'm getting closer to having WOT boostest inline with actual boost between 5000 and 5500 rpms.


The fuel settings, however, are another story. My wideband is leaner than the targeted AFRs in the Direct Access tables. I'm trying to get the wideband and DA tables to match.

My fuel pressure, dead time, and global fuel settings should be set up correctly based on the ECMLink set-up videos.

I plan to only tune using the DA tables and not the sliders. Is it a band-aide to adjust the fuel sliders to get the DA table values to match the actual AFR on the wideband? If so, what else should I check?


Thanks,
Craig

[yokotabrat]

Craig, I'm still new at this as well, but I'll give it a shot.

Have you got your boostest properties setup correctly for altitude?

Which wideband do you have?

Which pin do you have the wideband wired in to the ecu?

How much leaner is the wideband from the DA tables?

How does your wideband match with AFRatioEst in dsmlink?

[DSMtuned]

Craig, I'm still new at this as well, but I'll give it a shot.

Thanks Bert.

Have you got your boostest properties setup correctly for altitude?

Yes the boostest is set-up for 5800ft, since I live next to Bandi.


Which wideband do you have?

I have a PLX M300 wideband.

Which pin do you have the wideband wired in to the ecu?

The wideband is wired into the EGR Temp pin.
How much leaner is the wideband from the DA tables?

The wideband is generally 3 points leaner than the DA Tables.

How does your wideband match with AFRatioEst in dsmlink?

This is also 3 points leaner than the AFRatioEst.


I know that the boostest is most accurately set at 80*. Do you know is that 80* at the MAS or at the throttle body?

-Craig

[yokotabrat]

This is also 3 points leaner than the AFRatioEst.

So 3 by 3 points you mean 14.7 vs 11.7 a/f ratios? Also, you're saying the AFRatioEst matches the DA tables?

I know that the boostest is most accurately set at 80*. Do you know is that 80* at the MAS or at the throttle body?

Are you running a GM maf? If not, then boostest should change based on temperature measured at the factory MAS. If you're heatsoaking the IC, I suppose the boostest numbers could be wrong. If you're using a GM maf, you would need to "adjust" boostest based on temperature if it's above or below 80F. I think you have to assume ambient temp = temp of air going into the engine to make things simple if you don't have an IAT sensor.

-Craig

How old is your PLX wideband? Is there any way to do a free air calibration? Have you verified voltages at the analog output vs what's being measured by the ecu? Maybe there is a ground offset that's causing the reading into the ecu to be incorrect.

[DSMtuned]

So 3 by 3 points you mean 14.7 vs 11.7 a/f ratios? Also, you're saying the AFRatioEst matches the DA tables?

Yes to both. The wideband says 12.1ish and the DA tables/AFRatioEst say 9.0ish.

Are you running a GM maf? If not, then boostest should change based on temperature measured at the factory MAS. If you're heatsoaking the IC, I suppose the boostest numbers could be wrong. If you're using a GM maf, you would need to "adjust" boostest based on temperature if it's above or below 80F. I think you have to assume ambient temp = temp of air going into the engine to make things simple if you don't have an IAT sensor.

I'm running a 2g unmolested MAS. There is no way I'm heatsoaking this giant FMIC. ;) Cool, so it sounds like I don't need to worry about temp variations then.

How old is your PLX wideband? Is there any way to do a free air calibration? Have you verified voltages at the analog output vs what's being measured by the ecu? Maybe there is a ground offset that's causing the reading into the ecu to be incorrect.

My PLX Wideband is about 3 years old. No free air calibration is needed. I haven't measured the output. I could do that, though. I'll have to double check the wiring. The WBO2 readings make sense, though. They are very close to 14.7 at idle and cruise, which leads me to believe they are correct.

[yokotabrat]

My PLX Wideband is about 3 years old. No free air calibration is needed. I haven't measured the output. I could do that, though. I'll have to double check the wiring. The WBO2 readings make sense, though. They are very close to 14.7 at idle and cruise, which leads me to believe they are correct.

Are you running a buffer into the egrt line? It's possible the PLX wideband can't drive the input impedance of the egrt line. Check this page out:

http://www.ecmtuning.com/wiki/ecuinputdata

Bert

[DSMtuned]

Are you running a buffer into the egrt line? It's possible the PLX wideband can't drive the input impedance of the egrt line. Check this page out:

http://www.ecmtuning.com/wiki/ecuinputdata

Bert


Hmm, I'm don't remember. I'll again have to check my wiring. I faintly remember putting in a resistor...

Thanks Bert.

[yokotabrat]

If you need help making one, let me know. You have to use an op amp, so it's not something someone typically has lying around. From your symptoms, it sounds like it could be your problem since cruise AF ratios are okay, but it's not showing rich AF ratios.

[DSMtuned]

If you need help making one, let me know. You have to use an op amp, so it's not something someone typically has lying around. From your symptoms, it sounds like it could be your problem since cruise AF ratios are okay, but it's not showing rich AF ratios.

Sounds good. I'll let you know.

-Craig

[biglady112]

Where do you live craig? What fire department do you work for?

[DSMtuned]

Where do you live craig? What fire department do you work for?

I'm in Lakewood at about Union and Jewell.


I volunteer for Pleasant View Fire Dept. and I also work for Action Care Ambulance. Still testing for career departments and sitting pretty on a couple hiring lists.



-Craig

[TerryLiv]

Craig,

I hope you don't mind me jumping in on your thread. I too have a wide band problem. I have an Innovate LC-1, and pocketlogger. Without manking any changes, I am off over a point from what my gauge shows. I can have the gauge match the logger at 11.7 A/F but only if I offset the value by about 1.25 points and then it isn't accurate at other values. My logger is also connected to the EGR Temp line.

Any others have the same problem?

Terry

[DSMtuned]

Craig,

I hope you don't mind me jumping in on your thread. I too have a wide band problem. I have an Innovate LC-1, and pocketlogger. Without manking any changes, I am off over a point from what my gauge shows. I can have the gauge match the logger at 11.7 A/F but only if I offset the value by about 1.25 points and then it isn't accurate at other values. My logger is also connected to the EGR Temp line.

Any others have the same problem?

Terry

Hi Terry. Nice to hear from you.

Have you tried a buffer on the EGRT input? I installed one and at idle, it seems to have helped. Are you logging WBFactor to see exactly how far off you are?


I am having a new problem where I can no longer connect the laptop to ECMLink. The car runs and drives normally, but for some reason I can't get connected. It may be a Vista com port problem...

-Craig

[TerryLiv]

Hi Terry. Nice to hear from you.

Have you tried a buffer on the EGRT input? I installed one and at idle, it seems to have helped. Are you logging WBFactor to see exactly how far off you are?


I am having a new problem where I can no longer connect the laptop to ECMLink. The car runs and drives normally, but for some reason I can't get connected. It may be a Vista com port problem...

-Craig

Craig,

Thanks for the reply. I am using pocketlogger and not ECMLink, so don't have an equilivent of WBFactor. My method of seeing how far off my logger is compared to my WB gauge (which appears to be accurate) is to look at the Palm and the gauge at the same time and get a good idea of how far off I am. At 11.7 A/F on the gauge it appears to be a little more than one point higher.

Where did you get your buffer? Is it the one made by ECMTuning?

The problems I have had trying to connect my Vista to my wide band for calibration purposes is caused by having HotSync running. If I close it down, I can connect. Innovate tech support solved that for me.

I can't understand why all of the other people on this site that have widebands are not talking about these problems? Is it so universally understood that they don't think it is worth mentioning, or are they ignorant of the problem as I was.

Later,

Terry

[yokotabrat]

Terry,

I also could not get my palm to read correctly with my lc1 compared to my xd16 gauge. That was with pocketlogger and mmcd. I ended up using mmcd and importing the raw data into a csv file and then applying my own formula to convert it to the proper a/f ratios. It was a hassle, but it was the only way that seemed to work for me. Now that I have dsmlink, my wideband reading seems to match. The wideband is still input on the EGRT line without a buffer. I haven't done anything but idle, yet, but that never even matched with my palm.

[DSMtuned]

Terry,

Here is is a link to ECMLink's info regarding the use of buffers: http://www.ecmtuning.com/wiki/ecuinputdata?s=buffer


I made one with this using only one of the op amps: http://www.radioshack.com/product/index.jsp?productId=2062592



-Craig

[TerryLiv]

If you need help making one, let me know.

Bert,

I would take you up on that. It looks like my biggest hurdle is finding a single op amp in quantity one. It is:

Texas Insturments: TLV2371iP In 8 pin DIP

I Have the pin outs figured out. Now just need the IC. Mouser is out of stock. Digi-Key is also out of stock. Can't find in Fort Collins. Lead time for Digi-key is September 2010. I think my best option may be to buy a pre made filter from Thomas.

Thoughts?

Terry

[TerryLiv]

Craig,

You posted while I was writing. Thanks, that should do me. They have it in Ft. Collins. Will let you know how I make out.

Terry

[yokotabrat]

Terry, just be sure to disable the other 3 op amps by tying the - input to the output and the + input to ground. That way you minimize noise induced into the chip.

[TerryLiv]

Terry, just be sure to disable the other 3 op amps by tying the - input to the output and the + input to ground. That way you minimize noise induced into the chip.


Bert,

Is the ground, the global ground for all 4 op amps and is it pin 11? Any other hints are appreciated.

Thanks,

Terry

[DSMtuned]

Bert,

Is the ground, the global ground for all 4 op amps and is it pin 11? Any other hints are appreciated.

Thanks,

Terry


Terry,

Yes, Pin 11 is the ground for all 4 op amps.


Bert, I was wondering what you think of the ground for the op amp. Some people recommend grounding it to a chassis ground vs the sensor ground. Any thoughts on that?

-Craig

[yokotabrat]

Terry, yes pin 11 is the power supply ground for the op amp. This can be tied to the other pins that need to be grounded (+ inputs).

All grounds for the op amp should then be connected to sensor ground on the ecu as that's what the 5V power supply is referenced from. Hope this helps.

[TerryLiv]

Back to my problem.

I have recently installed an Innovate LC-1 wide band with digital gauge. I am trying to log it into my pocket logger. Having trouble getting the pocket logger numbers to match the gauge. I am logging into the EGR-T line.

The gauge seems to be accurate. The logger is significantly off. I can by using the logger's "off set" command, get the two units to match up at one A/F number only.

I was told I need an Input buffer to get the two devices to read the same. I installed one I bought from ECMTuning, but it didn't seem to change anything.

Has anyone had experience with these devices and resolved the problems other than what Bert did by writing code? Craig, did yours really work out?

__________________
Terry,
91 AWD

[yokotabrat]

Terry, what version of pocketlogger are you using? I'm guessing 2.03b?

I looked up info on that version and it assumes you have a PLX wideband, which has a different voltage vs a/f ratio curve than the analog out on the lc-1. Luckily the analog out on the lc-1 is programmable. You need to hook the lc-1 up to your laptop and reprogram the lc-1 for the following parameters:

0V = 10:1
5V = 20:1

Then make sure your offset is 0 again and this should put you a lot closer. The offset will just shift everything up and down, not change the slope like the change above will do.

Here's the plx analog output curve for reference:
http://www.plxdevices.com/products/wideband/WidebandOutputGraph.jpg

Hope this fixes your problem.

[TerryLiv]

Terry, what version of pocketlogger are you using? I'm guessing 2.03b?

I looked up info on that version and it assumes you have a PLX wideband, which has a different voltage vs a/f ratio curve than the analog out on the lc-1. Luckily the analog out on the lc-1 is programmable. You need to hook the lc-1 up to your laptop and reprogram the lc-1 for the following parameters:

0V = 10:1
5V = 20:1

Then make sure your offset is 0 again and this should put you a lot closer. The offset will just shift everything up and down, not change the slope like the change above will do.

Here's the plx analog output curve for reference:
http://www.plxdevices.com/products/wideband/WidebandOutputGraph.jpg

Hope this fixes your problem.

Bert,

Thanks again for helping.

My version of Pocketlogger is 2.03G, newer than B. I had tried the recalibration you suggested before, and tried it again after your suggestion. It doesn't help compared to having the LC-1 set up for the default setting of 7a/f at 0 volts and 22 a/f at 5 v. I am still about 1.25 A/f off at 11.7 A/F compared to being on at 14.7.

According to Pocketlogger, it is supposed to accept a straight line curve the same as LC-1 or the PLX units that Pocketlogger recommends.

I have talked to a tech at Innovate and he is supposed to be asking his engineer for suggestions. Did you read what Hal suggested on my NABR post? I don't quite understand it, but will investigate.

Hal wrote:

You could try using the LC-1 in place of the oem sensor (narrow band simulation).

That would give you the LC-1 data on a decent input port.

Unlike Dave, I've never found the egr temp line to be reliable.

Thanks,

Terry

[Overkill]

Have you tried using mmcd version for palm logging? My understanding is Pocketlogger is different and I have seen correct wide band logging in mmcd.

Just throwing another option out there.

[yokotabrat]

Bert,

Thanks again for helping.

My version of Pocketlogger is 2.03G, newer than B. I had tried the recalibration you suggested before, and tried it again after your suggestion. It doesn't help compared to having the LC-1 set up for the default setting of 7a/f at 0 volts and 22 a/f at 5 v. I am still about 1.25 A/f off at 11.7 A/F compared to being on at 14.7.

According to Pocketlogger, it is supposed to accept a straight line curve the same as LC-1 or the PLX units that Pocketlogger recommends.

I have talked to a tech at Innovate and he is supposed to be asking his engineer for suggestions. Did you read what Hal suggested on my NABR post? I don't quite understand it, but will investigate.

Hal wrote:

You could try using the LC-1 in place of the oem sensor (narrow band simulation).

That would give you the LC-1 data on a decent input port.

Unlike Dave, I've never found the egr temp line to be reliable.

Thanks,

Terry

The only problem with Hal's suggestion is you would need special code in the ecu to run the 5V input into something the ecu is expecting to be 1V. I don't know how to do that, but it's probably already been done on the dsm-ecu group. It's easy with dsmlink, though.

When you are updating your lc-1, is the configuration change taking effect? If you were to go and look at the analog out configuration now, do you see the default setting or the one you programmed in that goes from 10:1 to 20:1. Only reason I ask was I was having problems with my lc-1 where I would change the setting and then disconnect and reconnect and the data was back to its default values. It wasn't saving the data. It's a common problem with the lc-1s.

Bert

[TerryLiv]

Bert,

Good thought, I will check.

Terry

[TerryLiv]

The only problem with Hal's suggestion is you would need special code in the ecu to run the 5V input into something the ecu is expecting to be 1V. I don't know how to do that, but it's probably already been done on the dsm-ecu group. It's easy with dsmlink, though.

When you are updating your lc-1, is the configuration change taking effect? If you were to go and look at the analog out configuration now, do you see the default setting or the one you programmed in that goes from 10:1 to 20:1. Only reason I ask was I was having problems with my lc-1 where I would change the setting and then disconnect and reconnect and the data was back to its default values. It wasn't saving the data. It's a common problem with the lc-1s.

Bert

Bert,

Hal posted again and said his idea would only work if I had Link, so eliminates trying to figure that one out.

My LC-1 does change when I reprogram the output to different voltage/ A/F values. I see the changed value when I go back to check. Innovate must have fixed that bug.

Does anyone know if MMCD is able to output a/f rather than voltage or Lambda? I had read that at one time it wasn't.


Terry

[TerryLiv]

Have you tried using mmcd version for palm logging? My understanding is Pocketlogger is different and I have seen correct wide band logging in mmcd.

Just throwing another option out there.

Jared,

Will my pocket logger cable function correctly with the mmcd software?

Thanks,

Terry