PDA

View Full Version : Afr spikes in BBL only



alian
July 31st, 2008, 09:22 PM
Hi guys, Been having trouble with my V2 when downloading the bbl file to the laptop.I have only had the V2 for about a month and this is the 1st time I have played with anything like this. But have been in contact with Simon(swingtan),Nate, Jezzab and Vx leather. I am running a Techedge wideband and that is where we thought the problem was,(in the set up).But after many attempt to recal the O2 to talk to the V2,we have discovered that when the laptop is linked with the V2 and the car and logging through to the computor it works fine(No afr spikes).Unhook the laptop(nothing else) and start BBL, download the file and :doh2:,the spikes are there. Have removed as many pids from the optifie as we thought that there was to many. Tried 2 SD cards and another V2. Still happening. Would love to hear if anyone out there has a theroy or something else that I could try. Many Thanks Ian

joecar
August 1st, 2008, 02:04 AM
Ian, post before/after log files.

scdyne
August 1st, 2008, 02:25 AM
This sounds like a floating ground.
Please post the log as joecar said and we'll see what the deal is.
I built 2 Te WB2.0 controllers so I have schematics and experience with them. While I have never connected one to my V2 (I have an LC-1 now) I have intimate knowledge of how they do and don't work. If I understand what you are saying here, when you don't have the laptop connected you see spikes in the log file. In my experience not having a good ground can cause "noise" spikes, but having the laptop connected would provide a better ground through the serial connector.

Simple check would to be either run an additional ground from the Te to the chassis or direct to the car battery.

joecar
August 1st, 2008, 05:05 AM
It seems like the connection to the laptop may be completing the ground path.

alian
August 1st, 2008, 08:29 AM
QUOTE=joecar;75867]Ian, post before/after log files.[/QUOTE]

3691

Just tried to attach file, hope it worked. This is a file of going arround the block whilst testing yesterday. Ian

alian
August 1st, 2008, 08:32 AM
This sounds like a floating ground.
Please post the log as joecar said and we'll see what the deal is.
I built 2 Te WB2.0 controllers so I have schematics and experience with them. While I have never connected one to my V2 (I have an LC-1 now) I have intimate knowledge of how they do and don't work. If I understand what you are saying here, when you don't have the laptop connected you see spikes in the log file. In my experience not having a good ground can cause "noise" spikes, but having the laptop connected would provide a better ground through the serial connector.

Simple check would to be either run an additional ground from the Te to the chassis or direct to the car battery.

Ok, I thk I get what you mean. So where do I attatch the extra earth to on the V2 so I can test it? Thanks Ian

vxleather
August 1st, 2008, 10:03 AM
refering to the ground for the tech edge not the v2

alian
August 1st, 2008, 12:40 PM
refering to the ground for the tech edge not the v2

Thanks Shane will look into this this afternoon. will take some time to get to the wideband where it is wired in

vxleather
August 1st, 2008, 01:39 PM
Guys got a chance to take a photo of Ian going out to work on his techedge this arvo... I reckon the techedge will loose hehe....

http://i222.photobucket.com/albums/dd200/vxleather/ianandhislikefortechedgewidebands.jpg

alian
August 2nd, 2008, 09:17 AM
[QUOTE=vxleather;75903]Guys got a chance to take a photo of Ian going out to work on his techedge this arvo... I reckon the techedge will loose hehe....

Wireing in another earth this morning to the o2. Fingers crossed.

ntae
August 2nd, 2008, 06:27 PM
We reloaded the Boot & firm in the V2 . Formated the V2 config , internal & Sd card and all is good now :cucumber:

alian
August 2nd, 2008, 09:35 PM
We reloaded the Boot & firm in the V2 . Formated the V2 config , internal & Sd card and all is good now :cucumber:

Yep, who would have thought, Thanks again Geoff. Ian

vxleather
August 2nd, 2008, 09:38 PM
just gota work out where to put all that acurate scaned info now..

Blacky
August 9th, 2008, 08:01 PM
The value that it "spikes" to is significant. The FlashScan firmware will set the AFR to 25.00 whenever it reads a data packet from the wide band controller that is not valid AFR data.

For example the PCM may have gone into deceleration fuel cutoff and the WO2 controller may be outputting a "free air" value rather than the AFR value.

Try logging the EXT.WO2STATE PID(s) as well to see what the data state is when the value goes to 25.00

What you are seeing may be legitimate - in which case you should use the data filters to filter out frames where AFR=25.00.

Regards
Paul

ntae
August 9th, 2008, 08:33 PM
Both people there V2 units work in passthrough ok . its just when you use BBL there is the problem . Both of these units had the same problem with the same techedge 2J1 unit a week ago and mine was ok & know of about 12 others that BBL with serial PLX & Techedge with no problems . I got the first one working as per my post . VX did not format the V2 internal memory. i think that is his problem today . He is on to it as we type


Geoff

vxleather
August 10th, 2008, 09:02 AM
ok have tried reformating the memory, procedure went full format of both memories, then re load boot block, then re load firmware and re formate it all again.. did this 3 times and it did not fix teh problem...

Blacky can you do aquick explanation of what it means when it hits 25 afr... i really dont understand... Free Air..???

And why do we see it in BBL only and not Pass threw, or when using non serial ????

Blacky
August 10th, 2008, 11:27 AM
Try logging the EXT.WO2STATE PID(s) as well to see what the data state is when the value goes to 25.00

Did you try this?

What this will tell you is whether the data being sent from the WO2 device is sending valid AFR data and FlashScan is not reading it correctly, or whether the WO2 device is sending invalid data which causes FlashScan to log 25 as a place holder for the invalid data.

Regards
Paul

vxleather
August 10th, 2008, 11:31 AM
Will give it a try as soon as i get a chance thanks for the promt..

Shane

vxleather
August 10th, 2008, 02:01 PM
Okay cant find the pid you are talking about!! Blacky,

I did find "External WideBand Sensor State" EXT.W02ST1 is this what you mean, or am I on the wrong path here...

Have attached a log with this pid being logged...

Shane

Blacky
August 10th, 2008, 03:41 PM
Yes, that was the PID I meant.

See how that PID goes to zero and the AFR goes to 25.00. The AFR only goes to 25.00 because the STATE PID is not "Lambda".

Now, the question becomes why does the state change from "Lambda" to something other than Lambda. The only reason I can see is that junk data is being received by FlashScan. Junk data could be caused by noise on the serial line or a bad cable, specifically the ground pin.

1. Make sure you are using a shielded cable.

2. Make sure the ground pin is connected properly between TE's serial port and FlashScan's serial port.

3. Try running a three way ground wire to these points:
- Pin 5 of the OBDII connector.
- The ground pin on TE's serial port.
- The ground pin on FlashScan's serial port.

Regards
Paul

vxleather
August 10th, 2008, 03:52 PM
Ok TE = Tech edge?? I am using a PLX I guess the same principle applies... thanks for you advice will go and see what i can do..

Shane

Blacky
August 10th, 2008, 04:57 PM
Blacky can you do aquick explanation of what it means when it hits 25 afr... i really dont understand... Free Air..???

Some controllers (i.e. Innovate) will stop sending AFR data and instead start sending other data if/when it is more appropriate.
For example, if the LC-1 detects free-air (i.e. the O2 sensor detects the same O2 inside and outside the exhaust, then it transmits %O2 instead of AFR. In that case, there is no AFR to log, so FlashScan substitutes a known value (i.e. 25.00) to indicate that the AFR is not valid.

I think the Tech Edge also transmits data other than AFR - but I can't be certain, I don't have the documents here with me right now.

I'm not sure what the PLX sends when there is no AFR to send. Maybe it just trends towards some upper bound when it detects free air.


And why do we see it in BBL only and not Pass threw, or when using non serial ????

As has been pointed out by others in this thread it is possible that is a hardware issue that is only manifesting when the laptop is not connected. I guess a voltage measurement between the WO2 controller's serial ground pin and FlashScan's serial ground pin (when they are not connected via a serial cable) could shed some light on any potential ground offset problems.

When logging analog AFR, ground offsets show up as +/- errors in the AFR reading. Compare that with a digital signal that is either correct or not, there is no +/- error.
So where you might see a error of +/-0.1V in the analog signal that would just simply translate into about a 2% AFR error.
When logging digital/serial data that voltage error may be enough to cause the serial signal to be corrupted. In that case the entire data packet is discarded and FlashScan will substitute the known value of 25.00 AFR.

Having said all that, I would not expect to see any data corruption in a correctly operating serial wideband setup with FlashScan.

Regards
Paul

vxleather
August 10th, 2008, 05:41 PM
Ok we have had 2 V2's spiking... We have managed to fix one being a software issue.. this unit belonging to Alian... now I have carried out the same fixs to my unit.. so i will duck round and see him and test my unit in his car.. if it works fine, I will take it that my wirring is suspect and needs further investagation. I will also try his unit in my car, as we know it is operating properly and also runs a older versioin in the Firmware and bootblock than mine...

I tried going back to ver .17 of the Firmware this arvo and it just stayed in dead poll and would not let me load the older firmware... it gave a error messge.

So by doing this we should be able to narrow it down to a Hardware or software issue..

alian
August 10th, 2008, 07:22 PM
Ok we have had 2 V2's spiking... We have managed to fix one being a software issue.. this unit belonging to Alian... now I have carried out the same fixs to my unit.. so i will duck round and see him and test my unit in his car.. if it works fine, I will take it that my wirring is suspect and needs further investagation. I will also try his unit in my car, as we know it is operating properly and also runs a older versioin in the Firmware and bootblock than mine...

I tried going back to ver .17 of the Firmware this arvo and it just stayed in dead poll and would not let me load the older firmware... it gave a error messge.

So by doing this we should be able to narrow it down to a Hardware or software issue..

Just got home Shane, come round and grab it to give it a go.Ian

TAQuickness
August 12th, 2008, 07:27 PM
VX - Quick recap:

Pass thru BBL and/or analog is fine
BBL analog is fine
BBL serial spikes

Have you tried BBL serial and analog?
Can you post a screen shot of the spikes?

vxleather
August 13th, 2008, 10:36 AM
Yeah Ta that is correct... I was running Analog it worked fine... I then built a Serial cable.. It works fine in pass threw... But not in BBL..

I have not changed anything in the power source to the PLX.. same earth same power feed as prevouise...

I am at a loss now to what it is.. what can the laptop realy be changing, it is not conected to the car in any way...

Screen shot and attachment of log...

http://i222.photobucket.com/albums/dd200/vxleather/spike25afr.jpg

Blacky
August 13th, 2008, 10:47 AM
When logging serial AFR in standalone mode, as soon as you see a spike occur, please stop the data log immediately and save a trace file and email it to me.
The reason you need to stop it immediately is because FlashScan only keeps the last 5-10 seconds of seconds of data in its internal buffer.

To save a trace file: F2 Scan Tool -> F3 Options -> F1 Save Trace File
Then, using EFILive_Explorer, copy the trace file from FlashScan to your PC and email it to me.

That trace file will show the raw serial data that is being received by FlashScan. I may be able to see what is happening from that.

Regards
Paul

vxleather
August 13th, 2008, 11:50 AM
Thanks blacky i will try and get this sorted tonight... when the wife brings the taxi / shopping trolley home from work...

Shane

vxleather
August 13th, 2008, 11:11 PM
okay email sent Blacky,,

TAQuickness
August 13th, 2008, 11:21 PM
Also, what is the peak value of the spike? Looks like it's going off the chart a bit.

While Blacky is sorting the trace file, would you have an opportunity to swap PLX devices,
serial cables, or V2's with the other guy that was having the spiking issue? Would be good to know what device the problem follows.

TAQuickness
August 13th, 2008, 11:21 PM
Also, what is the peak value of the spike? Looks like it's going off the chart a bit.

While Blacky is sorting the trace file, would you have an opportunity to swap PLX devices
Serial cables
V2's
with the other guy that was having the spiking issue? I'd like to see if the proble follow's the PLX or the cable.

vxleather
August 14th, 2008, 01:02 AM
TA the other guys issues was with a techedge 2 unit.... the values are going to 25 afr, and the WO2ST1 pid changes form Lamba to unkown when it spikes to 25. Jezzab when he first saw it commented that it looked like the PLX was losing power...

Once again i said in return .... only does it in BBL serial.. put it into pass threw or analog and all is okay.... I noticed that the pass threw AFR line holds a much smother look about it... where as in the bbl mode it is very very jaged... dont know if this means anything...

I am sort of getting to the point of just putting it back to analog and leave it alone... to hard basket is what it feels like at the moment..

joecar
August 14th, 2008, 02:13 AM
When in passthru mode, how is the laptop being powered...?

TAQuickness
August 14th, 2008, 02:31 AM
I'm suspecting there may be something marginal about your cable or the PLX device that the laptop is absorbing while in passthru mode.

Do you have the means to construct a new cable?

Blacky
August 14th, 2008, 08:15 AM
I can clearly see the faulty data.

I think the problem occurs when FlashScan is writing logged data to its internal memory. (Which does not happen in pass-through mode)

Are you using an SD card to log data? If not, can you try that please.

Also can you please try selecting the "Display data" instead of "Record data" and monitor the WO2 PID to see if it shows any spikes.
If it does, please sop the display and save the trace file.

Regards
Paul

swingtan
August 14th, 2008, 08:43 AM
I was talking to Alian last night and had a look through his latest log. There was still evidence of WB issues with the log data, though it was a little different.

With the log file I saw, the WB data would regularly flat line before continuing on again. there were a few spikes, both rich and lean. The spikes in this case only occurred after a time of the WB data flat lining. My thoughts on this are...


Intermittent power supplied to the WB controller.
Intermittent connection in the WB controller to sensor cable.
Intermittent connection in the WB controller to V2 serial cable.


Of course that doesn't explain why the data seems fine in pass-through mode, unless there is some sort of filtering that occurs in pass-through that id not applied in BBL mode.

Is there any chance that the A/D filter settings get applied to the serial data when in pass through, but not in BBL?

All that being said, I'm running a TE unit with the V2 and have only had issues with intermittent connections. Never any interfacing issues.

Simon.

Blacky
August 14th, 2008, 09:22 AM
Of course that doesn't explain why the data seems fine in pass-through mode, unless there is some sort of filtering that occurs in pass-through that id not applied in BBL mode.

Filtering is applied to A/D inputs as per the ADFilter setting in FlashScan.
Filtering is not applied to serial data.


Is there any chance that the A/D filter settings get applied to the serial data when in pass through, but not in BBL?


The difference between pass-through and black box logging is that during bbl data is written to the SD card or internal memory.

Writing to the SD card is a non-intrusive process and should not affect the reception of serial data. I do not expect writing to the SD card to affect serial data reception.

Writing to the internal memory is a little more tricky. It is in intrusive process and it is possible that serial data is being dropped when FlashScan is writing to internal memory. An interesting observation would be: Does the serial AFR flatline/spike only when the blue LED is illuminated? (i.e. when data is being written to flash memory).

Regards
Paul

vxleather
August 14th, 2008, 10:32 AM
okay my head is officaly spinning now hehe...

Joecar to answer your ?? it is power via its battery, no connection anywhere to the car...

Blacky I will attempt to do what you asked this arfternoon... Ntae is coming over and going to give me a hand.. he is a little more technicaly minded than me.. in the electronics department.

Shane

Blacky
August 14th, 2008, 01:36 PM
I am making a firmware change that will detect if/when the received data is corrupt. FlashScan will discard the data and resynchronize with the PLX. It may result in one or two frames of AFR data being carried over from the previous frame to cover for the discarded data, but it should eliminate the data spikes.

Regards
Paul

vxleather
August 14th, 2008, 01:54 PM
okay, you still want the the trace files done this arfternoon???

Blacky
August 14th, 2008, 02:13 PM
Yes, but with the new firmware, I'll email it to you as soon as it is done.
Paul


okay, you still want the the trace files done this arfternoon???

vxleather
August 14th, 2008, 02:17 PM
okay thanks, I might be able to do some loging after all for our trip to townsville over the weekend... okay i am getting a little excited again...

vxleather
August 14th, 2008, 03:58 PM
Blacky, I have just uploaded the update you sent... and noticed this again..i remeber in the past you asked us to delete the "settings" folder out of the config file...(cant remeber why) I noticed mine keeps appearing even after I have deleted it...

http://i222.photobucket.com/albums/dd200/vxleather/untitled12.jpg

Blacky
August 14th, 2008, 04:05 PM
Blacky, I have just uploaded the update you sent... and noticed this again..i remeber in the past you asked us to delete the "settings" folder out of the config file...(cant remeber why) I noticed mine keeps appearing even after I have deleted it...

FlashScan will recreate it with the install defaults if it is not there. So you should expect it to re-appear after you delete it.

The only times you would need to delete it are:
1. When a new firmware is released that uses a different format for the settings file (hopefully that won't need to happen again).
2. If you want to reset all the settings to their defaults.

Regards
Paul

ntae
August 15th, 2008, 08:09 AM
My V2 works in Shane's car with the PLX serial setup ok .as soon as we load 8.1 22/6 the problems starts . the update you sent yesterday is showing problems as well , shane is sending the trace file to you . is there a back door to reload pre 8.1 22/6 untill the problems are worked out

vxleather
August 15th, 2008, 08:36 AM
Trace file and logs sent last night Paul, ntae thanks for your efforts last night much appreciated.

Shane

alian
August 15th, 2008, 02:59 PM
I was talking to Alian last night and had a look through his latest log. There was still evidence of WB issues with the log data, though it was a little different.

With the log file I saw, the WB data would regularly flat line before continuing on again. there were a few spikes, both rich and lean. The spikes in this case only occurred after a time of the WB data flat lining. My thoughts on this are...


Intermittent power supplied to the WB controller.
Intermittent connection in the WB controller to sensor cable.
Intermittent connection in the WB controller to V2 serial cable.


Of course that doesn't explain why the data seems fine in pass-through mode, unless there is some sort of filtering that occurs in pass-through that id not applied in BBL mode.

Is there any chance that the A/D filter settings get applied to the serial data when in pass through, but not in BBL?

All that being said, I'm running a TE unit with the V2 and have only had issues with intermittent connections. Never any interfacing issues.

Simon.

Did another log and, sent through to Simon, No spikes or flat lines this time. might have been a bad conection. Will keep an eye on it. Ian

Highlander
May 16th, 2009, 08:12 AM
I am seeing this and I am suspecting a sensor timing problem... I will exchange sensors and see... my problem comes on pass thru though...

Thanks

joecar
May 16th, 2009, 08:25 AM
Post pics/screenshots.

Highlander
May 16th, 2009, 08:57 AM
Here it is... its only at WOT...

I guess its a sensor issue because there is more "pressure" at WOT than there is at Part Throttle, so...... sensor is having issues... I will try a different sensor tonight...