PDA

View Full Version : PLX, V2 Voltage Offset - HELP!



JezzaB
May 24th, 2007, 08:51 PM
Hey guys,
Im tearing my hair out. I have posted that awhile ago my V2 suddenly went out of calibration. I recalibrated it and thought that it was ok but its not. Ive tried outputing the voltages with

-Batteries
-Variable power supply
-The PLX (2.34 unplugged and 5v plugged in free air)

They all work bang on for the 2 reference voltages but it reads around 100mv out everywhere else. It reads lean and it just makes it richer and richer when doing an autove. Embarrasing when you tune a car and its using more fuel.

I tuned a car before the calibration dramas and its LTFT's were great.

One example is I held a load point and constant voltage (within 50mv) 1650mv in efilive and when I pull it off the V2 and put the multimeter on it, it reads around 1520mv, then put the V2 back on and its 1650mv.

I just dont know what else to do.

Jez

joecar
May 25th, 2007, 02:21 AM
Does post #2 of this thread help: showthread.php?t=4593 (http://forum.efilive.com/showthread.php?t=4593)

dc_justin
May 25th, 2007, 03:43 AM
Does post #2 of this thread help: showthread.php?t=4593 (http://forum.efilive.com/showthread.php?t=4593)

PLX's aren't programmable like that, unfortunately.

I'm experiencing the same voltage offset with mine, but it tends to drift around a bit. One time, 5v will be 5.15v in FS. The next, 5.36v. Had it hooked up to an autronic a few weeks back and it read dead on. WB power was hooked directly to the battery, so that might be part of the difference, but it otherwise appears that the WB is the culprit...

JezzaB
May 25th, 2007, 12:30 PM
Does post #2 of this thread help: showthread.php?t=4593 (http://forum.efilive.com/showthread.php?t=4593)

Yup. Ive followed that guide to the letter for every calibration using different power sources.

As said above, the PLX can only really output 2 voltages

2.34v if the WBO2 is unplugged from the controller and
5.00v free air reading

Got me stumped.

Jez

SinisterSS
May 26th, 2007, 02:11 AM
Double check your earth/ground connection. For best results I have the EFILive box ground point wired to the same point on the chassis as the PLX WBO2.

Doc
May 27th, 2007, 04:43 PM
I agree, ground offsets can be a bitch.

Blacky
May 27th, 2007, 08:58 PM
The guys that designed FlashScan have been hard at work figuring out what can be done to limit external influences on the WB voltage as read by FlashScan.

The best solution is to use the RS232 serial link into FlashScan and get an exact digital wide band AFR value. The serial interface is being developed for the next major FlashScan firmware update. The LC1 will be the first wide band supported by FlashScan's RS232 interface.

The next best option is to return your FlashScan unit to EFILive. We can replace some of the components in FlashScan to higher specification parts. Parts with smaller errors and better resilliance against ground offset issues. These parts cost more but testing has shown that with the higher quality parts we can reduce ground voltage influences that can otherwise cause up to 50mV variance.
This upgrade is not free, but EFILive has decided to subsidise the upgrade and the final cost should be less than $20 (plus shipping).
I highly recommend getting this upgrade done. We will announce a formal upgrade procedure very soon on our web site.

Another option is to make sure all ground points are as resistance free as possible. Running a low resistance ground wire from the wide band directly to the battery negative terminal helps.

Lastly, some grounds change then the ignitions is switched on. What appears to be a goo dground with the igntion off can have a small voltage when the ignition is turned on and even more so when the engine is running and the alternator is charging. So when you are checking for ground issues, check under all three cases: Ignition off, Ignition on and Engine on.

Regards
Paul

JezzaB
May 27th, 2007, 09:27 PM
The guys that designed FlashScan have been hard at work figuring out what can be done to limit external influences on the WB voltage as read by FlashScan.

The best solution is to use the RS232 serial link into FlashScan and get an exact digital wide band AFR value. The serial interface is being developed for the next major FlashScan firmware update. The LC1 will be the first wide band supported by FlashScan's RS232 interface.

The next best option is to return your FlashScan unit to EFILive. We can replace some of the components in FlashScan to higher specification parts. Parts with smaller errors and better resilliance against ground offset issues. These parts cost more but testing has shown that with the higher quality parts we can reduce ground voltage influences that can otherwise cause up to 50mV variance.
This upgrade is not free, but EFILive has decided to subsidise the upgrade and the final cost should be less than $20 (plus shipping).
I highly recommend getting this upgrade done. We will announce a formal upgrade procedure very soon on our web site.

Another option is to make sure all ground points are as resistance free as possible. Running a low resistance ground wire from the wide band directly to the battery negative terminal helps.

Lastly, some grounds change then the ignitions is switched on. What appears to be a goo dground with the igntion off can have a small voltage when the ignition is turned on and even more so when the engine is running and the alternator is charging. So when you are checking for ground issues, check under all three cases: Ignition off, Ignition on and Engine on.

Regards
Paul

This is great information and I will be looking into this upgrade when its available.

Jez

69camaro5speed
May 28th, 2007, 02:43 PM
sounds like my lm1.my own car i used a lc1 worked great.when in closed lop would read 14.6 to 14.7 and the comanded allways stayed very close to the lc1. on my lc1 i used the extra ground the the chassis. i thought about running one on my lm1 but it didnt say to with it.i could have my lm1 reading dead on 14.7 open loop doing the auto tune just enable closed loop and it would change.open loop again back to 14.7 same day just mins apart tryed it on 4 differant auto tune cars all did the same with my lm1.the lc1 with extra ground didnt do it.so guessing running a extra ground from the lm1 to car chassis should be ok to try ? dont remember reading it needed to be done.but did wounder my not

Bruce Melton
May 28th, 2007, 09:29 PM
Seems that V1 is accurate and V2 less so?
If I logged same car with both units the V2 seems skewed. Did a exhaustive calibration using power supply which helped, but V2 still seems to be variable compared to PLX display.

ringram
June 11th, 2007, 03:19 AM
Great, Id love to hook up direct to the LC-1..
Innvoate updated their own units recently because of similar issues with offsets and voltages blowing the unit etc.

mr.prick
September 1st, 2007, 07:00 AM
The guys that designed FlashScan have been hard at work figuring out what can be done to limit external influences on the WB voltage as read by FlashScan.

The best solution is to use the RS232 serial link into FlashScan and get an exact digital wide band AFR value. The serial interface is being developed for the next major FlashScan firmware update. The LC1 will be the first wide band supported by FlashScan's RS232 interface.

The next best option is to return your FlashScan unit to EFILive. We can replace some of the components in FlashScan to higher specification parts. Parts with smaller errors and better resilliance against ground offset issues. These parts cost more but testing has shown that with the higher quality parts we can reduce ground voltage influences that can otherwise cause up to 50mV variance.
This upgrade is not free, but EFILive has decided to subsidise the upgrade and the final cost should be less than $20 (plus shipping).
I highly recommend getting this upgrade done. We will announce a formal upgrade procedure very soon on our web site.

Another option is to make sure all ground points are as resistance free as possible. Running a low resistance ground wire from the wide band directly to the battery negative terminal helps.

Lastly, some grounds change then the ignitions is switched on. What appears to be a goo dground with the igntion off can have a small voltage when the ignition is turned on and even more so when the engine is running and the alternator is charging. So when you are checking for ground issues, check under all three cases: Ignition off, Ignition on and Engine on.

Regards
Paul
when is this going to be available?

Blacky
September 1st, 2007, 08:36 AM
when is this going to be available?
The A/D hardware upgrade option is available now, we just have not updated the web yet. If you want the upgrade done immediately then send your unit back to EFILive for the hardware upgrade. Please contact us before sending your unit back for the upgrade.

We will set up a web page that explains how to send the unit back, what it costs (and payment options), what gets upgraded and how long it should take.

Regards
Paul

Blacky
September 1st, 2007, 08:38 AM
when is this going to be available?
The LC1 (and possibly other wide bands) serial interface option will be available as a free software update once BB logging has been released.
BB logging is scheduled to be released (possibly as a pre-release version) prior to SEMA this year. The serial interface option will be released after that, most likely before the end of the year.

Regards
Paul

mr.prick
September 2nd, 2007, 04:20 AM
The LC1 (and possibly other wide bands) serial interface option will be available as a free software update once BB logging has been released.
BB logging is scheduled to be released (possibly as a pre-release version) prior to SEMA this year. The serial interface option will be released after that, most likely before the end of the year.

Regards
Paul
oh man i can`t wait to get rid of the analog outs!
mine flutter like crazy.

Blacky
September 2nd, 2007, 08:01 AM
oh man i can`t wait to get rid of the analog outs!
mine flutter like crazy.

We have made a firmware fix for jittery A/D inputs. The FlashScan firmware now 10x oversamples the A/D inputs and averages the input data every 100ms. That conditions the incoming data and provides a more stable A/D value to the EFILive software.

I've attached the firmware with that update here. Unzip the file into \Program Files\EFILive\V7, then run the firmware programmer to update FlashScan's firmware with the new V64 file.

Regards
Paul

Bruce Melton
September 2nd, 2007, 11:13 AM
That is happy news!
Thanks Blacky

Chalky
September 7th, 2007, 08:23 AM
Can't get the zip file to download. Can someone send me a copy?

dchalk@wi.rr.com

Chalky
September 7th, 2007, 02:06 PM
Thanks Bruce!

Chalky
September 21st, 2007, 08:10 AM
We have made a firmware fix for jittery A/D inputs. The FlashScan firmware now 10x oversamples the A/D inputs and averages the input data every 100ms. That conditions the incoming data and provides a more stable A/D value to the EFILive software.

I've attached the firmware with that update here. Unzip the file into \Program Files\EFILive\V7, then run the firmware programmer to update FlashScan's firmware with the new V64 file.

Regards
Paul

What's wrong with the link? Not much value if you can't download

joecar
September 21st, 2007, 09:07 AM
Link appears to be working...

Chalky
September 21st, 2007, 09:13 AM
Link appears to be working...

Does now!

Chuck CoW
October 2nd, 2007, 06:47 PM
We have made a firmware fix for jittery A/D inputs. The FlashScan firmware now 10x oversamples the A/D inputs and averages the input data every 100ms. That conditions the incoming data and provides a more stable A/D value to the EFILive software.

I've attached the firmware with that update here. Unzip the file into \Program Files\EFILive\V7, then run the firmware programmer to update FlashScan's firmware with the new V64 file.

Regards
Paul


Hey Paul... I'm suffering with this.... Ross just e-mailed me and wants to know why I'm e-mailing with him and what I'm going up at 3:30am us tiime....

ANSWER: I'm dying! Please get us serial AFR into V2 soon!:master:

Is this firmware current or do I already have it from another update?

Thanks
Chuck

Gelf VXR
November 6th, 2007, 04:30 AM
Does post #2 of this thread help: showthread.php?t=4593 (http://forum.efilive.com/showthread.php?t=4593)

I've checked the offsets on both my wide bands, both read around 0.03 volts lower than the programmed 1V and 4V output from the LC-1's in EfiLive.

Would the process you describe remove the 0.03V negative offset?

joecar
November 6th, 2007, 07:48 AM
I've checked the offsets on both my wide bands, both read around 0.03 volts lower than the programmed 1V and 4V output from the LC-1's in EfiLive.

Would the process you describe remove the 0.03V negative offset?IIRC, it should get it to within 0.02V which compared to 5.00V full scale is 0.4% error.

Ninety8C5
November 6th, 2007, 12:33 PM
If it is consistently -0.03 off, would adding 0.03 to the calculation in sae_generic.txt make it read correctly?

Right now it is:
# PLX WBO2
# ========================
*CLC-00-920
V 0.0 5.0 .1 "{EXT.AD1}"
AFR 10.0 20.0 .1 "{EXT.AD1}*2+10"

Why not make it:
# PLX WBO2
# ========================
*CLC-00-920
V 0.0 5.0 .1 "{EXT.AD1}"
AFR 10.0 20.0 .1 "({EXT.AD1}+0.03)*2+10"

dc_justin
November 6th, 2007, 12:36 PM
A better bet would be to add .03 to a calculated PID in calc_pids.txt. This file will not be overwritten with future software updates whereas sae_generic.txt will.

Ninety8C5
November 6th, 2007, 12:52 PM
Sounds like a better idea.

joecar
November 6th, 2007, 01:28 PM
I like that idea also.

Gelf VXR
November 7th, 2007, 11:14 PM
Great, I'll make the changes in the calculated PID, thanks

Bruce Melton
November 7th, 2007, 11:29 PM
Standing by, please let us know how that works.*

* M300 and SM user.

TNT76
December 8th, 2007, 11:50 AM
I had been using an innovate up until a few weeks ago until it shat itself. So I bought a PLX to trial it.
First up I noticed it was a fair bit off in the readings compared to an Autronic that we have on the dyno, so looked into why.
Checked the voltage with the O2 sensor unplugged and it was 2.3 or very close, which was shown through the V2 A/D input in EFILIVE, thought that was pretty close. So plugged in the O2 sensor again and it started at 2volts or so and climbed upto the free air reading but went upto 5.5volts by the efilive. Thought that was a bit far out since it should of been 5v max.
So then I got the multimeter out and measured actual volts. With the O2 sensor unplugged the reading showed 2.32v on the multimeter and through efilive, about spot on. O2 sensor back in showed 4.95 on meter and 5.5 on efilive. Gave the car a run for a few minutes and shut it off and the volts showed 2.35v on the meter and 2.8v through efilive.
So what I found was a voltage reading difference only when the O2 sensor was plugged in.
Recalibrated efi to show same volts as multimeter and all is sweet.

Chalky
December 8th, 2007, 12:18 PM
What was involved in calibrating PLX in EFILive? Are you using SM-AFR?

I have had some discussions with the folks from PLX. They claim that EFILive has a max of 40 Milli amps while PLX only supports 20 milli amps. PLX did an upgrade on my SM-AFR to support 40 milli amps. I didn't notice any difference in AFR readings. PLX thought the difference in amperage might have caused a misreading in AFR.:confused:

TNT76
December 8th, 2007, 12:40 PM
Its in the V2 control panel (ctrl + f11). once you get a stable voltage on the plx, punch in what your multimeter says into the first voltage (usually around 2volts) then a higher volt around 5v then hit calibrate. it gives instructions on the screen.
I'm not sure if this will sort the whole range or just between the two ref voltages. but it sorted the range I needed as well as down to at least 12:1. As long as you get a stable reading.
Its an SM-afr.

oztracktuning
December 8th, 2007, 08:39 PM
The issue is caused by earth offset voltage resulting from the wiring resistance of the PLX device. This is why it varies in its error. The best way to make it stable is to use low resistance wiring more direct the battery.

TNT76
December 9th, 2007, 01:27 PM
Yeah thought of that as I was unrolling about 3 meters of the supplied power lead of about 1mm wire. Was much easier with a few key strokes to offset the offsets.
Will look into some heavier power lead.

Gelf VXR
February 16th, 2008, 02:48 PM
A better bet would be to add .03 to a calculated PID in calc_pids.txt. This file will not be overwritten with future software updates whereas sae_generic.txt will.

I found the location of the calc_pid text, its practically empty, do I cut and paste the file from the sae_generic file?

Thanks