How to accurately get rid of WB ground offset?
10-19-2010, 09:29 PM
#1
11 Second Club
Thread Starter
iTrader: (14)
Join Date: Nov 2004
Location: Michigan (Macomb or Lansing)
Posts: 1,653
Likes: 0
Received 0 Likes
on
0 Posts
How to accurately get rid of WB ground offset? I have a PLX wideband that seems to work well. I have it wired through my EGR (also used the included filter capacitor) circuit and lately i think i have a ground offset.
I am doing an OLSD tune on a 404 h/c/i and I did my VE table off of the wideband, however it seems to run about 13.9 at idle and part throttle according to wideband (When 14.7 is commanded), obviously too rich.
How can I accurately figure out what my offset is and then correct for it? MY narrowbands work and seem to switch correctly when i command 18 a/f.
At this 18 a/f commanded, my narrowbands switch at 450 mv and my wideband reads 15.5ish.
Should i just take my wideband a/f formula and subtract .8 from it?
Is there any other way to accurately do this?
I am doing an OLSD tune on a 404 h/c/i and I did my VE table off of the wideband, however it seems to run about 13.9 at idle and part throttle according to wideband (When 14.7 is commanded), obviously too rich.
How can I accurately figure out what my offset is and then correct for it? MY narrowbands work and seem to switch correctly when i command 18 a/f.
At this 18 a/f commanded, my narrowbands switch at 450 mv and my wideband reads 15.5ish.
Should i just take my wideband a/f formula and subtract .8 from it?
Is there any other way to accurately do this?
10-19-2010, 09:58 PM
#2
get the power and ground connections right and you wont have ground offsets...
take all Grounds to the Battery-including the ground for the analog output and the ground from the EGR cable.
get power from the battery
use a relay to turn on the wideband using any ignition on source.
proper wiring solves electrical problems.
take all Grounds to the Battery-including the ground for the analog output and the ground from the EGR cable.
get power from the battery
use a relay to turn on the wideband using any ignition on source.
proper wiring solves electrical problems.
10-19-2010, 11:56 PM
#3
11 Second Club
Thread Starter
iTrader: (14)
Join Date: Nov 2004
Location: Michigan (Macomb or Lansing)
Posts: 1,653
Likes: 0
Received 0 Likes
on
0 Posts
While I completely agree with you completely on this, and I plan on doing it, I don't have access to my soldering and electrical tools for a couple weeks.
Im pretty sure that I did a good job on the install. Thinking about it, i think i do have a dedicated accessory relay for just my wideband and nitrous stuff. I know for a fact i grounded the egr and wideband to the same ground which is from the body. i can try to get it directly to the battery sometime.
In the meantime, is there any way to do it with software?
BTW. I am a bit confused now... I tuned my car OLSD, stfts and ltfts turned off and i did % error corrections on my VE table from my wideband: Resulting in the 13.9 readings on my gauge at idle and part throttle.
I enabled closed loop to see what stfts would do. They wanted me to trim my VE table down by up to 20% in some areas for low throttle. I did this. Now for some reason my STFTs are +/- 3% and my wideband is settled with +/- 3%.
What could be causing this? Why would tuning off of the wideband make my car way rich and then when i used my stfts, both my wideband and stfts agree?
Im pretty sure that I did a good job on the install. Thinking about it, i think i do have a dedicated accessory relay for just my wideband and nitrous stuff. I know for a fact i grounded the egr and wideband to the same ground which is from the body. i can try to get it directly to the battery sometime.
In the meantime, is there any way to do it with software?
BTW. I am a bit confused now... I tuned my car OLSD, stfts and ltfts turned off and i did % error corrections on my VE table from my wideband: Resulting in the 13.9 readings on my gauge at idle and part throttle.
I enabled closed loop to see what stfts would do. They wanted me to trim my VE table down by up to 20% in some areas for low throttle. I did this. Now for some reason my STFTs are +/- 3% and my wideband is settled with +/- 3%.
What could be causing this? Why would tuning off of the wideband make my car way rich and then when i used my stfts, both my wideband and stfts agree?
Last edited by pHEnomIC; 10-20-2010 at 12:01 AM.
10-20-2010, 10:15 AM
#4
Moderator
iTrader: (11)
Join Date: Mar 2002
Location: East Central Florida
Posts: 12,605
Likes: 0
Received 6 Likes
on
6 Posts
In software you could set your wideband to have less
range (say, 11:1 - 16:1 for 0V-5V out) so that any
ground skew is smaller relative to real output. Some
people propose figuring your ground offset mV but I
think this is too variable (follows heater current, which
follows environment) to be trustworthy.
My setup is pretty crappy for ground offset, a LM-1
with 10' of cig lighter power cord and two connectors,
plus the natural cig lighter to OBD-II port sheet metal
ground offset.
But I have made a little widget that takes all that out.
Just an AD620 instrumentation amplifier and two 9V
batteries. Float the amplifier, and it takes the input
voltage at the LM-1 output jack (which is tight), and
outputs the difference referred against the EIO remote
ground. Ten bucks, hour of soldering, all done. Except
I keep forgetting to turn it off so I have to remember
to bring fresh batteries. "Someday" I'll do another one
with isolated 12V power and fix that.
range (say, 11:1 - 16:1 for 0V-5V out) so that any
ground skew is smaller relative to real output. Some
people propose figuring your ground offset mV but I
think this is too variable (follows heater current, which
follows environment) to be trustworthy.
My setup is pretty crappy for ground offset, a LM-1
with 10' of cig lighter power cord and two connectors,
plus the natural cig lighter to OBD-II port sheet metal
ground offset.
But I have made a little widget that takes all that out.
Just an AD620 instrumentation amplifier and two 9V
batteries. Float the amplifier, and it takes the input
voltage at the LM-1 output jack (which is tight), and
outputs the difference referred against the EIO remote
ground. Ten bucks, hour of soldering, all done. Except
I keep forgetting to turn it off so I have to remember
to bring fresh batteries. "Someday" I'll do another one
with isolated 12V power and fix that.
10-27-2010, 02:37 AM
#5
11 Second Club
Thread Starter
iTrader: (14)
Join Date: Nov 2004
Location: Michigan (Macomb or Lansing)
Posts: 1,653
Likes: 0
Received 0 Likes
on
0 Posts
I have hptuners pro now, if i wire straight to the hptuners interface, will that effectively get rid of any ground offsets?
10-27-2010, 04:09 PM
#6
Moderator
iTrader: (11)
Join Date: Mar 2002
Location: East Central Florida
Posts: 12,605
Likes: 0
Received 6 Likes
on
6 Posts
That's more about the wideband current loop. How much
pedestal the heater current puts under the output ground
reference and signal, relative to OBDII port. You can fix
some of that by taking wideband power return ground
off right by the connector but there can still be ground-rise
that depends on its wiring and connectors, etc.
"Get rid of", no. "Reduce", yes. "Good enough"? Maybe.
Another thing to look at it to expand the range so you
are skewed less by the pedestal error. Like don't make
0V=10:1 and 4V=18:1, when you don't care about either
of those ends - it makes 100mV into 0.1 AFR.
Make it 0V 12:1 and 4V 16:1, and you've got the sweet
spot covered pretty much - and drove down error to 0.05
AFR per 100mV. Assuming your wideband lets you vary the
transfer function.
pedestal the heater current puts under the output ground
reference and signal, relative to OBDII port. You can fix
some of that by taking wideband power return ground
off right by the connector but there can still be ground-rise
that depends on its wiring and connectors, etc.
"Get rid of", no. "Reduce", yes. "Good enough"? Maybe.
Another thing to look at it to expand the range so you
are skewed less by the pedestal error. Like don't make
0V=10:1 and 4V=18:1, when you don't care about either
of those ends - it makes 100mV into 0.1 AFR.
Make it 0V 12:1 and 4V 16:1, and you've got the sweet
spot covered pretty much - and drove down error to 0.05
AFR per 100mV. Assuming your wideband lets you vary the
transfer function.
