Changed Widebands, Values Differ Greatly
07-26-13, 01:53 PM
#1
Eye In The Sky
Thread Starter
iTrader: (2)
Join Date: Feb 2001
Location: In A Disfunctional World
Posts: 7,892
Likes: 0
Received 114 Likes
on
66 Posts
Changed Widebands, Values Differ Greatly
I have been tuning with PLX M-300 WBs since 1999 without any problems or blown engines. They use the Bosch LSU 4.2 sensor.
Just got a NGK Powerdex AFX WB with a NTK sensor.
After calibration to air, I compared it to the PLX.
I used a gas mixture that gave about 11AFR to the AFX readout.
The AFX was reading about .8 AFR richer than the PLX.
Travis notice about the same thing when he went from a PLX to the AFX.
Has anyone notice something similar?
Just got a NGK Powerdex AFX WB with a NTK sensor.
After calibration to air, I compared it to the PLX.
I used a gas mixture that gave about 11AFR to the AFX readout.
The AFX was reading about .8 AFR richer than the PLX.
Travis notice about the same thing when he went from a PLX to the AFX.
Has anyone notice something similar?
07-26-13, 02:35 PM
#2
I noticed the same thing when we tried out an AEM that use the Bosch LSU 4.2 sensor. I can't remember for sure the AEM reading richer or leaner but it was surely different than my NTK.
The NTK L1H1/L2H2 uses a twin-pump (plus a catalyst system) to measure oxygen content. In short, the NTK sensor contains a narrow band (nernst cell) and wideband sensor (diffusion cell) in one unit. They're suppose to be a much more robust unit but you do have to deal with a "narrow window" of 9-16afr.
The NTK L1H1/L2H2 uses a twin-pump (plus a catalyst system) to measure oxygen content. In short, the NTK sensor contains a narrow band (nernst cell) and wideband sensor (diffusion cell) in one unit. They're suppose to be a much more robust unit but you do have to deal with a "narrow window" of 9-16afr.
07-27-13, 10:54 PM
#6
Eye In The Sky
Thread Starter
iTrader: (2)
Join Date: Feb 2001
Location: In A Disfunctional World
Posts: 7,892
Likes: 0
Received 114 Likes
on
66 Posts
The first thing I do with a new unit is connect it to the DL and to its software in my laptop. Using a jar with the sensor in it, and sealed with a mixture of air and propane to display around 11 AFR; I modified the AUX setup so that the DL logs what the unit displays.
Both units were setup that way. When both sensors were exposed to the same gas mixture at the same time, the AFX reads .7 - .8 AFR richer.
My old PLX M-300 and newer PLX M-300TE logged richer than their displays.
The Powerdex AFX logged leaner than its display.
Thus both have different compensations made to their setups.
07-27-13, 11:49 PM
#7
What are you trying to say, that an NTK sensor and driver are superior to an LSU sensor and driver? Most dynos that I know of use the exact same sensors (one ore the other) with a somewhat different (everyone has to reinvent the wheel, some chintz on having a sensor ground versus element etc) driving element as the in car counterparts.
There are issues with a lot of vehicle installs such as I've mentioned (piggybacking sensor and heater grounds on one another) but to say that it is never wise to rely on a vehicle installed wideband versus a dynos 12v source powered wideband I think doesn't give the whole picture on why one should be skeptical of one AFR to another.
Trending Topics
07-28-13, 07:24 AM
#8
Racing Rotary Since 1983
iTrader: (6)
i have run some sort of Bosch WB setup since 2004 and none of them were in sync w the NGK system on the Beyond Redline dyno. i finally switched to the NGK AFX and my car lines up to a tenth w the dyno.
Luke also mentioned that the NGK is significantly better than the Bosch when reading rich area AFRs such as in the low tens and high elevens.
of course that doesn't necessarily mean that the NGK is reading absolutely correct, just that the two are in sync.
given the newer ECUs abilities to auto-tune, having the right AFR being generated is especially valuable.
hc
Luke also mentioned that the NGK is significantly better than the Bosch when reading rich area AFRs such as in the low tens and high elevens.
of course that doesn't necessarily mean that the NGK is reading absolutely correct, just that the two are in sync.
given the newer ECUs abilities to auto-tune, having the right AFR being generated is especially valuable.
hc
07-28-13, 07:53 AM
#9
It's just common practice to tune by AFR rule-of-thumb, because really that's a very practical way of doing it without a gazillion dollars in other instrumentation. So when we say something like "I have to see my AFR richer than a nominal 12.0:1 or it'll blow up, but I have 3 sensors all saying something different, what do I believe?"
For what it's worth, I've compared Denso (on-board sensors found on newer engines), Bosch (LSU 4.2 and 4.9), and NGK sensors to actual emission benches in steady-state engine dyno conditions. They're all pretty close for the most part. These were using lab-grade equipment--specifically, ETAS and ECM controllers. Now when you get into that rich of a range on a boosted engine, the heater control as well as temperature and pressure sensitivity does vary.
I've said before that sometimes I feel like there's a little too much hand-wringing over AFR's, at least when we get to the point where we're debating widebands like this. Timing blows up engines much more easily than AFR. I've seen that again and again. So I think about widebands the same way I think about an engine with multiple boost gauges/pressure sensors/MAP sensors: just pick ONE and stick with it. Any of them will work when there's a good tuner, provided the sensor isn't damaged. Otherwise you'll drive yourself nuts making these comparisons.
For what it's worth, I've compared Denso (on-board sensors found on newer engines), Bosch (LSU 4.2 and 4.9), and NGK sensors to actual emission benches in steady-state engine dyno conditions. They're all pretty close for the most part. These were using lab-grade equipment--specifically, ETAS and ECM controllers. Now when you get into that rich of a range on a boosted engine, the heater control as well as temperature and pressure sensitivity does vary.
I've said before that sometimes I feel like there's a little too much hand-wringing over AFR's, at least when we get to the point where we're debating widebands like this. Timing blows up engines much more easily than AFR. I've seen that again and again. So I think about widebands the same way I think about an engine with multiple boost gauges/pressure sensors/MAP sensors: just pick ONE and stick with it. Any of them will work when there's a good tuner, provided the sensor isn't damaged. Otherwise you'll drive yourself nuts making these comparisons.
07-28-13, 10:16 AM
#10
Eye In The Sky
Thread Starter
iTrader: (2)
Join Date: Feb 2001
Location: In A Disfunctional World
Posts: 7,892
Likes: 0
Received 114 Likes
on
66 Posts
The reason I am concerned is that I have been tuning my car with the PLX for years and have not had any knock problems.
Is this because it reads leaner and I have been running richer safer than my targets?
Now with the new WB which says to run leaner by almost 1 AFR, will I be in trouble?
Example, I have been running 11.2 AFR at 16 psi without WI for years. Or with the PLX has that been really 10.4. With the AFX, will it now be 12?
It is scary. You expect equipment to be accurate and not vary that much. You loose faith in equipment.
Thus I will initially only retune my lower 2 boost rows and see what happens with knock.
My EBC has 4 settings and setting 4 is my highest with water/alky.
Is this because it reads leaner and I have been running richer safer than my targets?
Now with the new WB which says to run leaner by almost 1 AFR, will I be in trouble?
Example, I have been running 11.2 AFR at 16 psi without WI for years. Or with the PLX has that been really 10.4. With the AFX, will it now be 12?
It is scary. You expect equipment to be accurate and not vary that much. You loose faith in equipment.
Thus I will initially only retune my lower 2 boost rows and see what happens with knock.
My EBC has 4 settings and setting 4 is my highest with water/alky.
07-28-13, 01:29 PM
#12
Senior Member
A couple years ago I went to the Mega-Meet for the megasquirt engine management.
Bruce Bowling, one of the creators of the MS system, put on a presentation about the inaccuracies of the currently available Wide Band units.
at the time he was working on a system of his own that was to be much better, but in the end the discussion boiled down to how you should not trust only the in-car wide band.
If you are tuning for a high specific output, as all of us with rotaries are, you need to use as many tools and all the experience you can find.
Bruce Bowling, one of the creators of the MS system, put on a presentation about the inaccuracies of the currently available Wide Band units.
at the time he was working on a system of his own that was to be much better, but in the end the discussion boiled down to how you should not trust only the in-car wide band.
If you are tuning for a high specific output, as all of us with rotaries are, you need to use as many tools and all the experience you can find.
07-28-13, 08:00 PM
#13
They're all sensitive to exhaust temperature and pressure.
Widebands have a time delay in representing what's happening in the actual combustion chamber. If your wideband reads some AFR, you have to consider the amount of time it took for the exhaust to travel to the sensor. It's one of the issues with using gas analyzers (emissions bench) during transients.
Widebands are sensitive to how much scavenging you are running (basically, port timing/valve timing). At a given AFR, the brake specific air consumption can be different depending on the amount of overlap. So the AFR reading can be deceptive--air can be blown right out to the exhaust if intake pressure is higher than exhaust pressure, and the engine is in an overlap condition.
I've seen it on piston engines with cam phasers on intake and exhaust. Dial in a 100 degrees of overlap from advancing the intake cam and retarding the exhaust cam and it will make your wideband read deceptively lean. Considering the amount of overlap rotaries have, especially ported ones, I wouldn't be surprised if you get the same effect at high boost.
Wideband have issues with sensor degradation over time. Did you know that widebands 'drift' , and normally the rear o2 sensor on a production car will detect and compensate? The front wideband O2 sensor can read too lean, and if the ECU doesn't compensate based on rear o2 sensor, you will have NOx emissions issues. I'd say that matters less for cruising range though.
I totally understand. So the options I see for you are:
1) keep using PLX (not sure if yours broke, but you could get another one), maybe replace the sensor and see if the readings change
2) Use NGK and additionally tune by EGT --doesn't really tell you knock, but you'd expect EGT to be same under controlled conditions if tune didn't change, regardless of wideband reading.
3) Use NGK and additionally tune by knock sensor--you're already doing this
4) Use NGK and tune to knock limit by combustion pressure trace--this isn't normally feasible without steady state operation on an engine dyno
If you've got deep pockets, here are the lab grade widebands I've used...
ECM products like the FA1000, which uses an NGK sensor
ECM - F/A1000
ETAS products like the ES635 which uses Bosch LSU 4.9 and can compensate based on exhaust backpressure
ETAS - ES63x
Widebands have a time delay in representing what's happening in the actual combustion chamber. If your wideband reads some AFR, you have to consider the amount of time it took for the exhaust to travel to the sensor. It's one of the issues with using gas analyzers (emissions bench) during transients.
Widebands are sensitive to how much scavenging you are running (basically, port timing/valve timing). At a given AFR, the brake specific air consumption can be different depending on the amount of overlap. So the AFR reading can be deceptive--air can be blown right out to the exhaust if intake pressure is higher than exhaust pressure, and the engine is in an overlap condition.
I've seen it on piston engines with cam phasers on intake and exhaust. Dial in a 100 degrees of overlap from advancing the intake cam and retarding the exhaust cam and it will make your wideband read deceptively lean. Considering the amount of overlap rotaries have, especially ported ones, I wouldn't be surprised if you get the same effect at high boost.
Wideband have issues with sensor degradation over time. Did you know that widebands 'drift' , and normally the rear o2 sensor on a production car will detect and compensate? The front wideband O2 sensor can read too lean, and if the ECU doesn't compensate based on rear o2 sensor, you will have NOx emissions issues. I'd say that matters less for cruising range though.
The reason I am concerned is that I have been tuning my car with the PLX for years and have not had any knock problems.
Is this because it reads leaner and I have been running richer safer than my targets?
Now with the new WB which says to run leaner by almost 1 AFR, will I be in trouble?
Example, I have been running 11.2 AFR at 16 psi without WI for years. Or with the PLX has that been really 10.4. With the AFX, will it now be 12?
It is scary. You expect equipment to be accurate and not vary that much. You loose faith in equipment.
Is this because it reads leaner and I have been running richer safer than my targets?
Now with the new WB which says to run leaner by almost 1 AFR, will I be in trouble?
Example, I have been running 11.2 AFR at 16 psi without WI for years. Or with the PLX has that been really 10.4. With the AFX, will it now be 12?
It is scary. You expect equipment to be accurate and not vary that much. You loose faith in equipment.
1) keep using PLX (not sure if yours broke, but you could get another one), maybe replace the sensor and see if the readings change
2) Use NGK and additionally tune by EGT --doesn't really tell you knock, but you'd expect EGT to be same under controlled conditions if tune didn't change, regardless of wideband reading.
3) Use NGK and additionally tune by knock sensor--you're already doing this
4) Use NGK and tune to knock limit by combustion pressure trace--this isn't normally feasible without steady state operation on an engine dyno
If you've got deep pockets, here are the lab grade widebands I've used...
ECM products like the FA1000, which uses an NGK sensor
ECM - F/A1000
ETAS products like the ES635 which uses Bosch LSU 4.9 and can compensate based on exhaust backpressure
ETAS - ES63x
07-28-13, 09:29 PM
#14
Junior Member
Join Date: Jul 2013
Location: louisville
Posts: 15
Likes: 0
Received 0 Likes
on
0 Posts
On any turbo car that I've owned with an AEM wideband, this has always been the case.
Personally I grew use to it, but some of my buddies prefer to go with an LC1
07-29-13, 11:41 AM
#16
Weird Cat Man
I have seen similar things in my limited experience.
See: https://www.rx7club.com/single-turbo...estion-154842/
That comment was after I had gone to KDR (Dave).
At a later time, I actually had my car with my TechEdge wideband side-by-side with Peter Farrell's Bailey brand. The Bailey read higher even on the same run. I'm not saying one is right, and one is wrong - Just saying they can read a bit different so you can't always directly compare readings from one to another with 100% precision.
See: https://www.rx7club.com/single-turbo...estion-154842/
That comment was after I had gone to KDR (Dave).
At a later time, I actually had my car with my TechEdge wideband side-by-side with Peter Farrell's Bailey brand. The Bailey read higher even on the same run. I'm not saying one is right, and one is wrong - Just saying they can read a bit different so you can't always directly compare readings from one to another with 100% precision.
08-03-13, 10:37 AM
#18
Eye In The Sky
Thread Starter
iTrader: (2)
Join Date: Feb 2001
Location: In A Disfunctional World
Posts: 7,892
Likes: 0
Received 114 Likes
on
66 Posts
In general, tail pipe sniffers can read leaner depending on conditions:
(1) low rpms tend to be leaner, usually OK at high rpms full power
(2) ricer large tailpipes tend to be very leaner, even at high rpms and full power
(3) strong tail winds make it leaner, any tail or side wind on a dyno make it leaner exceept at full power
(4) If a dyno shop does not change their sensor frequently or calibrate it frequently, it can not
be trusted.
I have used a tail sniffer with good results for full power tuning in 3rd and 4th gears when the tail pipe is smaller like with the RB models. It did not work with a huge single tip ricer pipe.
(1) low rpms tend to be leaner, usually OK at high rpms full power
(2) ricer large tailpipes tend to be very leaner, even at high rpms and full power
(3) strong tail winds make it leaner, any tail or side wind on a dyno make it leaner exceept at full power
(4) If a dyno shop does not change their sensor frequently or calibrate it frequently, it can not
be trusted.
I have used a tail sniffer with good results for full power tuning in 3rd and 4th gears when the tail pipe is smaller like with the RB models. It did not work with a huge single tip ricer pipe.
08-03-13, 10:43 AM
#19
Moderator
iTrader: (3)
Join Date: Mar 2001
Location: https://www2.mazda.com/en/100th/
Posts: 30,805
Received 2,578 Likes
on
1,831 Posts
Thread
Thread Starter
Forum
Replies
Last Post
trickster
2nd Generation Specific (1986-1992)
25
07-01-23 04:40 PM
fastrx7man
3rd Generation Specific (1993-2002)
33
09-02-15 09:42 PM