ADVrider

Go Back   ADVrider > Bikes > GSpot > Parallel Universe
User Name
Password
Register Inmates Photos Site Rules Search Today's Posts Mark Forums Read

Reply
 
Thread Tools Search this Thread Display Modes
Old 05-25-2013, 07:57 PM   #436
MTrider16
Ridin' in MT
 
MTrider16's Avatar
 
Joined: Oct 2006
Location: Eastern Montana
Oddometer: 1,611
Thanks for sharing everything Ebrabaek. I got a call at work that surprised me, a OEM wants to do some experimenting on their AFR for some of their industrial engines we own. Looks like it will be an interesting experiment.

David
__________________
'13 VFR1200D, '13 XVS950, '09 F800GS, 07 CRF250X
Riding roads in Montana - Big Sky Country
www.mtrider16.smugmug.com
Mountains, Moose, and Miles: a Montanan's Alcan Highway Story
Continental Divide and More: the "No Dust" Tour of WY and MT
MTrider16 is offline   Reply With Quote
Old 05-25-2013, 09:11 PM   #437
ebrabaek OP
Beastly Adventurer
 
Joined: Jan 2010
Location: Grand Valley, Colorado
Oddometer: 4,672
Quote:
Originally Posted by MTrider16 View Post
Thanks for sharing everything Ebrabaek. I got a call at work that surprised me, a OEM wants to do some experimenting on their AFR for some of their industrial engines we own. Looks like it will be an interesting experiment.

David
Absolutely David. It has been an interesting journey. May yours be as fascinating.....
__________________


Erling
ebrabaek is offline   Reply With Quote
Old 05-26-2013, 10:01 AM   #438
roger 04 rt
Beastly Adventurer
 
roger 04 rt's Avatar
 
Joined: Nov 2011
Location: Massachusetts
Oddometer: 2,150
Hi EB, You have done a great job of documenting, testing and examining alternative theories for some of the results you've seen. We are all far richer in our knowledge of the F800GS for the work you've done. You have a configuration that works well and I hope you don't catch my disease of perpetual examination and testing.

At this point, I'm still a bit perplexed, as you are, for why the PCV is adding so much fuel. What's happening, is: in order to get the oxygen content in the exhaust down to a level consistent with the 13.2:1 AFR target, there is a lot of the map with 20-30% fuel additions in the mid-range. Here are the explanations I see:

1. Extremely lean fueling by the BMSK.
2. Large amounts of additional air in the cylinder charge from the header/exhaust mods.
3. The SAS valve is introducing oxygen.
4. Air Reversion at low throttle angles and RPMs below 5000
5. Air leakage somewhere else.

For comparison, below is some data taken from my stock (exhaust/intake) R1150RT which shows how the Motronic MA 2.4 manages a WOT run. Looking at the first chart, the Motronic injects a lot of fuel for a second or two, then relaxs it a bit, then builds toward the end of this run (at which time I was doing about 100 MPH), during the run, the Motronic advances the spark slowly. Looking at the second LC-1 AFR log, the AFR builds slowly from closed loop of 13.8:1 to about 12.1:1 mid-way and then holds there. A few seconds more of cruising and the AFR is back to 13.8:1 and then I drop the throttle and the AFR goes to 22:1, it's max value. The BMW systems look very well behaved with no tendency to lean-out the mixture during acceleration, rather they add a lot of fuel (Point #2 on the first chart). I have looked at data from Terry's R1200GSA and his AFRs during acceleration look just the same.

RB



roger 04 rt is offline   Reply With Quote
Old 05-26-2013, 11:18 AM   #439
itsatdm
Beastly Adventurer
 
itsatdm's Avatar
 
Joined: Jun 2004
Location: Nor Ca.
Oddometer: 4,733
Do you have any dyno runs with AFR of the F800? I remember seeing one on the internet, and the lowest AFR was 14.4. I can't find it again.

That might be important, because motorcyles were included in Euro 3 emission standards in 2006 and I think Euro4 went into effect in 2012.

I suspect we all get the same maps.
__________________
BMW Motorrad USA customer service: "We make superior motorcycles and continue to improve them."
itsatdm is offline   Reply With Quote
Old 05-26-2013, 12:20 PM   #440
roger 04 rt
Beastly Adventurer
 
roger 04 rt's Avatar
 
Joined: Nov 2011
Location: Massachusetts
Oddometer: 2,150
Quote:
Originally Posted by itsatdm View Post
Do you have any dyno runs with AFR of the F800? I remember seeing one on the internet, and the lowest AFR was 14.4. I can't find it again.

That might be important, because motorcyles were included in Euro 3 emission standards in 2006 and I think Euro4 went into effect in 2012.

I suspect we all get the same maps.
I don't have one for the F800 but if you look at the chart below for an R1150RT I know for a fact it isn't correct. Dyno operators seem to have a hard time getting their probes into a place where they can make a reliable AFR reading.

For my money, the only way to get a good AFR reading is with a wideband located ahead of the cat in the exhaust pipe. I haven't seen anything else which looks very realistic. See the 1150 chart below. I know from my own measurements that a stock 1150 gets to about 13:1. But the chart below starts very high (not realistic at all) and doesn't make it to 13:1. I'll bet a dollar that this is a tailpipe AFR probe. Go back two posts and compare it to my measurement made pre-cat. The chart below should start at 14.7:1 and move smoothly down to about 13:1 on a stock bike.


roger 04 rt screwed with this post 05-26-2013 at 12:28 PM
roger 04 rt is offline   Reply With Quote
Old 05-26-2013, 12:53 PM   #441
ebrabaek OP
Beastly Adventurer
 
Joined: Jan 2010
Location: Grand Valley, Colorado
Oddometer: 4,672
Quote:
Originally Posted by roger 04 rt View Post
Hi EB, You have done a great job of documenting, testing and examining alternative theories for some of the results you've seen. We are all far richer in our knowledge of the F800GS for the work you've done. You have a configuration that works well and I hope you don't catch my disease of perpetual examination and testing.

At this point, I'm still a bit perplexed, as you are, for why the PCV is adding so much fuel. What's happening, is: in order to get the oxygen content in the exhaust down to a level consistent with the 13.2:1 AFR target, there is a lot of the map with 20-30% fuel additions in the mid-range. Here are the explanations I see:

1. Extremely lean fueling by the BMSK.
Yes....for sure, as we saw that " riding the Dyno"

2. Large amounts of additional air in the cylinder charge from the header/exhaust mods.
3. The SAS valve is introducing oxygen.
4. Air Reversion at low throttle angles and RPMs below 5000
5. Air leakage somewhere else.

For comparison, below is some data taken from my stock (exhaust/intake) R1150RT which shows how the Motronic MA 2.4 manages a WOT run. Looking at the first chart, the Motronic injects a lot of fuel for a second or two, then relaxs it a bit, then builds toward the end of this run (at which time I was doing about 100 MPH), during the run, the Motronic advances the spark slowly. Looking at the second LC-1 AFR log, the AFR builds slowly from closed loop of 13.8:1 to about 12.1:1 mid-way and then holds there. A few seconds more of cruising and the AFR is back to 13.8:1 and then I drop the throttle and the AFR goes to 22:1, it's max value. The BMW systems look very well behaved with no tendency to lean-out the mixture during acceleration, rather they add a lot of fuel (Point #2 on the first chart). I have looked at data from Terry's R1200GSA and his AFRs during acceleration look just the same.

RB



As I studied the resources available, for the F800GS ( not a whole lot) I compared it with the info from DJ, and they all really added a lot of fuel. If you look at the DJ base maps, they are pouring in fuel as well, so there must be something to that. But on the flip side of that, why with a about 30% fuel addition, why am I not seeing a detrimental fuel economy....???? My answer to that, is along my tuner's answer.... It takes a certain amount of fuel/air to move a certain weight or should I say mass along. And I can say with certainty, that with this installed, it takes a whole lot of less throttle, than it did before.
__________________


Erling
ebrabaek is offline   Reply With Quote
Old 05-26-2013, 12:55 PM   #442
ebrabaek OP
Beastly Adventurer
 
Joined: Jan 2010
Location: Grand Valley, Colorado
Oddometer: 4,672
Quote:
Originally Posted by itsatdm View Post
Do you have any dyno runs with AFR of the F800? I remember seeing one on the internet, and the lowest AFR was 14.4. I can't find it again.

That might be important, because motorcyles were included in Euro 3 emission standards in 2006 and I think Euro4 went into effect in 2012.

I suspect we all get the same maps.
Not sure if you are asking me, but the dyno graphs I posted has the AFR on them earlier in the thread, stock.....pipe....header...... full meal deal.... in steps.
__________________


Erling
ebrabaek is offline   Reply With Quote
Old 05-26-2013, 03:49 PM   #443
ebrabaek OP
Beastly Adventurer
 
Joined: Jan 2010
Location: Grand Valley, Colorado
Oddometer: 4,672
Gas mileage..... Just filled up.... 132 miles, and 3.01 gal. ( US....=3.78 Liters) 43.85 mpg. Computer said 50 mpg. This is the first time they have disagreed. One day, I will have to put this thing on the dyno, with both SAS On, and OFF...... Puzzled here.
__________________


Erling
ebrabaek is offline   Reply With Quote
Old 05-26-2013, 09:12 PM   #444
roger 04 rt
Beastly Adventurer
 
roger 04 rt's Avatar
 
Joined: Nov 2011
Location: Massachusetts
Oddometer: 2,150
As you know I'm not surprised by the actual gas mileage being less than the trip computer. That's how it should work but we need an explanation of why they were the same until this last tank.

On my chart below, the point I'm making is that curve 2 (the injection time) adds fuel immediately, even over-fueling for a couple seconds. Then at wot it builds to a very high level. There's no real mechanism for AFR to go above 14.7, or in my case 13.8. And when you see the result on the AFR curve it richens to about 12:1.

Since your BMSK hips highly likely to follow the same pattern, its hard to believe that your 18:1 off idle is real. Not that you didn't measurement, just that O2 is coming to the O2 sensor from other sources than the combustion chamber.

My results can be confirmed by someone with a GS-911 and an f800GS. Attach the GS AND 911, turn on cvs logging and go make a couple wot runs in forth gear. Then we will know.

roger 04 rt is offline   Reply With Quote
Old 05-26-2013, 09:40 PM   #445
ebrabaek OP
Beastly Adventurer
 
Joined: Jan 2010
Location: Grand Valley, Colorado
Oddometer: 4,672
Quote:
Originally Posted by roger 04 rt View Post
As you know I'm not surprised by the actual gas mileage being less than the trip computer. That's how it should work but we need an explanation of why they were the same until this last tank.

On my chart below, the point I'm making is that curve 2 (the injection time) adds fuel immediately, even over-fueling for a couple seconds. Then at wot it builds to a very high level. There's no real mechanism for AFR to go above 14.7, or in my case 13.8. And when you see the result on the AFR curve it richens to about 12:1.

Since your BMSK hips highly likely to follow the same pattern, its hard to believe that your 18:1 off idle is real. Not that you didn't measurement, just that O2 is coming to the O2 sensor from other sources than the combustion chamber.

My results can be confirmed by someone with a GS-911 and an f800GS. Attach the GS AND 911, turn on cvs logging and go make a couple wot runs in forth gear. Then we will know.

Mind you that the high AFR was seen not at WOT, but as " driving the dyno" And not seen until open pipe, and headers. I do not think stock will yield that high AFR, but again, the stock runs went by very quick.
__________________


Erling
ebrabaek is offline   Reply With Quote
Old 05-27-2013, 04:47 AM   #446
roger 04 rt
Beastly Adventurer
 
roger 04 rt's Avatar
 
Joined: Nov 2011
Location: Massachusetts
Oddometer: 2,150
Quote:
Originally Posted by ebrabaek View Post
Not sure if you are asking me, but the dyno graphs I posted has the AFR on them earlier in the thread, stock.....pipe....header...... full meal deal.... in steps.
Here is that dyno graph. On the left-hand side of the AFR chart, the lines all start at AFRs that aren't real, but an anomaly created by the tailpipe O2 sensor. You'll see that the BLUE line (stock) has recorded horsepower between 3000-4000 RPM, but the the BLUE AFR line is way above 16:1 below 5000 RPM. That is the air reversion Dyno error I'm talking about. (BTW, does the "stock" run include a catalytic converter?)

Also note that all the AFR lines fall around or above 14:1. I know from measurements my bike (see above chart) and the data from Terry's GSA BMSK that the BMSK will be fueling to at least 13:1 at WOT. For some reason the dynos tailpipe sniffer doesn't get it right there either.



Quote:
Originally Posted by ebrabaek View Post
As I studied the resources available, for the F800GS ( not a whole lot) I compared it with the info from DJ, and they all really added a lot of fuel. If you look at the DJ base maps, they are pouring in fuel as well, so there must be something to that. But on the flip side of that, why with a about 30% fuel addition, why am I not seeing a detrimental fuel economy....???? My answer to that, is along my tuner's answer.... It takes a certain amount of fuel/air to move a certain weight or should I say mass along. And I can say with certainty, that with this installed, it takes a whole lot of less throttle, than it did before.
I see that DJ maps often show large numbers. In the cruising area where closed loop usually applies, I can see 10% more for the drop to 13.2 from 14.7. (I can understand 4% more for ethanol if the tuned bike is using E10 but yours isn't) and perhaps 5-10% for a change to exhaust tuning like you saw with SAS off. This gets me to 15-20% for some parts of the map. The numbers you're seeing in the 20-30% range which is hard to explain without post-combustion fresh air entering.

It would be very illuminating to see the record of actual fuel pulses as measured by a GS-911 for a given condition, before and after a change. For example, say your idling, looking at AFR, pull the SAS connector, does the actual BMSK pulse get longer, shorter or stay the same?

(I read on another site that if you don't add trims to the base map it doesn't add them to the fueling the next time you start. Recent you did Accept and Save trims and now you do have the actual vs trip computer error. Is this a possible explanation for why actual vs trip computer has previously agreed?)

Quote:
Originally Posted by ebrabaek View Post
There were 1/10'th variation in the idle AFR. That would be too small of a change in the AFR to alert me towards reversion air. Bike just likes this magic BMW SAS being active, me thinks. I am hoping that one day Joel would see this thread, and explain that the SAS really consist of a flux capacitor, and Photon torpedo's..... Gotta be somming like'at......

( PS) video of air reversion test uploading to you tube at the moment. Be here shortly.
Watching your video, I see an average AFR of 15 dropping to 14.3/4/5.


Quote:
Originally Posted by ebrabaek View Post
Mind you that the high AFR was seen not at WOT, but as " driving the dyno" And not seen until open pipe, and headers. I do not think stock will yield that high AFR, but again, the stock runs went by very quick.
Since it wasn't seen until the open pipe and headers, with SAS was enabled and the VE near idle can't be changed by more than a few percent, isn't it clearer to you that it's fresh air coming in by the tailpipe or a leak?

Quote:
Originally Posted by ebrabaek View Post
Gas mileage..... Just filled up.... 132 miles, and 3.01 gal. ( US....=3.78 Liters) 43.85 mpg. Computer said 50 mpg. This is the first time they have disagreed. One day, I will have to put this thing on the dyno, with both SAS On, and OFF...... Puzzled here.
This is interesting and has occurred after you Accepted Trims and Saved them to the Fuel Map.

At this point I would ride what runs best and then as you have time and interest, try and isolate one issue at a time. The goals being to end up with trim values that are consistent and can be explained and an even better running bike that others might emulate.

If you could get the entire stock exhaust back on the bike, reset the BMSK, fuel map to 0 and then start Autotune, we could work through the open questions one at a time. Maybe by that time someone will install and log LC-1 data on the F800GS on a stock bike so that there is a baseline for independent comparison.

On Terry's GSA we've got tens of hours of recorded data showing how predictably the BMSK fuels his motorcycle. It's only leaner than 14.7/13.8 when decelerating, as you would expect. (On mine, I have hundreds of lc-1-hours and two bungs weled in so that I can test sensor A against sensor B.)

Enjoy the holiday, rains stopped here, time to ride.
roger 04 rt is offline   Reply With Quote
Old 05-27-2013, 06:46 AM   #447
ebrabaek OP
Beastly Adventurer
 
Joined: Jan 2010
Location: Grand Valley, Colorado
Oddometer: 4,672
Quote:
Originally Posted by roger 04 rt View Post
Here is that dyno graph. On the left-hand side of the AFR chart, the lines all start at AFRs that aren't real, but an anomaly created by the tailpipe O2 sensor. You'll see that the BLUE line (stock) has recorded horsepower between 3000-4000 RPM, but the the BLUE AFR line is way above 16:1 below 5000 RPM. That is the air reversion Dyno error I'm talking about. (BTW, does the "stock" run include a catalytic converter?)
Yes, it does. First two runs includes the CAT.

Also note that all the AFR lines fall around or above 14:1. I know from measurements my bike (see above chart) and the data from Terry's GSA BMSK that the BMSK will be fueling to at least 13:1 at WOT. For some reason the dynos tailpipe sniffer doesn't get it right there either.
I am not accepting this to be a detriment.






I see that DJ maps often show large numbers. In the cruising area where closed loop usually applies, I can see 10% more for the drop to 13.2 from 14.7. (I can understand 4% more for ethanol if the tuned bike is using E10 but yours isn't) and perhaps 5-10% for a change to exhaust tuning like you saw with SAS off. This gets me to 15-20% for some parts of the map. The numbers you're seeing in the 20-30% range which is hard to explain without post-combustion fresh air entering.

It would be very illuminating to see the record of actual fuel pulses as measured by a GS-911 for a given condition, before and after a change. For example, say your idling, looking at AFR, pull the SAS connector, does the actual BMSK pulse get longer, shorter or stay the same?
At some point, I will need to get the GS-911.


(I read on another site that if you don't add trims to the base map it doesn't add them to the fueling the next time you start. Recent you did Accept and Save trims and now you do have the actual vs trip computer error. Is this a possible explanation for why actual vs trip computer has previously agreed?)
The only time I accepted trims, is experimenting with SAS OFF. I did this to try to eliminate any possibilities that the PC-5/AT was not running right.



Watching your video, I see an average AFR of 15 dropping to 14.3/4/5.
It behaved the same as without the towel.




Since it wasn't seen until the open pipe and headers, with SAS was enabled and the VE near idle can't be changed by more than a few percent, isn't it clearer to you that it's fresh air coming in by the tailpipe or a leak?
No. But perhaps by the SAS.



This is interesting and has occurred after you Accepted Trims and Saved them to the Fuel Map.

At this point I would ride what runs best and then as you have time and interest, try and isolate one issue at a time. The goals being to end up with trim values that are consistent and can be explained and an even better running bike that others might emulate.
Yeppers..... I will ride for at least 100 miles, before accepting trims, and go from there. See where this takes me.

If you could get the entire stock exhaust back on the bike, reset the BMSK, fuel map to 0 and then start Autotune, we could work through the open questions one at a time. Maybe by that time someone will install and log LC-1 data on the F800GS on a stock bike so that there is a baseline for independent comparison.
Not an option for me. Sorry.


On Terry's GSA we've got tens of hours of recorded data showing how predictably the BMSK fuels his motorcycle. It's only leaner than 14.7/13.8 when decelerating, as you would expect. (On mine, I have hundreds of lc-1-hours and two bungs weled in so that I can test sensor A against sensor B.)

Enjoy the holiday, rains stopped here, time to ride.

Thanks Roger. I wish we could get some rain here in West Texas, as I am not sure how it looks anymore...... Have a great holiday as well.


__________________


Erling
ebrabaek is offline   Reply With Quote
Old 05-28-2013, 11:51 AM   #448
roger 04 rt
Beastly Adventurer
 
roger 04 rt's Avatar
 
Joined: Nov 2011
Location: Massachusetts
Oddometer: 2,150
Understand what you're saying but the AFRs 18:1 and above, the SAS being enabled and the amounts of fuel being added still suggest to me that you have fresh air entering the exhaust, leading to less than the best result.

I looked at the thread where you built the exhaust, very neat! And well done.

Dynojet is very aware of these "fresh air issues" as are many tuners. Here is a good video from Dynojet discussing wideband sensor location. If you could get a bung in one of the pipes, say 18-20" from the valves, block SAS and then let Autotune work on a base map with 5-10% fuel added you might get a better results.




Roger
roger 04 rt is offline   Reply With Quote
Old 05-28-2013, 02:56 PM   #449
ebrabaek OP
Beastly Adventurer
 
Joined: Jan 2010
Location: Grand Valley, Colorado
Oddometer: 4,672
Quote:
Originally Posted by roger 04 rt View Post
Understand what you're saying but the AFRs 18:1 and above, the SAS being enabled and the amounts of fuel being added still suggest to me that you have fresh air entering the exhaust, leading to less than the best result.

I looked at the thread where you built the exhaust, very neat! And well done.

Dynojet is very aware of these "fresh air issues" as are many tuners. Here is a good video from Dynojet discussing wideband sensor location. If you could get a bung in one of the pipes, say 18-20" from the valves, block SAS and then let Autotune work on a base map with 5-10% fuel added you might get a better results.




Roger
Good Video Roger. Thanks for the pipe praise..... A very unique piece indeed. The only way I could get a O2 sensor that close, would be to have two of them, as the stock location, is just as the two headers comes together, So That is not an option. It will be a while for more riding..... June 21'st to be exactly, as We leave for Scandinavia on Thursday.
__________________


Erling
ebrabaek is offline   Reply With Quote
Old 05-28-2013, 03:29 PM   #450
roger 04 rt
Beastly Adventurer
 
roger 04 rt's Avatar
 
Joined: Nov 2011
Location: Massachusetts
Oddometer: 2,150
The dynojet people seem to think that you could just pick one pipe.

I'm borrowing a PC III USB (no Autotune) and planning to run it in my second bung to see how it compares to the LC-1. Should have some data by the time you're back.

Safe travels.
roger 04 rt is offline   Reply With Quote
Reply

Share

Thread Tools Search this Thread
Search this Thread:

.
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On

Forum Jump


Times are GMT -7.   It's 06:44 PM.


Powered by vBulletin® Version 3.8.5
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright ADVrider 2011-2014