TTS Tuning Member 7Remmag's Bike

[mayor]

no worries 7mm.    Keep in mind, if you want to slow up the tuning pace (meaning skip some days, and just enjoy the ride)...you can always flash your last datarecording tested cal CAA176-03-A0-newcal-SWAG-003, which is set for everday riding. 

[7remmag]

It sounds and looks like it's almost there so I'll stick with it for now. Tonight I'll probably stick around here and do a little maintenance and cleaning on the bike if I can get away with it.  Sounds like I'm grillin and chillin with the little lady, but am sure I can grab a brewski and get a little clean up time in between flippin the tasties .  Then tomorrow    Getting ready to go to the Doc so hopefully everything is all clear there. Tomorrow is high of 93 and windy so will get up early and get everything done before it gets to hot.
Later
7

[7remmag]

Hey mayor I read that thread that you posted on timing.  Very interesting although it's still confusing to me were and how to look for it at times,  but you said that timing was a dark world of it's own. (Maybe not those exact words).  I do remember a couple times I couldn't wait to get home or didn't know if I was going to get home, because of vibration in certain areas.  I wondered how all that was gone the next time  .  Looking over some of the previous post on my tune again I am starting to understand a bit more.   I'll get that run done tonight,  the little lady is going to a relay for life thing so I'll get plenty of quality riding time.
Thanks 7

[mayor]

I think your VE's are dialed in pretty good now:
   

I sent you 1 cal last night, and two this morning to try data recordings on. I removed some timing where you where having the timing pulled on the last data recording, and added some timing in areas that we were trending low from the base cal (when compared to original, keep in mind we started this particular cal with timing derived from the last cal we tried from DPC), and I also took some upper rpm areas out of closed loop (higher map at 4k, and all of 4.5k rows) on the cal-004. 

[7remmag]

I'll probably get started around noon,  I'm on call till then
Thanks

[mayor]

no worries, I won't be able to look at data recordings until later tonight anyway.   The wife and I are about ready to head south on a ride for lunch.   

[mayor]

it's been a couple of days since I've updated this thread.  Since my last posting, 7mm has run 6 more calibrations. Most of these cals were baby steps testing various areas of concern.  We've been continuing the plan to get the timing back to the base cal levels, and see if the engine would take more timing. 

just to catch up the folks just joining, or those that haven't read the whole way through the thread.  We started this cal based on the timing that we were running from another cal.  That other cal was pulling timing like crazy right in the sweet spot (2-3k rpm 50-70 kPa ranges), to the point that we gave up on trying to make that cal work. One the latest data recording, 7mm had no timing being pulled so we will continue our efforts to make the bike as happy as possible.   

here's the data recording:
http://www.box.net/shared/u9glirxqnu

here's how much the timing has changed since the first vtune run 7mm made on this cal:
front:


rear:


We have been able to add a substanial amount of the timing we were pulling from the previous cal back in.  Luckily we caught the open loop (lean) trend on the previous cal, or 7mm's bike may have been very unhappy with the timing.   


and here is how we compare to the base original calibration timing tables:

front:

rear:


We still have some areas that are trending lower than the base cal timing map, but 7mm will be doing another data recording soon to see if the engine responds well to the timing changes in those locations.  We have more work to do, but we are getting there. 

data recording -010 should be attached.   

[attachment removed after 60 days by system]

[7remmag]

Hey Mayor
I ran 11 and 12 and sent them to ya.  I think I got some pretty decent runs,  11 was a little weaker in the lower RPMS 1800 to 2800.  Both seemed strong above 3400.  12 had a little ping around 2600-2800 when opening the throttle, but not much and not everytime.  I need to be a little careful now since my closed course is getting crowded.   
Later
7

[mayor]

I think I got some pretty decent runs

you did real good.  I had made a good bit of timing adjustments on cal 12, and it looks like you did a good job of testing them. 

There was some timing being pulled on cal 12 where you heard the ping, I'm still thinking it is a ve issue since the surrounding cells trend higher.  I'll make some corrections and send you another cal to try. You're in luck, only one cal tomorrow.    

[mayor]

there was no timing being pulled on cal -010 (although, keep in mind that doesn't mean every cell was checked at every load), so we decided to double down and start seeing how much we can push the timing.  Both -011 and -012 cals are versions of cal 10.  Cal -011 has some slight ve changes in open loop and some minor timing adjustments.  Cal -012 has more aggressive adjustments.  We kept the changed to cal -011 milder since no timing being pulled in cal -010 means we are getting close to a nice base line of a safe tune.  What we're trying to get now is a better tune. 

here's the changes to cal -011 from -010:

data run -011 Front VE adjustments from cal 10:


data run -011 Rear VE adjustments from cal 10:

We added VE's to some areas that we had concern with low trending VE's. The areas that we reduced the VE's above is areas that we thought might be high, but admitidly we are using SWAG to to us what to do since this is open loop tuning areas.  We're not to concerned though  , since all we're trying to do is get those areas in the suitable wide open throttle window (12.8~13.8 ). 

data run -011 Front Timing adjustments from cal 10:


data run -011 Rear Timing adjustments from cal 10:


if you look at the above, you'll notice the changes from cal -010 to cal -011 was minimal.  This was more of a test to see if we can get back to back cals that didn't pull timing, and correct a few areas that we didn't have trending well.  Unfortunately, the data recording did show one area that we had timing being pulled. 

Here's what the data recording showed on cal -011:






If you look at the above data, you will see that the timing is being pulled after a quick twist of the throttle and a subsequent switch to open loop.  The other thing to note is the VE's changed very little during that process.  Add the the fact the the rear didn't object to similar timing, and we're leaning towards a lean condition causing the trigger. 

data recording for cal -011:
http://www.box.net/shared/m8zadn7mfu4pjm8jikpn


Our field operations manager (7mm) has been commenting that the mid rpm areas have not felt as strong as he would like, so we're trying to see if we can improve those areas. The VE's look like the are trending low in the area that he said is weak, so we decided to push the VE's up on this cal and push the timing.
Here's what we did with cal -012 compared to cal -010:

data run -012 Front VE adjustments from cal 10:

We added to the Ve's in the areas that were extended VE vtune values, and we just tried to pretty up the 3D VE chart. We also used simple logic to tell us that a cam that has a peak tq around 4k should have a reasonably strong ve pattern at 3-3.5k.  My guess is these are rich values in the 80-100% columns, but shouldn't be pig rich.  Once we get the open loop tendancy to totally go away, we can work on lower the 80%-100% ve table values.

After the adjustments, the front VE table looked like this:

compare the red circled area to the blue.....and this is after we raised the ve's some in that area.  I don't find it coincidental that there timing being pulled in that area. 

data run -012 Front Timing adjustments from cal 10:


data run -012 Rear Timing adjustments from cal 10:


We used the "looks like we can add some timing here method" and also used the base line timing to make see if we could add back into areas that were previously pulling.  Our goal was to try to improve the weak feeling of the mid rpm ranges, so that's where we targeted the timing.

The data recording was encouraging, but we did end up with some pulled timing in the VE dip area of the front cylinder.  The good news was we didn't have any pulled on the rear despite the increases we made in timing.  Here's what we saw:







The case above looks like we had a low VE amount which was masked by closed loop running, but as soon as we went into open loop there was a trigger.  It could also be caused by a higher starting timing level which could have triggered an advanced timing trigger. 







The case above looks like we had a timing trigger.  It could have also been a VE trigger, but this looks like timing.  I think we might have gotten a little over zealous in the 70 kPa column.   

data recording for cal -012:
http://www.box.net/shared/mgd300dehzctpu05adq6

we used the data that we gathered from the above cals to create cal -013.  More on that one tomorrow. 

Cal -012 should be attached to this post. 

[attachment removed after 60 days by system]

[mayor]

We were still pulling a little timing on the front cylinder in cals -011 and -012, but there wasn't any pulled on the rear.  We wanted to see if we could improve the performance in the mid rpm areas, since our field operations manager feels this area is weak.  We were guessing that area was still lean, so by adding fuel (increasing VE's) we were hoping that we would be limiting the timing being pulled and should improve the power in that area. 

here's the only change to cal -013 from -012 (front ve table):

We decided to see if raising the VE's in the problem area would limit the timing being pulled.  We also decided to trim back the VE's in the heavier throttle areas in the mid-RPM's, and add just a touch to the upper rpm's. 

Here's what the front VE table for -013 looks like after the change:

as you can see, we are not adding peaks.  We are just filling in dips in areas that we would expect the VE's to be at least similar to surrounding areas. We are also using SWAG to adjust the VE's in a general arching shape at WOT that might represent what the tq curve would look like.  This is really just a guess, the only way to know for sure that we are on the right track is an external measurement.  We're not skeered though, that AFR window for power potential is pretty broad at wide open throttle (12.8~13.8 ). 

Here's what the above looks like in 3D:

I circled the area that we still have some concerns with.  This area was populated with vtune data, and the vtune process showed that this was less than 2% off.  At this point, we're not sure we agree with that.   








From the above it appears that 7mm twisted the throttle in the area that we still have concerns with, and timing was eventually pulled.  This could really be either timing or AFR issues, since the VE's trended low on the front before the trigger and the timing was also more advanced on the front at the time as well.  We're going to assume a little of both when we adjust cal -014. 







We think the above is more than likely a over advance issue. 


For the first time in several runs, we had timing being pulled on the rear cylinder.  We did not make any changes in this area on the last cal, so this was probably an area that was missed over the last data recording (this is why multiple data runs on your finished cal is important).  Here's what we saw happening on the rear cylinder:






The above appears to be timing too far advanced.  It looks like we might have been a little over zealous in that area.   

data recording for cal -013:
http://www.box.net/shared/jlz2rzm93xc14m5v36u5

cal -013 should be attached below.  We used the above information two develop cals -014 and -015.  On those cals we made added another change...but we'll keep you in suspense until 7mm gets some data runs done on those cals.   

[attachment removed after 60 days by system]

[axtell]

very well done!!!!!!!!!!

[mayor]

Thanks Ron.      I'm curious to see what our next round of changes produces as far as pulled timing.  We made a good bit of changes (mostly associated with timing), and the only change we made to the front VE chart was filling in the dip that shows up in the cal -013 3D VE chart (the circled area).  It will be interesting to see if this helps the pulled timing area.  Here's what the 3D chart for the up coming cal-014 looks like (this cal has not been tested yet):

The VE flow is much smoother in the problem area (red circle).  I circled the area that had VE's added in purple, which shows that we are not creating peaks but rather smoothing out the flow of VE's. 

[mayor]

We produced two new cals for 7mm to try based on the data we learned from cal -013.  We introduced a new change on these cals though, or more like a change back to original.  Early on when we were having problems with the timing being pulled, we decided to add an addition degree of retard to the temp correction table.  We figured it was best to go back to the base cal settings in this chart, so we ended up with this change on cal -014 when compared to -013:

We could have probably left it like it was, but we decided to use the individual areas to adjust timing since that table makes a more general decrease.  I think adjusted that table is better when you adjust the timing under cool conditions, and can know for sure that hotter conditions is causing timing to be pulled.  In this case, we missed the cooler whether by a couple of months.   

Here is the only front VE table changes we made on this cal:

We did that just to fill in the dip that we were seeing on the 3D ve chart, which just happened to coincide with an area we were getting pulled timing on earlier cals. 

Here's the changes we made to the timing charts:

Front:

The 60-70 kPa column reductions is almost covered by the 1 degree less retard we now have at temp. 

Rear:


We made some pretty good reductions, but keep in mind that we will be running one degree less offset due to the temp correction chart modifications in many of those areas too. Furthermore, many of the really big change areas is already well above base line cal timing for those cells.  Our goal in this case was to find a happy place where timing and afr's could live together in harmony, or in layman's terms...make a cal that didn't pull timing. 

Here's what the current front timing looks like compared to the original calibration:


Here's what the current rear timing looks like compared to the original calibration:


The good news was there was no timing pulled with this cal.   

the data recording for cal -014 can be found here:
http://www.box.net/shared/yc5r4fglhjvgkbex5tyh


On cal -015 we decided to see if we could invite more wide open throttle timing to the party.  We didn't make big jumps, but we made plenty of changes. 

Here's what we did to the front timing of -015 compared to cal -014:


Here's what we did to the rear timing of -015 compared to cal -014:


Here's how the timing charts from cal -015 now compare to the base calibration charts:

Front:


Rear:


This calibration did not pull any timing either.     That means we had two back to back cals where no timing is being pulled. 

the data recording for cal -015 can be found here:
http://www.box.net/shared/9ag1rkgmbry8cq4jtry5

Now that we appeared to have corrected the timing being pulled, we are going to see if we can push the timing a little in the 50-70 kPa columns on cal -016. 

[mayor]

since our last update, 7mm has been busy testing cals that we have been tweaking based on what we were seeing on data recordings.  We were making some good progress on the knock retard, but we ended up introducing another problem- vibration.  The bike wasn't pulling timing in the problem areas anymore, but it wasn't running smooth there either.  We think that it was due to raising the VE's in those areas, to a point where the o2 readings were not lining up with what was written on the VE tables close enough.  We decided to try vtuning again, and this time see if adjusting the EGR tables would elevate the VE's in the problem areas.  The problem with this train of thought is the EGR tables mostly affect the sub 60 kPa areas, but we had to test to see if this would work since we are still not where we want to be as far as the tune goes.  In looking at the 3D VE charts from the last vtune recording (vtune rd 3 run 5), you can clearly see that there is a trough in the VE's around the problem area:

Front (VTrd3rn5):


Rear (VTrd3rn5):


During the last set of data recordings, we have been slowly filling in that trough in the VE's and it has been working to improve the knock retard but it also started causing some engine non-smooth issues. We decided to see if we could adjust the EGR tables during another set of vtuning to raise the VE's up in the problem areas to flow better the surrounding rows.

here's what we've seen so far after 2 vtune runs:

Front:

The EGR table in the picture is the amount of change from the left chart to the right chart.

We used one of our data recording cals that had the VE's adjusted as our base for the vtune run, since our goal was to raise the VE's during the Vtune to where we thought the ve's should be based on surrounding trends.  Even after raising the EGR in the problem area, we didn't see much movement in the VE's there.  This confused us on whether we should raise or lower the EGR settings, then we looked at this (EGR Effect Chart vt4 run 1:

Based on this chart, we didn't have good 60 kPa data around the problem area.  We figured this is probably skewing what we were seeing.  We decided to use what we were seeing on the left side of the EGR affect line, and based on that it appeared we needed to raise the values there.   Here's what we saw after the next change:

It appears our hunch was right, the Ve's are raising throughout the length of the row.  We're guessing this means we are on the right track.  Here's what the 3D VE chart looks like after two Vtune runs:

The Red circled area has less of a trough, although a trough is still present.  The black circled area was where we raised the open loop ve's previously.  We decided not to auto-extend, since we were trying to match these two areas up. 

Rear:

The EGR table in the picture is the amount of change from the left chart to the right chart.

Not surprising, we saw more positive movement on the rear than we did the front.  I say not surprising, since the bulk of our issues was on thr front cylinder, not the rear.   


The results on this cylinder was very encouraging.

The 3D chart on the rear after two vtune runs looks much better too:

The interesting thing to note on the 3D chart on the rear is there is a much less noticable transition from the problem area to where we added.  We think this means we are close with our EGR settings on the rear cylinder. 

Based on what we were seeing outlined above, it appears that the EGR tables can help raise the entire rpm row but it may be dependant on at what MAP range the data was collected.  Or it least that's what it appears.