TTS - 96" with 255 cams

[07heri]

Steve I didn't see your post until now.  AFTER I made another vtune this morning.

Steve and Mayor:  attached is the VE change graph from todays run.  Highlighted it also.

It looks like the high cells calmed down but I now picked some up in higher rpm ranges.  I don't ever see this bike going over 4000 rpm's so not sure how the high rpm cells will affect the tune.

I'm not real happy with todays run.  Somehow I missed a bunch of cells that I really wanted to get. 

Plan to make another run in the morning after I hear from you guys.

I do have a couple questions though.  What causes the high VE's?  I looked at a bunch of 96" tables and it seems mine are quite a bit higher than norm.

Also, how will the high VE's affect MPG?  I'm guessing not in a positive way which kinda concerns me.

[attachment removed after 60 days by system]

[Steve Cole]

The VE values will change based on how the engine performs and the combination of parts. Dropping the EGR down will lower the value that is needed int he VE table and since your rear cylinder was running higher values than the front I knew you could drop that range to give you a little more room to work with.

[mayor]

Also, how will the high VE's affect MPG?  I'm guessing not in a positive way which kinda concerns me.

no worries, your milage will not suffer due to the high numbers in the majority of the ve tables.  in short- most of the high numbers are not real, the values in those cells are being affected by other tables (hence the bold numbers).  What may affect the milage is not getting good data further out in the TPS % in the early rpm's. 

[07heri]

The VE values will change based on how the engine performs and the combination of parts. Dropping the EGR down will lower the value that is needed int he VE table and since your rear cylinder was running higher values than the front I knew you could drop that range to give you a little more room to work with.

Steve, the cells I was concerned about actually came down on the 3rd run.  But, they're still red/pink so another run may turn them white or drop them a tad more.  Should I still lower the EGR values or make another run and see what happens? 

[07heri]

Also, how will the high VE's affect MPG?  I'm guessing not in a positive way which kinda concerns me.

no worries, your milage will not suffer due to the high numbers in the majority of the ve tables.  in short- most of the high numbers are not real, the values in those cells are being affected by other tables (hence the bold numbers).  What may affect the milage is not getting good data further out in the TPS % in the early rpm's.

Ok here's my question.  Lets say I go out and work on 3000 and below to get the cells filled out to the right.  I figure out a way to make it happen and get them white.  When I run VTune it will only show data in cells 3000 and below correct?

The data above 3000 rpm's will still remain valid because it wasn't changed?

So, could my final tune actually be developed from more than one actual vtune run?  Could I put all the white cell values on a spreadsheet and keep adding white cell values as I turn them white on subsequent vtune runs. 

The reason I'm asking is it's obvious I have more work to do.  But if I can focus on certain weak areas without redoing the same white cells every time it would save time and gas.  When I look at the histogram I'm hitting the same cells 500 times every run for no reason if they're already white. 

So, basically what I'm asking is can the final map be piecemealed together as the cells turn white without having the final vtune look perfect?  If so, I'll go out and spend more time in the cells that need to be worked and not worry about the cells that are already white.

[mayor]

Ok here's my question.  Lets say I go out and work on 3000 and below to get the cells filled out to the right.  I figure out a way to make it happen and get them white.  When I run VTune it will only show data in cells 3000 and below correct?

The data above 3000 rpm's will still remain valid because it wasn't changed?

no, not necessarily.  You may inadvertently collect data in the areas above 3,000, and that data may be used in the subsequent vtuned cal.  If you do try to just target certain areas, what you will want to do is copy the good values from your last vtuned calibration and paste them in the unintentionally populated cells (look for yellow boxes on your vtune ve charts). Just keep in mind that if you adjust the egr tables at 3,200 rpm that the 3,500 rpm ve rows will likely be affected (meaning the values there may no longer be accurate in the affected areas (highlighted cells).

So, could my final tune actually be developed from more than one actual vtune run?  Could I put all the white cell values on a spreadsheet and keep adding white cell values as I turn them white on subsequent vtune runs. 

technically..probably...but things like weather conditions and gas blends can easily change what might be a white cell over to a light pink cell.  I might be in a minority, but I think in the case of vtuning typical closed loop cells light pink is not a big concern.   

Should I still lower the EGR values or make another run and see what happens?

I think you can safely lower the front EGR table 10-15% from 3,600 up to 6,400, and lower the rear EGR table 10-15% from 3,200 up to 6,400.  Just remember, if you adjust that table you must retune to dial in the new values.  You do not have to adjust the EGR tables, but I personally feel that limiting the change between EGR affected and non-EGR affected cells helps prevent potential lean transitions when switching from EGR table affected cells to non EGR table affected cells. 

[07heri]

Mayor the bike runs worse than it did as a Stage 1 without the cams.  Even in the lower rpms where the cells are all white the bike runs terrible.

In your opinion what do I need to do next? 

[Steve Cole]

Do a recording in DataMaster of a cold start up on the bike. You need to record the O2 sensors. Just start the motor and let it sit an run give it no throttle. You should see the O2 sensors start working in about 2 minutes and they should start switching up and down and keep doing it. If it takes too long or they do not come on and stay on then that maybe an issue.

[07heri]

Steve:  Here's the scan.  What voltages should we be seeing?

[attachment removed after 60 days by system]

[mayor]

this one's all Steve....I haven't passed my o2 sensor value correspondnace class yet.     after looking at your ve's again, the rear really doesn't look right.   

[Steve Cole]

Looking at the data there is something going on that's not right. If you look at the length of time from startup until the sensors get going and keep going, it's too long and the front is not switching properly. Then toward the end of the recording the rear stops switching, then goes for a burt and then stops again. My guess is that the sensors are not reading the exhaust correctly. This could be due to sensor placement or the sensors are starting to go bad. You can try to record another test and once the O2's start switching run the engine up to 2000 and hold it for awhile and see if they settle down and work correctly at the high engine speed. If they do then I would say it poor sensor placement if they do not then I would guess at sensors going south.

[07heri]

The head pipes are stock Harley.  Have you seen OEM head pipes that had placement issues?

Also Steve, what voltage should we be seeing?

If it was running at exactly 14.6 as the map is set for what would be an acceptable range?

My CLB is set at 759 so how would that affect the voltage?

I guess these questions could be asked at both idle and also at higher rpm's, if it even matters.

I hope I'm not asking too many questions but it would help me, and other too i bet, if we had a good idea of what the correct voltages look like and what to look for down the road if we suspect a bad sensor.

Lastly, are these Bosch 12028 sensors that I can get at the local parts store?  Either way they have 20,000 miles on them so I will change them tomorrow and run another data log.

[Steve Cole]

I hate to see people buy parts to guess at the solution. YES, HD stock pipes have bad sensor positions. It works for them only because they come stock with restrictive mufflers which keeps the exhaust pressure high enough at low engine speeds to force the exhaust in and around the sensor. The reason for running engine speed up is that will increase the exhaust pressure and if the sensors settle down it is a good indication of what is going on. There is no set number that one could use as a pass/fail test for the O2 voltage in this case. We just have to look at how it is responding and in a perfect world the graph would show both cylinders working the same so the O2 voltage should look the same. In your case what I see if them dropping in and out at idle and the pattern on the graph not near the same.

One other thing I did not mention but I probably should have was to check for exhaust leaks, so please make sure that's OK.

CLB setting is an offset the computer uses to set when to switch, so in a perfect world that would be the average voltage of the output of the O2 sensor with no additional offsets. The only issue is that the Delphi code has various offsets based on model year and model of bike and to tell you the truth  I have to look up the offsets, if any, as I do not remember them off the top of my head. The only question that would be too many is the one you do not ask. As for the Bosch 12028 I cannot answer that one, but if it's a replacement sensor that plugs in WITHOUT cutting wires I would think it would be OK. I do not like the universal cut the wire O2's for any application.

[07heri]

I'm not a big fan of throwing parts at a problem either.  Here is another data log for the O2's, warm up, idle, 2000 rpm's.  If you need more data at idle and 2000 let me know and I will run it longer in each area after I swap exhaust seals and plugs.

I have a set of new plugs and exhaust gaskets so I will swap them out this afternoon.

Afterwards I will run another data log and post to see if there are any changes.

[attachment removed after 60 days by system]

[Steve Cole]

Take a look at the Rear O2 sensor voltage on the graph in DataMaster and you can really see a problem when the engine RPM comes up on the rear cylinder. The sensor almost stops switch most of the time. Try swapping the sensors front to rear and rerun the test. If the problem follow to the front cylinder than I would replace that sensor.

[07heri]

Steve....new data log

Exhaust gaskets replaced
All exhaust joints sealed
New plugs
Sensors reversed

[attachment removed after 60 days by system]

[07heri]

Did you guys give up?

[mayor]

nope....just don't know what that data is telling me.    

[Steve Cole]

Just have been buried at work. I will try to look at this over the weekend but the weather is nice and I'm heading out for a ride up the coast.

[Steve Cole]

Well enjoyed several hundred miles of riding in beautiful weather this weekend. Did about 300 yesterday and about 400 more today. Now maybe work tomorrow will not seem so bad.

Looking at your recording, you ran the higher RPM for a little over 1 minute. During that time the O2 sensors were very inconsistent in working in both the front and rear cylinder. Since it seems to be both cylinders and you have already reversed the sensors in there location it would be my guess that the sensor placement maybe the cause of your problem. My next move would be to cut the length of the bung down to get as much of the sensor into the exhaust stream as you can. I really cannot see it being both sensors bad and doing the same thing but it is possible.

[Rider57]

I would have to agree with SC on this one.(rare_lol)
But be careful to get it square.

[On Deck]

Just last week we had a bike from another shop come in...running like a POS... it had a 103 with 204 Map in it after checking with the other Dealer we found that the bike was a 96 ...not a 103...and they used the 103 map due to the 204 cams they had installed ... after we Reduced the Engine Displacement to 96.7 in and reflashed it...and did a few Smart tune runs ( using a SESPT ) it ran fine... Data runs after showed no timing being pulled from any place it the runs.....so as Mayor said you still need to check Data runs to check timing...the 255 Maps are far diff timing wise then a 204 would be ....again as he said a complete Dyno tune would be the most complete option

Ha!  Sounds like my experience a couple months back with a local dyno dude.  Have a 96" w/ 204's and he loaded a 103 map.  Kept stalling out during idle, so I loaded CAA176 map.  After 3 Vtunes it runs only OK.  Need to better understand and do a few more.

[07heri]

Well enjoyed several hundred miles of riding in beautiful weather this weekend. Did about 300 yesterday and about 400 more today. Now maybe work tomorrow will not seem so bad.

Looking at your recording, you ran the higher RPM for a little over 1 minute. During that time the O2 sensors were very inconsistent in working in both the front and rear cylinder. Since it seems to be both cylinders and you have already reversed the sensors in there location it would be my guess that the sensor placement maybe the cause of your problem. My next move would be to cut the length of the bung down to get as much of the sensor into the exhaust stream as you can. I really cannot see it being both sensors bad and doing the same thing but it is possible.

It isn't going to happen with the stock bungs.  I measured them when I swapped fr to rr.  I have approx 1/4" on the short side of the bung.  If I knock that down to just shy of flush with the pipe there won't be any threads left.  What a total hackjob by the folks at HD.  There's plenty of room where they could have mounted the bungs at a 90 degree angle to pipe, yet they chose to mount the bungs at an angle.  I have a good half inch in the pipe on the short side but the long side is still in the bung. 

[07heri]

Steve...I read and read and read all weekend.  My head is spinning....lol.  I basically started from scratch with the tune.  I can't tell for sure but I believe most of my problems came from exhaust leaks.  The exhaust is tight after sealing all the joints and new head seals.  This is an 07 Heritage.  Where the balance tube connects to the front and rear header pipes is a simple push fit seal similar to the head seals.  So I run down to the local HD boutique and look at a newer Heritage.  Low and behold it has clamps to squish the seal.  I had the parts guy look at the 07 diagram and it doesn't list any clamps.  So Harley must have figured out they would leak and did a fix.    I'm sure my head seals were leaking because they have been off without replacing the seals every time.  I know, bad practice, but it is what it is.  Anyway, I got the exhaust all sealed up tight.

Next experiment was to do a little shadetree mod to the baffles.  They are pretty much open baffles, Rineharts...stuffed in a 3" cans.  There really wasn't much restriction to speak of.  Good example of "sounds great...runs like crap".  What I did was cut 3 sides of a 1 1/4 square into the baffle, and bent the tab down into the exhaust flow of the baffle.  Granted, this was a very UNscientific idea but I was grabbing at straws.  I couldn't think that adding a little restriction would be a bad thing so I did it.  I would compare it to how the Cycle Shacks and the SE have the lip inside that is bent inward. 

I ran 3 different maps and the CPZ176 ended being the one that I went with.

I ran the 5/2 Estimated cam setting, as well as the 4/2 and 3/2.  The cam analysis said 3 but I wanted to "feel" it for myself.  The cam data run ( 3/2 ) was the one I started with and still using.  There wasn't really a huge difference but the idle sounded more stable and tighter, (for lack of a better word) with exactly what the cam data analysis told me.

Got in 4 VTunes and the bike is running is so much better.  It has never been as smooth as it is now.  The rear VE's came down a bunch, the front came down a little too, and both are a hell of alot closer to each other than what they were when I started.  Rear still a little higherbut way better.  I'm not a tuner but I would bet that the exhaust leaks played a big part in the wild numbers I was initially getting.  Still need 4 or 5 more VTunes to get it where I want it.  I should get those done in the next day or two and I will post here for the critics...lol. 

Overall, I'm much happier today than I was last week...and obviously the bike is too.  I think once the VE's are done and I tackle the timing it's going to run pretty nice for a pretty much stock bike. 

The only thing I did notice is a slight rich smell after it starts.  Let it idle for a few minutes, or smell it after a run, and I don't smell anything.  Just right when it starts.

I want to thank you and Mayor for helping me with this.  I'm far from being finished so stand by for more questions when it comes time to get the timing squared away.   

Stay tuned.....(no pun intended)

[mayor]

sounds like you are on the right path.  I would think all the things you mentioned you did, should help make the bike run better.