Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[FLTRI]

Yep Steve, it is hard to believe the factory set it up that way...  Who would have thunk it?  That motor had a PC5 for it sitting on my desk before the I picked it up from the dealer.

Issues jump right out at you when you have the tools to spot them.... 

Andy

Are you going to use the new Power Commander V PT-i (w/3bar Map sensor) to tune the turbo motor?
Bob

[hrdtail78]

That's all good stuff whittlebeast.  Are you going to show narrow band sensors for WOT tuning.  I thought your name was brought up because you did this and maybe, accusingly, somebody said you couldn't?

[whittlebeast]

Narrow bands are wicked fast and very accurate at 14.7 AFR.  For the most part I do not use them at any other voltage.  As far as using them for WOT, if I ever see them swing below 14.7, I instantly add fuel there in the MAP.  Like 10%

Any more, I do not tune with narrow bands but if I had to do a Harley with NBs, I would tune thru 80 KPA at 14.7 to get the VEs dialed in and then go to say 13.0 AFR and do things with MegalogviewerHD and work off the MAPxRPM/100 VS Pulsewith where the motor was going over 80 KPA.  Most Harleys spend so little time up they, why worry about it.  If you really are that worried about that part of the map, purchase a PC5/Autotune/LCD-200 and do it right.  I have not played with the Powervision with the wideband stuff lately, but I am sure it will also work fine.  When I did the Sporty with the narrow bands and then did the early development work on the Dynojet wideband, the tune changed very little.  Most of the changes were in the overrun part of the map.

Bottom line is the only way I know of to do this stuff right at WOT is get the widebands or do it with math and get plenty close enough given the motors we are talking about here.  Generally, I stay off dynos.  I like data loggers.

Doing a turbo soon may wind up taking home grown Harley tuning to a new level.  I wonder how hard it will be to keep the heads cool....

Andy

[whittlebeast]

Are you going to use the new Power Commander V PT-i (w/3bar Map sensor) to tune the turbo motor?
Bob

That motor will be 100% Megasquirt if I am still involved.  Life is way too short to fight anything less capable.  We are going to try to do it with a MicroSquirt.  If we need more processor, we will most likely go MS3 Pro.  We are trying to keep the packaging reasonable.

Andy

[hrdtail78]

I have played with megalogviewer.  Still boils down to data you can collect and put into that program. Filtering data is the name of the game here and if your program now gives me more control over filtering.  I will have to revisit it.  Something I have always like with Innovate Logveiwer is you can easily take apart data and delete the area you don't want.  Meaning go collect data, and take out all steady state, and decal and just look at acceleration.

And we are still restricted on how the ECM puts out the data.  Filtering what we have collected and filtering how we collect is yet another topic.

[whittlebeast]

The data that I get off the PC5/Autotune/LCD-200 setup is coming at me a 40 samples per sec.  About 10 times faster than the Harley CAN network.  The o2s appear to be about 100 MS delay at least on the other motors I mess with.  Plenty fast enough to get really close on the AE settings.

[FLTRI]

The data that I get off the PC5/Autotune/LCD-200 setup is coming at me a 40 samples per sec.  About 10 times faster than the Harley CAN network.

Are you sure the CAN network only supplies 4 samples per second transfer rate?
I believe you have it confused with the 4hz j1850 transfer rate.
Best I can tell is the CAN transfer is at least 10 times faster than the j1850 protocol.
This would give a transfer rate of approx 40hz.
Bob

[hrdtail78]

That is faster then what I have found.  Still too slow for my liking. The engines I deal with rev to 6500 and see 41 firing events at 5000 rpm. More at 6000. In my perfect world. I would like to have a sampling rate of 2-1 type thing, and we still need to deal with the 40 the PC5 does do is at what piston location?

I brought up accel for example only. Steady state is a bit easier for sample rates and engine position.

And like you say. HFT

[Steve Cole]

The data that I get off the PC5/Autotune/LCD-200 setup is coming at me a 40 samples per sec.  About 10 times faster than the Harley CAN network.  The o2s appear to be about 100 MS delay at least on the other motors I mess with.  Plenty fast enough to get really close on the AE settings.

Is this 40 samples REAL data? I bring this up as before you posted datalogs off the Dynojet unit and the data was for the most part garbage. They were not updating the records properly, it was clear as when reviewing the data you supplied the RPM report was the exact same for 10 - 12 frames of data as were other data records and we all know that is not possible during an acceleration event. The CAN based HD bikes collect data much faster than the J1850 bikes. Currently we have the J1850 bikes going about 5 frames per second and the CAN bikes about 20 frames per second.

I also wonder why you would use a made up formula? "MAPxRPM/100 VS Pulsewith" what do you feel that MAP*RPM/100 is getting you? The divide by 100 just changes the scale some, so the real question is what do yo feel that MAP * RPM gets you?

[whittlebeast]

Is this 40 samples REAL data? I bring this up as before you posted datalogs off the Dynojet unit and the data was for the most part garbage. They were not updating the records properly, it was clear as when reviewing the data you supplied the RPM report was the exact same for 10 - 12 frames of data as were other data records and we all know that is not possible during an acceleration event. The CAN based HD bikes collect data much faster than the J1850 bikes. Currently we have the J1850 bikes going about 5 frames per second and the CAN bikes about 20 frames per second.

I also wonder why you would use a made up formula? "MAPxRPM/100 VS Pulsewith" what do you feel that MAP*RPM/100 is getting you? The divide by 100 just changes the scale some, so the real question is what do yo feel that MAP * RPM gets you?

Cool, feel free to email me a log off one of the new CAN setups running the street or even better off a road race circuit.  About an hours worth would be great.  The harder the bike is being ridden, the better.

Regarding the 40 samples per sec, that is off a PC5 setup using a Wideband2 and recording on a LCD-200.  That is has nothing to do with any Harley communications.

MAPxRPM come from the basics of engine modeling.  Try it on any motor that has a MAF, MAP signal and RPM.  Feel free to try it.  Did you reed the webpage I posted?

Have Fun Tuning (HFT works for me....)

Andy

[Steve Cole]

Cool, feel free to email me a log off one of the new CAN setups running the street or even better off a road race circuit.  About an hours worth would be great.  The harder the bike is being ridden, the better.

Regarding the 40 samples per sec, that is off a PC5 setup using a Wideband2 and recording on a LCD-200.  That is has nothing to do with any Harley communications.

MAPxRPM come from the basics of engine modeling.  Try it on any motor that has a MAF, MAP signal and RPM.  Feel free to try it.  Did you reed the webpage I posted?

Have Fun Tuning (HFT works for me....)

Andy

Andy

The data you post before was off the Dynojet unit too, it was not and is not an issue with the HD data buss that is why I asked. I would assume this means you have not checked it out to see if its real or not. As far as RPM*MAP that is just two variables getting multiplied. How about what you expect it to show you as you seemed to skip over answering any of the questions as to what you hope it's going to show you that they do not already show by themselves.

[whittlebeast]

Steve

May I suggest that you create the function and compare that to total fuel flow of the two injectors on a 50 or so different motors.  Also try comparing to the MAF readings.  You may want to reconsider your conclusions.

Just out of curiosity, I just ran one of my old Harley tunes thru a few calcs and applied a few filters and this is what I got.  It works at expected...



http://www.ncs-stl.com/PVTune/MAPxRPM.png

Andy

[whittlebeast]

Here is a plot that is a little more interesting.  If you create a new field called

TPSsqrt  and define that as 10 * sqrt([TPS])

Things get even more clear whats going on



http://www.ncs-stl.com/PVTune/Throttling.png

HFT

Andy

[Steve Cole]

Steve

May I suggest that you create the function and compare that to total fuel flow of the two injectors on a 50 or so different motors.  Also try comparing to the MAF readings.  You may want to reconsider your conclusions.

Just out of curiosity, I just ran one of my old Harley tunes thru a few calcs and applied a few filters and this is what I got.  It works at expected...

Andy

Andy

Since a HD engine doesn't use a MAF so that calculation is meaningless here. You keep saying it "works at expected" but I'm just trying to understand just what is it you expect? How about a straight answer.

[redmtrckl]

Jeez, I think if I drink me a bunch of moonshine to kill off all my weak and defective brain cells then with my healthy cells doing all the work I may be able to understand all these modern art graphs.
Maybe my headache will go away too.