My current setup is 07 Dyna (HDI model), SERT, SE a/c and SE 07 mufflers. Right now I'm using the 159AE004 map from the SERT CD which fits my setup perfectly, and the bike runs great.

I am replacing the exhaust system with a Rinehart 2-1, and I don't have a map for the new combo (Rinehart, SE a/c, 07 FXD and SERT). Dyno tuning is not an option over here.

So I'm planning about using my current map as a starting point and tweak it for the new setup with Data Mode.
I have read the SERT manual and if I understand correctly, most important tuning issues are AFR, Spark Advance and VE.

AFR and Spark Advance sound quite clear to me. As far as VE is concerned, I should record a few runs with data mode and compare Target VE (front and rear) with New VE (front and rear). Then I should replace Target VE cells with the figures of New VE recorded by Data Mode, which should be the "right" values. So the ECM is capable of determining the "right" VE figures.

My question is:
If the ECM is capable of determining the "right" VE values, why do we have to use a VE "target" table?? Couldn't the ECM take into account "right" VE values and calculate the correct amount of fuel from the "desired" AFR values from the AFR table?

Thanks a lot for your replies.

 

The V/E is a fixed table. You need to import the New V/Evalues to get it into proper range to control the AFR in open andclosed loop parameters.
It`s easy to do and it sounds like your comprehension of the manual and procedures is very good. I like this type of thread. We all need to perform more in depth procedures on our bikes instead of paying the dealers to hack it and treat your bike as a learning guinea pig. JMO.
Keep us informed of your progress. I need to do the same thing with my 08 Ultra, but I`m working on cam replacement on my 07 FLHX with the DTT/WEGO IID setup. You should have no problems based onwhat you have posted and your levelcomprehension....

 

drdiesel1
thanks for your reply


I'm afraid I don't understand your answer completely. What do you mean by "fixed table"?
My understanding is that, when tuning a bike with SERT, the tuner (in Tuning Mode) must fill in the cells for the front and rear VE (which is the VE that we expect, for that setup, at a given rpm value and throttle position). Then, in Data Mode, the tuner records some parameters of a few bike runs. New VE is one of these parameters, which, if I understand correctly, is the VE that the ECM measures at a given rpm value and throttle position. So the next step for the tuner is REPLACING the VE figures with the New VE figures which have been detected by the ECM.

This is what is obscure to me:
"if the ECM can detect New VE at a given rpm value and throttle position, why does the tuner have to copy these figures into the VE tables?"

I mean, we should not need VE tables at all. All we need is filling in the AFR table with the figures we prefer, and let the ECM calculate the "real" VE and therefore (given the AFR figures in the AFR table) the correct amount of fuel for each cylinder....

But this is not the case so I wonder where am I wrong.

One more question.
As a rule of thumb, what is better to tune first:
- AFR?
- Spark Timing?
- VE?

Thanks again.

 

Your not wrong at all.
The V/E tables are there to control closed and open loop parameters. This allows simple settings without complexity of tuning. You avoid data conflicts as well.
You right on the money. Don`t worry about it...My method is the K.I.S.S. method...Keep. It. Simple. Stupid.....[8D]Not referring to you. Just don`t over think/complicate things.
My personal opinion is to do the V/E table then the timing table. Fuel needs will override spark control as well as cause spark knock issues, so do the AFR stuff first. Hell when your done maybe I could use your map too.....

 

The AFR values should be set to a safe fixed amount beforehand, then the dyno run at every cell is made and the appropriate VE is entered where the measured AFR matches the set one.

Then the AFRs can be set to whatever values you want and they should be good since they're based on the VE settings.

The VE New only gets updated during closed loop and is the result of multiplying the set VE by the AFV (adaptive fuel value), which itself is determined by the difference between resultant and set O2 sensor target voltage.

In theory, you should not obtain a VE New that's different from the set VE if you run under the same atmospheric conditions and the running gear is in the same state of tune/cleanliness. That's if the VE was set properly to begin with.

So the order is 1) AFR, 2) VE, 3) AFR, and 4) Spark timing. As I understand it myself, at any rate. It's how I would do it.

 

The ECM uses the VE New values that it calculates, and it is calculating constantly.When you get the VE tables real close to the VE New values, the ECMarrives at the VE New values much quicker as it doesn't have to correct as much. As with any closed loop control, the less error the controller has to correct, the tighter the controlled processwill be.You'll never get the VE tablesperfect, as you will see minor variations in the VE New values at the same rpm, AFR desiredand throttle position. Below is a sample from a data run showing variations of as much as 2.5% in VE New values. (edit: as I suspected, the numbers didn't format properly)

RPM............AFR Desired...........Throttle %........VE Front.........VE New Front..........VE Rear...........VE New Rear
2041.............14.6........................6.8.. .................90.....................88.5...... ...................93.5................93
2041.............14.6........................6.8.. .................90.....................86........ ....................93.5................91.5
2041.............14.6........................6.8.. .................90.....................86........ ....................93.5................91.5
2041.............14.6........................6.8.. .................90.....................88........ ....................93.5................91.5

 

You should use a fixed-width font to tabulate your data in a text editor (notepad or such), then cut and paste it here between the "code" tags. Thus (quote to see):
I understand what you're saying, but you're not going to get every VE table cell tuned that way, are you? Full throttle, full load, 4000 RPM at 14.6 (14.1 if CLB maxxed-out) AFR? Sounds fairly dangerous to me.

The user's manual covers the initial VE setting process fairly succinctly, as I recall.

ste71: if you cannot get to a dyno you really should seek help finding a suitable map to start from.

 

Thanks for the tip on code tags, I figured there must be an easier way.
I am in no way suggesting that anyone set all the cells in the air-fuel table to 14.6 just to obtain VE New values (you do get VE New values that are different than the VE tables at AFR desiredother than 14.6, though). All I'm saying is that the VE New values obtained from data mode can vary somewhat at the same throttle position and rpm.When youexportdata and open the csv file in Excel, you can first sort the data by rpm then by throttle position. You can then typically find several instances of VE New values at the same rpm and throttle position which can easily be averaged. The averaged value is what would be used in the VE tables. I'm aware of the proceedure for VE table tuning on a dyno. But if one chooses not to put their bike on a dyno, this is an alternative.
ste71 asks a valid question as to why the ECM doesn't store the VE new values instead of constantly calculating new values and using them. I believe that it probably has the capability to do so.

 

It stores the last "good" AFV (adaptive fuel value) derived while in closed-loop. It's merely a multiplier against any particular VE table and is an extremely quick calculation (.867 [VE] × .925 [AFV] = .802 [VE New], as an example). The VE New is then used to calculate the required injector duty cycle.

That's why you get VE New values even in open loop. It's derived from the offset against the hard-coded VE required to obtain the desired O2 sensor reading while in closed loop.

I forgot to finish the thought. Any VE New values in open-loop areas are not indicative of what that particular cell actually needs. All it means is that when in closed-loop, in whatever area which produced the latest AFV, that area was in need of correction. It's totally assumed that the entire VE table has been populated with good figures on a dyno with exhaust gas analysis. The only way you'll be able to correctly populate the entire VE table without a dyno would be to set everything to 14.6, run it in every cell long enough for a new AFV to be generated, and emplace the resultant corrected VE value. And I'm not suggesting that be done.

 

If the ECM learns and stores an "adaptive fuel value" for each cell in the VE tables to be used as a correction to the programmed VE values, would one not expect the VE New values to be the same every time at a particular throttle position and rpm (assuming operating inclosed loop)? I'm seeing as much as 5% variation in VE New values in the same data file at the exact same throttle positions and rpms.
Is the AFVstored in non-volatile memory, or is it relearned every time the bike is started and run? If it is stored in non-volatile memory,the ECM would need to know that it needs to learn a new AFV if the programmed VEvalues are changed. Perhaps when you program the ECM with a new calibration, the ECM starts over with the AFV's?
I'm not challenging what you're saying, just trying to understand how this works. There was a race tuner guy from Harley at Sturgis this summer preaching the benefits of collecting VE New numbers in data mode and correcting the VE tables accordingly. He said when you do this, the desired AFR will be achieved much sooner as the ECM travels from one VE cell to another.