ECU

[jrt]

The meter really sikes on the throttle transitions, but I guess you could learn to read the steady states and average out the ups and downs. (sorry there is probably some proper jargon for that)

42181[/snapback]

 

"interpolation"?

[dlaing]

"interpolation"?

42195[/snapback]

OOps! sikes should have been spikes...I guess I have a weak inky, I mean pinky.

So let us re-phrase,

The meter really spikes on the throttle transitions, but I guess you could learn to ignore the large spikes and interpolate the average.

To do effective road dyno work, I think you would need a pretty steep hill. Just accellerating on the flats does not give the sensor time to give a good reading.

If you look at that graph, when decellerating the sensor appears to dip rich. When accellerating the sensor rises lean. It would be difficult to convert this to map changes. Although it may indicate the bike is running lean at high RPMs, it could also just be one of the inadequacies of the wide-band oxygen sensor.

But if the throttle can be held at a steady throttle position and RPM for about five seconds, the data is more useful for creating a map. The graph jumps too much for us just to look at just one or two second increments.

In any case, we also need throttle position data and there is supposed to be an option to log the voltage output of the TPS.

[Paul Minnaert]

Well this is a ecu topic, so I orderd a Moto Spezial injection. This one is build in the tacho housing. No external boxes, all in one. I'm going to make a new cableset to clean up guzzicable mess

[Cliff]

If you look at that graph, when decellerating the sensor appears to dip rich. When accellerating the sensor rises lean.

Perhaps they are real changes in mixture. Why do you think that have throttle pumps on carbs and emulation on PC. Possibly this ECU doesn't have a throttle pump emulation.

 

There's too much missing data to draw any real conclusions about this graph

[dlaing]

Perhaps they are real changes in mixture. Why do you think that have throttle pumps on carbs and emulation on PC. Possibly this ECU doesn't have a throttle pump emulation.

 

There's too much missing data to draw any real conclusions about this graph

42205[/snapback]

I can determine by looking at the map that unless a given state is held for more than three seconds, it is very difficult to know what to read.

Perhaps with more data, and more importantly with a map that has a flattened or known lambda, one could determine the value of the O2 reading when the rpm and throttle are NOT held at a steady state.

For now, it is greek to me.

Your suspecting a need for throttle pump emulation appears to be on the money.

How does a throttle pump emulator work?

I am not even sure why it is needed if the map is done properly, but perhaps with this meter, I could discover why!?!

[dlaing]

Well this is a ecu topic, so I orderd a Moto Spezial injection. This one is build in the tacho housing. No external boxes, all in one. I'm going to make a new cableset to clean up guzzicable mess

42202[/snapback]

Wow! amazing that they fit that all in one tachometer case!

Can you give us more details?

How are you going to map it?

Does it come with software?

Does it cost a lot?

May I have one?

[Cliff]

How does a throttle pump emulator work?

The implementation details can vary. Basically the idea is to be like the carbureter accelerator pump. It squirts extra fuel in as the throttle is opened.

One way if doing this is to compare the instantaneous throttle value with an averaged throttle value and use this difference to enrich the mixture. Many variations of how much more and for how long are possible.

[Paul Minnaert]

How are you going to map it? Does it come with software?

Well I get a pretty close map with it, from an engine with same tuning state. That map is procided in an big Excel file. Can be uploaded with a serial connection. There is an optimizer program with it, 2 lamda sondes.

It can be had in one case like the picture, or more traditional look with rev and tacho split over 2 cases. In the one case type, you can choose tach big and rev counter small or the other way around. There is a digital display, that can give different things, switching between them can be done with an external pushbutton. Of with a magnet outside the case.

 

 

Does it cost a lot?

I think it's a lot, but when you count marelli price & tacho's, that's also not cheap.

Think around 1300 euro. + lamda sondes+ some cables.

May I have one

Well you can buy one:-)

 

Zebulon has one:

 

[al_roethlisberger]

Man, that's a nice setup..... I every time I see Zeb's install

 

Very clean.

[JuhaV]

There is an optimizer program with it, 2 lamda sondes.

42235[/snapback]

 

Hi Paul,

 

Does this mean that you can use it in a closed-loop mode ? Have you already considered what kind of lambda probes you are going to use ?

 

The unit looks very nice, although the price is pretty high !

 

br, JuhaV

[dlaing]

Thanks Paul, more to consider!

So, our choices now are.

Stock ECU. US$100? on eBay/

plus PCIII US$180 second hand

PCIIIUSB US$257-$332

PCIII can be use a downloaded map, be tuning linked $200++, and or further tuned+++.

You may soon also be able to get an ignition module for it.

 

Tuneboy US$309 can download a map, theoretically tuning linking is possible, but more likely slow tune on the dyno will be used, prices vary, but more than the tuning link tune.

Also ships with valuable diagnostics and has ignition mapping.

 

TechnoResearch Direct Link, US$450? you could interpolate a PCIII map, borrow a PCIII to tuninglink, or more likely slow tune on the dyno+++.

Also ships with valuable diagnostics. and I believe ignition mapping.

 

 

Cliff's ECU Australian$700 you could interpolate a PCIII map, borrow a PCIII to tuninglink, or more likely slow tune on the dyno+++.

Also ships with valuable diagnostics.

And has closed loop capabilities and ignition mapping.

 

Moto Special

1300Euro Slick package replaces tach and or speedo.

you could interpolate a PCIII map, borrow a PCIII to tuninglink, or more likely slow tune on the dyno+++. Not sure of exact features, but it support dual lambda for closed loop and comes with optimizing software.

 

FIM

Probably only a choice if it is your tuners preference.

Not really an option for the home tuner.

 

We sure do have a lot of choices for a low sales model bike.

[moto]

I would rather lose a little power if it helped the bike run better.

41786[/snapback]

Huh?

[dlaing]

Huh?

42298[/snapback]

I said, "I am concerned about airflow contamination from one cylinder to the next via the crossover.

I would rather lose a little power if it helped the bike run better."

This was regarding my apparently mistaken belief that crossovers created asymetrical flow between the two cylinders on V twin engine.

If my belief was correct, then I would rather have had a balanced engine where each cylinder made 40HP, rather than a more powerful engine where one made 39HP and the other 43HP.

Everyone seemed to agree that crossovers and two into ones, are good things and that two into twos produce less power.

[moto]

It is my understanding, that the tuning link software does every throttle position in the table at either 500 rpm or 250 rpm increments depending on the model of PCIII you have.

Right, but to be precise the PCIIIUSB uses 500 rpm increments by default. 250rpm increments become available when one clicks on "Maximize Map Resolution" under "Tools". I am interested to know whether Powercommander Authorized Tuning Centers automatically tune at 250 rpm when the option is available, or if you have to ask for it. If they don't do it automatically, I wonder how much extra it costs. A word of caution: If you have the tuning done at 500 rpm increments, the software just fills in the gaps in the map with an average. My opinion is that this is not a very good idea, as adjacent cells sometimes require radically different values for optimum running.

It makes more sense to me that the computer would  use a simple PID loop (controls engineering speak) to quickly hone in on the correct fuel adjustment to achieve the desired output at the O2 sensor at each RPM step as the eddy current brake releases to let the engine rev. I have been told that if the shops computer is slow, one can definatly hear is pause at each RPM position as it sweeps up thru.

I don't think it makes the adjustments as it's letting the engine rev between steps, but rather during the steady state exactly at the steps. I think that's what's accounting for the pauses you hear.

That frankly is PCIII's claim to fame - adjustments are made realtime where with programs such as tuneboy, FIM, etc, data is taken, adjustments are made, map is uploaded, more data taken, more adjusmtents, more uploading etc.  You cannot directly work with the map in the ECU in realtime mode while the bike is running.  It must be flashed everytime an adjustment is made.

My version of TuneBoy allows me to tune in realtime. Granted, it's an alpha version, but I'm fairly certain Wayne intends to release a realtime version to the public at some point in the future.

Alledgedly, the tuninglink software is making all the adjustments and measuring while the bike is running.

If there are indeed two methods (see below), it could have been either way. Could you ask them? I am very curious about this.

MOST places will not save the monitored data for each throttle position.  They give you the last full roll on pass because us silly Americans want to see that final HP figure.  Plus, they don;t like keeping all them files ont he computer!!  If asked, I'm sure the shop could save the run data for each throttle position.  I know that Racer X likes to keep that sorta data on his bike.  The shop that did mine only saves the final full throttle roll on curves.

There may be a method (which you alluded to) wherein the "A/F ratio" is measured during an inertia run at a certain TP, the map is modified, and then anther inertia run is performed to verify the validity of the changes. Maybe the varification runs are not usually performed, and that's why no one has any to show. I also wonder how the software is set up to function during realtime "autotuning" via step tests. Can you save the data from the beginning of the test in order to compare to the result, or do you have to perform one set of step tests beforehand? So maybe the before runs would cost more money? If so, I wonder how much? They should at least be able to provide the after runs for all throttle positions. I can't imagine the data taking up too much space on a drive. Even so, they could erase it after printing.

Anyway, when one of my buds get his new Duc, I am going to try and get it tuning linked at a friends HD shop. If so, I'll be standing right next to it while it is done, then we will no for sure what really happens.

Except that everyone might not do it the same.

Recently, dynojet has added some new features to their controls to let operators simulate riding - that is, going up hills etc.  What this means is that a mechanic can take a 'test ride' on the dyno when there is 3 feet of snow on the ground to help sort out problems with a bike.

The ability to perform controlled sweep tests (at however many ft/sec/sec you want) has been a standard feature on Factory Pro eddy current dynos for at least 7 years now. However, sorting out problems on a dyno is a lot easier with an EGA (which doesn't really work during sweep testing). Say you have a bike that craps out after freeway riding for a while and you want to diagnose the problem. If you get the problem to occur on the dyno and it's fuel starvation related, you'll see the CO fall and fall until it quits. If it's ignition related, HC would go way up as it was crapping out. Or if you had a misfire, HC would go up at the same time as O2 and flux.

Previously the operator could only monkey around with the eddy current brake when in tuning link mode.......

I can't imagine that you couldn't manually tune during step tests on a DJ dyno. Why not?

[moto]

I said, "I am concerned about airflow contamination from one cylinder to the next via the crossover.

I would rather lose a little power if it helped the bike run better."

This was regarding my apparently mistaken belief that crossovers created asymetrical flow between the two cylinders on V twin engine.

That's not really a mistaken belief, because whether or not an assymetrical flow is created depends on the length, placement, and diameter of the crossover.

If my belief was correct, then I would rather have had a balanced engine where each cylinder made 40HP, rather than a more powerful engine where one made 39HP and the other 43HP.

Sometimes this is done deliberately in order to broaden the powerband.

Everyone seemed to agree that crossovers and two into ones, are good things and that two into twos produce less power.

Agreed in general terms, but on a vee twin, secondary pipe diameter is critical with two into ones due to the uneven firing.