Jump to content

The MyECU thread


raz

Recommended Posts

  • Replies 416
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

I work around the original guzzi's files found inside the eprom from a Quota,the same like other guzzi model...

Maybe somebody found inspirations or mistakes!!!

Your spark advance table has "InjAdv" in it's headings instead of "SpkAdv", that's just a typo of course.

 

I'm not too sure of your injection advance table. Earlier I said I assume that InjAdv is 'injection advance in degrees before TDC of combustion stroke' just like ignition advance (IgnAdv). Now thinking about it a bit more, that cannot be quite correct as the normal figure is 525 all over the map. Maybe it's a figure correlating to the enclosed graph, with higher value meaning later injection. In that case, 525 seems a bit late and maybe it could be lowered towards 360 but not much more. The advantage of having it a bit high (above some 400) might be to effectively avoid losing fuel in the valve overlap. All this are just layman guesses, I admit having little real knowledge or experience.

 

If my assumptions are correct I think your InjAdv numbers ought to be bad. At 300 (and especially as low as 45), you are injecting fuel while intake valves are closed!

 

Then again, if you can say it did good after some trial and error, it is very interesting.

 

V11_cam.jpg

V11 cam graph, thanks to motoguzznix

Link to comment
Share on other sites

In that case, 525 seems a bit late and maybe it could be lowered towards 360 but not much more. The advantage of having it a bit high (above some 400) might be to effectively avoid losing fuel in the valve overlap. All this are just layman guesses, I admit having little real knowledge or experience.

I found an article at http://www.springerlink.com/content/6b8yk17e2xdjhu7p/ with the following in the abstract (testing at stoich in a four valve engine):

The indicated mean-effective cylinder pressure was found to be constant with initiation of injection from 150° to 630° of crank angle after top-dead-centre of intake and with a 10% reduction between 30° and 60° which coincided with maxima in the covariance in pressure and in the emissions of unburned hydrocarbon. There was also a tendency for performance to decline with injection after 660°.

 

Measurements with laser- and phase-Doppler velocimeters showed that the number of droplets entering the cylinder was much reduced with injection at crank angles corresponding to closed inlet valves due to evaporation, and that the few large droplets which emerged did not survive until top-dead-centre of compression.

 

In contrast, some of the many droplets associated with injection with the valves open survived to the crank angle of ignition and it is likely that these led to an inhomogeneous charge with poorer flame-front propagation responsible for reduction in performance.

This might confirm 525° is a sane figure, and 300° is not too bad either (but I still think valve overlap may promote using a higher number). But 45° should be too low.

 

The article mentions "initiation of injection ... crank angle after top-dead-centre of intake" though. If I read that right it is still not what I thought, and I'm a little confused. Or is that the same as TDC of combustion stroke? :wacko:

 

Some other article I found are playing with "phasing of the end of injection" but otherwise like the valve graph I posted above. We need Cliff to know what we are talking about here!

Link to comment
Share on other sites

I,m a beginner in engine injection an tecniques,for me is better electronics! ;)

I send the raw tables,not adapted to Cliff's ECU'S.....this comes directly from a tuning program in a friend workshop.

He tell me is normal in certain conditions to inject the fuel when the inlet valve is closed,the pressure,2 Bars, of the fuel injected create a bad turbolence inside the cylinder chamber.....i don't know if this is true or not,he goes many time to mandello for training stages,maybe know somethings secret! :lol:

Maybe i make a mistake in my translation of the InjAdv table,the original work with positive and negative degrees,the Cliff's table not.

For translate i have to cut a row because the throttle breaks are 16 X 16 in the oginal map,in Cliff's map 15 x 16.

I hope to help somebody!!

P.S the tuning program show the other correction parameters for pressure and temperature to InjDur and InjAdv!!

 

guzzistd_INJ.pdfguzzistd_INJADV.pdf

Link to comment
Share on other sites

TDC Intake is the beginning of the intake event, our exhaust is also open so we all commonly refer to this as overlap. Our 525 should be the number of crank degrees after that event that the injector begins its injection duration. At 525 after TDC intake( or TDC exhaust, your call) we are just beginning our exhaust stroke. This is when most FI systems inject. Further reading of the above article explains, read the last few paragraphs.

 

http://resources.metapress.com/pdf-preview...mp;size=largest

Link to comment
Share on other sites

....guzzistd_SPKADV.pdf.....

 

That's what I'd call conservative ;) This map won't ping even with lighter fuel.

 

The idea of varying the injection advance, which Ivan brought up, is a nice one. I actually see a very stable consumption of 5.4 l/100 km, and regardless what spark advance or lambda targets I dial in - somehow the front wheel stays solidly grounded :( So playing around with injection advance will definitely be the number one topic next spring.

 

Hubert

Link to comment
Share on other sites

TDC Intake is the beginning of the intake event, our exhaust is also open so we all commonly refer to this as overlap. Our 525 should be the number of crank degrees after that event that the injector begins its injection duration. At 525 after TDC intake( or TDC exhaust, your call) we are just beginning our exhaust stroke. This is when most FI systems inject. Further reading of the above article explains, read the last few paragraphs.

Thanks, so we can safely assume that the 525 is to be interpreted in this way. Edit: this is wrong, see below. Hopefully Cliff will confirm that too, I sent him an email.

So using 525, we are injecting at 165° in Motoguzznix' graph. I had no idea there were advantages of injecting before intake valve opens. At higher throttle it's needed of course, to have enough time to complete the injection pulse. But at cruising I wouldn't have guessed it.

 

The idea of varying the injection advance, which Ivan brought up, is a nice one. I actually see a very stable consumption of 5.4 l/100 km, and regardless what spark advance or lambda targets I dial in - somehow the front wheel stays solidly grounded :( So playing around with injection advance will definitely be the number one topic next spring.

Agreed, thanks Ivan. Nice new thing to experiment with. My vague idea of dynoing (if I ever get around to it) is more and more leaning to this approach:

  1. Make a WOT sweep for reference, and because everyone else does :rolleyes:
  2. Choose a midrange cell (like half throttle at 4000 rpm) and have the dyno operator lock in at that speed.
  3. Vary the pulsewidth until best power. Note the O2.
  4. Set the O2 target to this value and go closed loop for steps 5 & 6:
  5. Tune injection advance for best power
  6. Tune ignition advance for best power
  7. Recheck if our O2 target is still the best
  8. Out of money, session ending with just one cell tuned :lol:

Hopefully something will be learned. If there are significant gains, I'll save up for another run and try tuning much larger portions of the map, concentrating on what seems most bang for the bucks.

Link to comment
Share on other sites

InjAdv is degrees before TDC that the Inj pulse ENDS.

 

This setting doesnt have a great deal of effect. This is why earlier MyECU just had a single entry rather than a table.

 

300 is just before the inlet opens and is the logical place to inject fuel. I found the bike felt crisper.

600 is well before and roughly where the OEM injects. This will give a few % better economy than the 300 as it allows more time for the fuel to vapourise.

Link to comment
Share on other sites

InjAdv is degrees before TDC that the Inj pulse ENDS.

...snip

...

300 is just before the inlet opens and is the logical place to inject fuel.

...snip

 

Just for getting a better understanding: in this case the last sentence could be: "...is the logical place where the fuel is completely injected" ?

 

Not to correct you, I just ask for better understanding, foreign language and so :)

 

And for the protocol: is it possible to enter values above 720 deg?

 

Hubert

Link to comment
Share on other sites

InjAdv is degrees before TDC that the Inj pulse ENDS.

 

This setting doesnt have a great deal of effect. This is why earlier MyECU just had a single entry rather than a table.

 

300 is just before the inlet opens and is the logical place to inject fuel. I found the bike felt crisper.

600 is well before and roughly where the OEM injects. This will give a few % better economy than the 300 as it allows more time for the fuel to vapourise.

Thanks for clearing this up. That's quite different from what we thought but 525 happens to end up in almost the same spot in Motoguzznix' table, at 195 degrees. And this ENDS thing will often make quite a difference of course.

 

This is interesting :nerd: even if there aren't huge amounts of hidden horses. I think I need to calculate some graphs visualizing timing. Like "milliseconds per degree rotation" and "ms per revolution", for various rpms.

 

It's a pity we can't tune the InjAdv while running. As far as I remember it's neither in the Optimizer or the ECUControl program, right? Testing how idle reacts to various values would be very interesting.

 

I want my own brake dyno for christmas. And a 5-gas. OK, Santa?

 

:xmas:

Link to comment
Share on other sites

The InjAdv can be tuned from the Optimiser. In the Injector menu, I think button 2 or 3 allows you to cycle between duration, advance and o2 target

That is great to know, I didn't even know I could change the O2 target that way :rolleyes:

 

I think I need to calculate some graphs visualizing timing. Like "milliseconds per degree rotation" and "ms per revolution", for various rpms.

So I did a little spreadsheet m@sturbation. These are all calculated at 525° ignition advance:

  • At 1000 rpm idle, I have a pulsewidth of about 4100 µs including deadtime. While injecting, the engine will rotate 25°. Thus the injector will be open between to 170° and 195° in Motoguzznix' chart.
  • At 4000 rpm and about half-open throttle, the pulsewidth is 11500 µs. Injection will take place during 276° of rotation, starting at 639° in the chart (just when the intake valve closes on the previous stroke) and continuing up to 195°.
  • At WOT and close to the rev limiter at 8000 rpm, the pulsewidth is 12300 µs, or 590° of rotation. The injector opens at 325° (about when the intake opens on the previous cycle) and stays open until 195°.

This is very interesting but I don't really know what to make out of it :P . The last case is pretty interesting. No, all three are.

 

The injector dead time will be significant at higher rpms. I have it set at 800 µs at 13V which is more or less a guess. At idle, this corresponds to just 5° but at 8000 rpm it takes 38° out of the 590°, scewing the above a little.

 

I'd think the InjAdv should benefit from being lowered towards 360 at high rpms. Just from gut feeling. But maybe it doesn't matter much. And maybe I have no idea really.

Link to comment
Share on other sites

He wrote "...where injection ends". So injection always ends at e.g. 520° = 200° b.t.d.c., no matter how long before it has started.

I'd expect bigger advantages with variations at lower revolutions, for instance when the engine gets started => 50°? Then, what values are used while cranking? First column?

 

Hubert

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...