direct injection on a desmodue?

Started by ducatiz, May 16, 2011, 09:55:00 AM

Previous topic - Next topic

ducatiz

gearheads, talk to me.

DI is a clear winner in the fuel delivery category.
Check out my oil filter forensics thread!                     Offended? Click here
"Yelling out of cars, turning your speakers out the window to blast your music onto the street, setting off M-80 firecrackers, firing automatic weapons into the airâ€"these are all well and good. But none of them create a merry atmosphere of insouciance and bonhomie quite like a revving motorcycle.

Artful

Sincerely doubt the existing system could even remotely handle the pressures involved...
Every time I meet a new group of your friends that understand you and your weird sense of humor I'm a little more amazed that there are other people in the world like you that lived through childhood - My loving girlfriend

ducatiz

Quote from: Artful on May 16, 2011, 10:00:48 AM
Sincerely doubt the existing system could even remotely handle the pressures involved...

the cylinders?  sure they could.  of course the EFI system wont, but no reason the valves and cylinders couldn't
Check out my oil filter forensics thread!                     Offended? Click here
"Yelling out of cars, turning your speakers out the window to blast your music onto the street, setting off M-80 firecrackers, firing automatic weapons into the airâ€"these are all well and good. But none of them create a merry atmosphere of insouciance and bonhomie quite like a revving motorcycle.

zooom

stop trying to apply automotive technology in a motorcycle....what the hell do you think your Duc is, a VW or a Honda or something?!?!?!
99 Cagiva Gran Canyon-"FOR SALE", PM for details.
98 Monster 900(trackpregnant dog-soon to be made my Fiancee's upgrade streetbike)
2010 KTM 990 SM-T

ducatiz

I'm just thinking...

If you can drill a hole for a second plug, then why not an injector?
Check out my oil filter forensics thread!                     Offended? Click here
"Yelling out of cars, turning your speakers out the window to blast your music onto the street, setting off M-80 firecrackers, firing automatic weapons into the airâ€"these are all well and good. But none of them create a merry atmosphere of insouciance and bonhomie quite like a revving motorcycle.

Dirty Duc

The only reason why not (aftermarket) DI is the development cost. 

With throttle body or port injection the timing of the injector pulse is not terribly important.  You can just throw the appropriate amount of fuel in the hole (at least once per combustion cycle) and it will go in the cylinder at the right time (when the valve opens).  (There are efficiencies available by getting the timing correct.)

With DI the timing is very important to avoid undesirable combustion events (detonation and/or compression ignition).  DI did not become economically feasible for gasoline engines until some massive improvement was mandated by increasing emissions and fuel mileage requirements.

At this time, to my somewhat limited knowledge, there are no programmable standalone ECUs that will run DI.  If you just want to control fuel and spark in some way not allowed by the plug and play options, http://www.microsquirt.info/

That site and the associated megasquirt site will tell you more than you ever thought you needed to know about fuel injection. 

If I were to attempt it, I would probably try to adapt a mechanical injection pump from a two-cylnder diesel (as found in chinese generators).  (Note: I am not a professional mechanic, just a guy with too much time on his hands)

bikepilot

DI would be awesome!  Its a bit expensive to implement, but is getting more common all the time.  To me one of the biggest advantages is being able to have huge compression without preignition (see e.g., Hyundai Sonnata with 11.3:1 or 9.5:1 and a bunch of boost).  I'd guesstimate that the 1100 desmo would happily burble along at 11.5-12:1 with direct injection without troubles on regular fuel.

The biggest annoyance with a mechanical injection pump is you can't easily vary the injection timing via software.  Also, at least some if not most or all diesel injection pumps rely partially on diesel fuel for lubrication.  I know the Standyne pump in my old Ford relies mostly on diesel fuel for lubrication and now that the EPA has yanked the slippery stuff from the diesel its pump will die very prematurely if I don't run additive for it.  Short of it is to pump gas you'd probably need to run two-stroke premix or find a pump that has its own lubrication source.  I know some folks who had a small gas four stroke running direct injection and compression ignition with gasoline at a very high compression ratio (and extremely lean like a diesel).  Last I heard it was far from operational, but was sorta promising in terms of hp and efficiency.


2009 XB12XT
2006 Monster 620 (wife's)
1997 TL1000S
1975 Kawasaki H1 Mach III
2001 CR250R (CO do-it-all bike)
2000 XR650R (dez racer)
2003 KX100 (wife's)
1994 DR250SE (wife's/my city commuter)

ducatiz

Seems like you could upgrade the timing and use a CRD-like setup.  If I understand the issue correctly, the problem would be ensuring the injector timing is right.  Anyone?
Check out my oil filter forensics thread!                     Offended? Click here
"Yelling out of cars, turning your speakers out the window to blast your music onto the street, setting off M-80 firecrackers, firing automatic weapons into the airâ€"these are all well and good. But none of them create a merry atmosphere of insouciance and bonhomie quite like a revving motorcycle.

Dirty Duc

#8
Nothing wrong with a little two-stroke oil... some of the good stuff even comes with soda pop mixed in!  [evil]

Agreed that the lack of programmability is a drawback, but... [shiny object] Hilborn, anyone?

All right, so the trouble is not really one of practicality, it is one of expense... find someone to reverse engineer a miller-cycle ecu, and adapt the programming to two cylinders... 15:1 compression... E85...

Fuel pump
http://thmotorsports.com/apr_tuned/apr_tuned_fuel_pump/ms100017/i-222610.aspx?googlebase=MS100017-08-2007

Fuel injector
http://compare.ebay.com/like/330424271714?var=bingv&ltyp=AllFixedPriceItemTypes&var=sbar&rvr_id=232826754631&clk_rvr_id=232826754631&crlp=1_263602_325002&UA=WVF%3F&GUID=633ff20d12a0a0e203128722fe72c762&itemid=330424271714&ff4=263602_325002

Some custom fuel line

and a Bosch HPI 1.1
http://www.bosch-motorsport.de/content/language2/html/3676.htm (which should also be able to do the ignition).

Some way of driving the fuel pump. (Add a external lobes to the camshaft drive gear and fab a mount?)

So that's an easy $3k in parts.

Plus tuning and install.  It's do-able. :)

chris1044

#9
Quote from: ducatiz on May 16, 2011, 02:29:25 PM
Seems like you could upgrade the timing and use a CRD-like setup.  If I understand the issue correctly, the problem would be ensuring the injector timing is right.  Anyone?

The big issue with why it wouldn't work (without a lot, and I mean a lot, of tuning time) is cylinder/combustion chamber design (and cams - cam timing plays a vital role as well).

In all direct injection engines, Diesel or Otto/Carnot cycle, the piston dome, ring land are, cylinder head, and even the intake ports are all very carefully designed to get the correct flame front propagation to occur in the cylinder during a combustion event.  Thousands of hours worth of simulation work, and then a ton more of dyno/vehicle work is done. 

Would it be sweet?  Yes.  But for it to be a true direct injection engine, the fuel would have to spray directly into the cylinder - not the intake runner.  I'm not a combustion engineer, but I can't imagine this working very well with the standard Ducati pistons.  Additionally, that's an extra hole that's got to be drilled into the cylinder head/cylinder to allow the injector to pass through.  I doubt the configuration of the engine (IE intake/exhaust port location, ring land depth, etc.) would work well with this in stock form, so plan on copious amounts of grinding/custom piston modeling as well.


Can it be done? Sure.  Anything can be done with infinite $$$ and time.  Practical?  No.  Really, power mods to our bikes aren't practical anyhow.  However, I'd venture to say Ducati's working on this as we type this...it's going to become far more prevalent on OEM automobiles over the next 5 years or so, which means it's only time before it trickles down into bikes.

If someone has a spare "junker" (aka old 900??) engine lying around, you could rig something up to work...HP pump off an auto engine would likely be too heavy and need more torque to run properly at low RPM's (dunno, just saying). 

bikepilot

Chris, that's for direct injection, compression ignition motors right?  Wouldn't a direct injection spark ignition motor (like virtually all new cages and many sleds use) be considerably simpler? Also, JE will make a set of pistons to spec for a pretty reasonable cost. 

The Motus motorcycle is direct injection  [thumbsup]

I know a bunch of two stroke sleds and skis are direct injection, I wonder if any of the four strokes have gone that route yet?  Many if not most cages are now directly injected (pretty much everything from Germany and Korea anyway and almost all of the smaller stuff from the U.S. and a hodge-podge from japan).
2009 XB12XT
2006 Monster 620 (wife's)
1997 TL1000S
1975 Kawasaki H1 Mach III
2001 CR250R (CO do-it-all bike)
2000 XR650R (dez racer)
2003 KX100 (wife's)
1994 DR250SE (wife's/my city commuter)

Dirty Duc

I think for proof of concept the special head design isn't necessary.  The key to the lean burn is to spray the atomized fuel at the spark plug just before the spark event (and probably just after TDC)... which can be done if you replace one spark plug in a DS head with a high pressure injector (I think).

To answer the injection pump concern: the cam-driven pump from the VAG 2.0 shouldn't take too much torque. 

I still see the actual problem not so much as a lack of parts or design, but a lack of a flexible enough ECU for a reasonable price (and a budget for R&D).

ducatiz

Quote from: mergatroyd on May 17, 2011, 08:44:51 AM
I still see the actual problem not so much as a lack of parts or design, but a lack of a flexible enough ECU for a reasonable price (and a budget for R&D).

.. for a one-off maybe.  But if this was part of the design by the factory, there is no reason it would be significantly more expensive.  Is there?
Check out my oil filter forensics thread!                     Offended? Click here
"Yelling out of cars, turning your speakers out the window to blast your music onto the street, setting off M-80 firecrackers, firing automatic weapons into the airâ€"these are all well and good. But none of them create a merry atmosphere of insouciance and bonhomie quite like a revving motorcycle.

Dirty Duc

I shouldn't think so.

See bikepilot's ref to MOTUS.

I can think of no reason not to do it except for the natural conservatism in the motorcycle community.  Maybe a cost-benefit analysis hasn't been done, or maybe it was done and it isn't worth the trouble until regulations and the market demand it.

ducatiz

Quote from: mergatroyd on May 17, 2011, 08:57:20 AM
I shouldn't think so.

See bikepilot's ref to MOTUS.

isn't the motus liquid cooled?  i wonder if it's been done on an air/oil cooled engine or if that presents an issue?
Check out my oil filter forensics thread!                     Offended? Click here
"Yelling out of cars, turning your speakers out the window to blast your music onto the street, setting off M-80 firecrackers, firing automatic weapons into the airâ€"these are all well and good. But none of them create a merry atmosphere of insouciance and bonhomie quite like a revving motorcycle.