Blake's Megasquirt thread..

[Coil Coyle]

While in Swap Shop considering CV carbs on EBay, Blake brought up MegaSquirt. I have moved his thought here to close the Swap thread.

"megasquirt works on a 12x12 map, which you personally program. giving you 144 different options for fuel requirements.

The sensors megasquirt uses to determine amount of air/fuel mixture:  Intake Air temp, Coolant temp, Throttle position sensor, and Map sensor.  The map and TPS are the main things, being as it is seeing the amount of throttle you have it open, and also the vacuum demand of the engine.  The temp sensors help to accurately adjust the amount of fuel to be sprayed out of the injectors (for the cold start the amount of fuel is something like 200% at 32degrees(coolant/air)  compared to the 100% mixture ratio needed at normal operating temperature.   Basically this system requires some fiddling around at first to get the amount of fuel needed in the right area, but from there you can adjust it ever so slightly to get the optimum performance.

One big thing that I really liked is the new version (3.0 board) has the ability to run a closed loop system using a WIDE BAND oxygen meter (http://wbo2.com/ is the main one used-I have the 2b0 but I see they just came out with a stripped down version intended for the megasquirt for much less!).

Oh.. also.. which is what I?ve been very happy about lately.  megasquirt also has the ability to control ignition from many means (optical, magnetic, etc).  and that also uses a map to control what you want the advance to be.. (not what you're interested in now, but just thought id throw it out.  I'm a horrible salesman I know  )

The system can fire the injectors either together or in banks (1 and 1 on the vision, or  4 and 4 on a v-8, etc).  you tell it the lb/hour of the injector and it will tell you the pulse width (time the injector is spraying) needed, and you work from there. 
-  It also can control an idle air control motor (new version has ability for step motor that opens in increments based on coolant temp, whereas old motors/version had simple full on/ full off switch."

? ? I've been thinking that the R1 system fires in two banks of two, so it should be possible to fire two banks of one each. It seems that the engineering is nearly done (R1 triggers into V case) and twenty years more modern than the V.

? ? Several times this Megasquirt was brought up and I checked their site. My impression was that they had a tremendously versatile system that required a great deal of "engineering" of componants just to get it in the Ballpark. There was also no feedback from the exhaust mix in the past.

? ? ?So, here we go Blake, a new Megasquirt system has feedback and idle control. Everyone interested can go to Megasquirt's page and we can thrash the idea around.

;) coil

[Coil Coyle]

Blake, when considering injector placement in the tract remember. There must be travel distance and turbulence for the fuel and air to mix for consistant burn.

$00,000.02   ;)       coil

[Tdub]

The Buell uses a very simple closed loop EFI system. Similar cylinder configuration and single throttle body using a single manifold. Might be worth checking out.

[Coil Coyle]

Tdub,
         If you built a single throttle body intake, would you "Y" the intake tubes or run the throttle body into a manifold with an inside airbox and separate tube intakes  like NASCAR 4bbl V8's do?

          In your R&D on the dyno have you ever run a single carb and manifold on the V?

???  coil

[Tdub]

coil, I think for the street, by going to a twin throttle body/carb setup with twin manifolds would allow you to use smaller carbs/throttle bodies. I think this would give better throttle response and streetability than a single setup with one large carb/throttle body. I dont think you can compare the setup on a V8. There are several schools of thought here(bear with me here).? Let's compare a single to a twin as an example with one carb. Every time the intake closes, flow essentially stops or interupts. The more cylinders, the less time the flow is interupted making for a smoother intake flow thru the single carb. There so now I contradicted my original statement LOL
let me think on this some more...I am not very adept at conveying what is going thru my head LOL Tdub

[Blake]

T-dub,


If i read you right..  your trying to explain the intake pulses.. with individual intakes/carbs, you have a go/stop flow working (suck in for so long, then suddenly halt everything, where it will reverb back up to the carb)

with multiple intakes/single carb, theres always some place for the mixture to go.  while one valve is open, another is closed, etc. 


is that right?   


I was talking to coil about reading about the ram air stuff using the intake pulses.  And for our powerband (2nd reverb between 9-10k)  i came up with right around 13" intake tracts for our intake (using formulas for car engines)  now if it would work like a cars intake, you could theoretically make a common plenum of .5L and  attach it to the two, with a single carb, and it should work the same.   but then again...thats what i learned about car intakes :-\   


What i was thinking was a couple of 40mm throttle bodies (up from stock 34) with the injectors right near the intake port mounted on a standard bosch bung (for throttle response).  i figure the 40's will allow more airflow at WOT, but arent too incredibly big (the new literbikes are using quad 40mm's on theirs(250cc cylinders), plus with the injectors right next to the intake you wont have to worry about lag like with a carb signal..

And you're pretty much right Coil, about the work having to get it initially in the ballpark.  My train of thought is that i'd rather spend a few extra days trying to get it to work initially than to use a system from another bike, and find out it wont meet our fuel requirements.   plus i seem to have extra time on my hands and I like making things work that originally werent meant to be.


Blake

[Coil Coyle]

Tdub,
Yeah, i know what you mean. ? ???? ? Thats without thinking of the fuel dropping out of suspension in an interior air box. ? :-\

I think that the NASCAR manifold is used in spite of fuel suspension because individual tuned port length is most important.

So that ( accepting bad fuel mix to get tuned port length )puts us back at 1 intake per cylinder with 1 fuel mixer. I'm trying the twinned R1 carbs on the "copper in the oil filter" 83 I'm in the middle of rebuilding.

I will probably try 1 40mm CV on a Y manifold for Lori's Standard Vision, just for drivability and economy.

[Coil Coyle]

Blake, when considering injector placement in the tract remember. There must be travel distance and turbulence for the fuel and air to mix for consistant burn.

$00,000.02? ?;)? ? ? ?coil

[Lucky]

Keep the YICS & it should help smooth out the pulses..

[Tdub]

Yamaha has used a version of the YICS since back in the '70s on the 2 strokes. Even some of the aftermarket companies came up with a version.

[Extent]

I heard that someone at SVRider created a YICS like device for their bike, I don't remember details though.

And something else I remember reading somewhere (either in the civic PGMEFI forums or in a motorcycle FI book) was that especially in motorcycle applications, when your injectors get closer to saturation in high load situations at high RPM they spend most of their time spraying against a closed valve.  At 10k RPM one rev comes in about 2.4 hundreths of a second, and only about half of that is with the intake valves open.

I've got more thoughts on the matter but that'll have to wait till I have more time to think about it. :p

[Coil Coyle]

Extent,
Re.: "when your injectors get closer to saturation in high load situations at high RPM they spend most of their time spraying against a closed valve."

The injectors should be spraying into a volume of air that has pressure waves and vortices richocheting chaotically. ;)
Similar to thoughts in my mind ::)

coil

[Blake]

when your injectors get closer to saturation in high load situations at high RPM they spend most of their time spraying against a closed valve.  At 10k RPM one rev comes in about 2.4 hundreths of a second, and only about half of that is with the intake valves open.

I can't believe i forgot to mention that too.. that was one explanation that came up quite frequently when people asked about progressive injection (is that the right term?  where injectors fire at their respective cylinder when the valve is open).  when you size the injectors for your specific system and determine how long the injector has to be open for that revolution (so many miliseconds), it is a lot longer than the amount of time an intake valve is open. 


Coil..

sorry i forgot to respond to the injector location.  The reason i decided as close to the intake as possible is because of looking at automotive multi-port injection locations.  Some new systems even have the injector directly mounted in the cylinder head, shooting directly into the cylinder.  But everything i've found has said (and seen) that the injector should be as close as possible to the intake port as possible.  With the high quality injectors now making quite a nice fog, and the ideally very high velocity air moving past, the period that the injector will be open will mix quite well with the air moving by, and the injectors own force of the fuel (from a few sources i read of the net...may or may not be true..but makes sense to me)..but the foce of the fuel coming out of the injector itself helps to create a turbulance to atomize the fuel even more.   But most places i have seen have said that the ideal place for the injector is to try to have the injector spray as directly down the center of the intake port as possible (however, not possible, but can be as close as can be).


wow..i even lost myself in there..    ;D

Cliff notes version: It's true distance helps to further all type of atomization(to a point), but i figure with the much closer to a "fog" injectors can "instantly" make soon out of the injector body, and with the help of the force of the fuel coming out of it (helping to create turbulance in the intake tract), the injectors can be as close to the intake port as possible and still be quite effective.  Reasoning?  just the conclusion i've come to from reading from lots of places and looking at other mp-efi systems.


Just curious coil,  how far do you think the injectors should be?  I figure the carbs (where the fuel enters) are about...2-3" from the intake?




Just throwing around ideas until i get back to school and can get cracking on this baby.  (need to make an engine stand so i can keep the engine on my desk so i can figure this all out in the warm :)


Blake

[Blake]

Here's a link that comes up with answers that are pretty close to what i was coming up with as for intake runner legnth:

http://www.bgsoflex.com/intakeln.html

bgsoflex is one of the guys that made the megasquirt


put in 13" into the calculator.. it'll show you 2nd 3rd and 4th harmonics (sonic pulses)

I was thinking of using that because you get the most powerful harmonic between 9-10k, then again at 6.7k-7.7k, then again at 5.2-5.9k.  basically where i spend most of my time depending on the style of riding for that road (drag(9-10), highway(6-8), cruising back roads with power to go through twisties by just rolling on (5-6) (About 50mph or so basically)

ALTHOUGH.. i have read a few times (i forget where..prob hot rod magazine.  haha) that the ideal time to open the valve is on the 3rd harmonic.  so in that case.. if you put in 10" into the calculator, the 3rd harmonic comes in from 8.8k to 10k. 

hmm..


I'm definately overthinking this  ;D

[Blake]

And one more i just found:


http://www.wallaceracing.com/runnertorquecalc.php


intake runner size (flow area in square inches) and also does length. vs peak RPM and engine size (33 or 34cid).   34 and 33 comes up with 9.33" long runner for a 9000 peak torque (road test data says peak is at 8500)

34" engine size (2 cylinders) says area should be 1.73".  (a 1-5/8" diameter pipe with .060" walls gives you a flow area of 1.77".  1.5" with .060" walls gives 1.4957" of flow)

[Coil Coyle]

Blake,
as to where the injectors go, where were they on the bodies you started with? I have not found a picture of yours yet.

It can be said that...

Honda puts 'em here http://cgi.ebay.com/ebaymotors/2001-Honda-CBR-600-F4i-throttle-body_W0QQitemZ4599908684QQcategoryZ10066QQssPageNameZWDVWQQrdZ1QQcmdZViewItem#ebayphotohosting
and then shoots another high speed injector down the intake horn from the air box for WFO.
http://cgi.ebay.com/ebaymotors/HONDA-CBR-600-RR-600RR-THROTTLE-BODIES-W-AIR-BOX-FUEL_W0QQitemZ4600081284QQcategoryZ35597QQssPageNameZWDVWQQrdZ1QQcmdZViewItem#ebayphotohosting
Triumph puts 'em here http://cgi.ebay.com/ebaymotors/TRIUMPH-955-DAYTONA-FUEL-INJECTION-THROTTLE-BODY-1999_W0QQitemZ4599982951QQcategoryZ35597QQssPageNameZWDVWQQrdZ1QQcmdZViewItem#ebayphotohosting

Yamaha puts 'em here http://cgi.ebay.com/ebaymotors/2003-2004-Yamaha-R6-throttle-bodies-body_W0QQitemZ4599484737QQcategoryZ35597QQssPageNameZWDVWQQrdZ1QQcmdZViewItem#ebayphotohosting

Kawasaki puts 'em here, and adds another injector at the inlet horn for WFO
http://cgi.ebay.com/ebaymotors/KAWASAKI-ZX-10-ZX10-THROTTLE-BODIES-FUEL-INJECTION-2005_W0QQitemZ4600081598QQcategoryZ35597QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

I'm going to guess that they all found the turbulence behind the part open throttle butterfly helped with mixture consistancy. ???
;)
coil

[Coil Coyle]

Here's a link that comes up with answers that are pretty close to what i was coming up with as for intake runner legnth:

http://www.bgsoflex.com/intakeln.html

bgsoflex is one of the guys that made the megasquirt


put in 13" into the calculator.. it'll show you 2nd 3rd and 4th harmonics (sonic pulses)

I was thinking of using that because you get the most powerful harmonic between 9-10k, then again at 6.7k-7.7k, then again at 5.2-5.9k.? basically where i spend most of my time depending on the style of riding for that road (drag(9-10), highway(6-8), cruising back roads with power to go through twisties by just rolling on (5-6) (About 50mph or so basically)

ALTHOUGH.. i have read a few times (i forget where..prob hot rod magazine.? haha) that the ideal time to open the valve is on the 3rd harmonic.? so in that case.. if you put in 10" into the calculator, the 3rd harmonic comes in from 8.8k to 10k.?

hmm..


I'm definately overthinking this? ;D

Blake, are they are calculating the returning positive wave from the first vacuum pulse?
Their formula does not consider your cam timing. Maximum cylinder filling occurs when the positive pulse returns slightly ahead of the intake closing. The formula needs cam timing and air temperature (speed of sound varies) to be useful.

$00,000.02? ? ?;)? ?coil

PS. Wallace is not using cam duration either, and cylinder count should not affect intake dimensions ???.
? ? ? Sequential injection fires like a spark plug, at at particular point that is unique for each cylinder. (different times, same piston and crank position of the cylinder being fired)
? ? ? Progressive injection uses the little injector behind the throttle plate for low flow and another injector that gives high flow in the same intake. See Honda and Kawaski above.

[Blake]

Hey coil,


I try and find the formula i used for the intake runner lengths/sizes.  I orignally did it for an intake for my cadillac engine (500cid) and i remember asking a guy who builds the engines and he said it "should be pretty good"..  not sure if i used cam timing or not.   I'll deifnately let you know.

Sequential injection!  thats what i was thinking  ;D  (really need to quite writing posts so late. my brain doesnt seem to work so well after 11)



Blake

[Blake]

Coil,

As for the throttle bodies..  Initially i was looking at them from twin cylinder bikes (tl-1000, sv650, ducatis) and happened to like the design of the ducatis because i figure i can easily modify the linkages on them to do what i want them to.



(picture is from a ducati performance web site-these are 54mm with the dual injectors)

But you see in the ducati bodies, the injector is behind the throttle plate (like the ones you showed me too).  But what gets me thinking, is how they all put it as close as possible to the end of the throttle body (closer to the intake). (my idea here...)  On production bikes (especially new ones) you want it to go together as easy as possible, have as as few of parts (assembly costs) as possible.  So by having the entire throttle body/injectors in one easy to install assembly, itll save cost.  Plus, there isnt much extra room in those modern sport bikes (especially with having to comply with emissions laws.  But if you think about it (at least how im looking at it) (look in the first link you sent with the honda bodies.) they mount the injectors as close as possible on the throttle body, and the only distance between them and the intake, is that of the rubber boot which is needed to hold everything onto the engine.  That distance may be what.. 1-3" max?  trying to visualize how id have to run the intake tracts in my head, thats probabaly the distance from the head i'd have to have it from the intake flange anyway.  The whole thing with what i wanted to try was the ram theory, where i place the throttle body further out (longer intake runner), but still keep the injector close to the intake to eliminate the lag.



Also.  here's the link that initially got me looking into the ram theory
http://www.chrysler300club.com/uniq/allaboutrams/ramtheory.htm

you are correct about the cam timing.  im not sure what timing i used when i figure out the intake length for the vision a while ago (simply because i dont know of it being available).  i probably used some other bikes duration (like a sv650) ??? 

Thanks for throwing around ideas and beating with me on this.  It's defiantely helping me to think about different aspects of this.  Helps to have someone else to give me some criticism.


Blake

[Extent]

Well I've done a bit of picking around trying to find out more about injector placement, and info is pretty slim, but I pulled a couple of nuggets.  Injecting before the TB isn't a problem (as I had originally thought) as long as the intake tracts are relatively straight.  This works great at high RPMs because it gives the fuel more time to evaporate fully, but it dosen't make for smooth low RPM performance.  Apparently several Superbike engines use two sets of injectors, one close and one far, that they switch between based on RPM (and Ferrari used sliding injectors in one of their F1 engines)

13" seems unreasonably long to fit on the bike IMO for runner length.  Why not use a tuned air box to make up for runner length?  It may not hit as hard as these runners are supposed to, but it seems much more doable within the space available.