248 replies to this topic

[Maribo TT Newbie]

Thanks, Mate.  Good info, all. It might or might not be worth doing.  That's what I'm trying to figure out.

[msiddalingaiah TT Silver Member]

HeadTrauma said:

If power consumption is a large enough concern, look for a smaller pump

That's the problem. There really aren't many small pumps out there. Most of the in-tank motorcycle pumps require about 4-5 amps continuous. The LT-R450 and Honda Rincon pumps require less than 2 amps, but as I say, they are not as widely available at reasonable cost. There are several late model CBR 600 RR fuel pumps on eBay now for less than $25 including shipping. I'm almost tempted to pick one up as a spare.

[Mbirt TT Member]

Maribo said:

Thanks, Mate.  Good info, all. It might or might not be worth doing.  That's what I'm trying to figure out.

http://www.diyautotu...-v22-p-381.html

It can be cheap AND good if you're willing to take on a challenge.  As for the easy, yet compromised route of retrofitting a KTM EFI setup, you're best off comparing part numbers for your engine and the donor engine to see which components from which subsystems can be swapped over.

[drillergeek TT Newbie]

Captain Midnight said:

I can build you one. Do you want to drive it off the cam or the crank?

I'm upgrading a KLR 250 and would like to convert to fuel injection.  My system output is only 140 watts. In this application what are the advantages and the drawbacks to consider when mechanically driving a fuel pump from either location.  I realize the cams turn at half the speed of the crank.

Can a mechanical fuel pump be operated manually to precharge the system prior to a starting attempt?

Thanks
Todd

Edited by drillergeek, 02 February 2011 - 07:53 AM.
incorrect stator output

[dmcca TT Silver Member]

Captain Midnight said:

It should adapt quite well. Although, for the cost, you could go buy another motorcycle. Ebay parts are very cheap. Here is a throttle body with injector, fuel rail, regulator, and throttle position sensor. All for $40.00. Worked very well.

what was this throttle body off??

After reading this thread ive caught the bug... Im working on EFI'ing my CR250R two stroke! Going to use an Ignijet ecu as it can also control the power valve servo as well as fuel and spark, getting the stator rewound to give enough power will be an issue but not impossible, fuel pump is easy, sensors are easy, probably the biggest question i have is regarding the throttle body...

I have heard that the pulses within a 2 stroke throttle body will disrupt the map sensor... I read the RZ350 site where a guy used 2 Constant Velocity throttle bodies off an R6 to set up his version of 2 stroke EFI... he reckoned that the CV slide helped smooth out the pulses and make the map sensor operate properly... the issue is that i need a single thriotlle body with tps and injector attached... the only CV ones are in banks of 3 or 4 and are not well sealed against dirt and water. The most practical units to use are ATV ones... they are well sealed, right size and have tps, injector, etc all in one body.

Captain Midnight... do you know is a non CV throttle body will work on a 2 stroke???

[msiddalingaiah TT Silver Member]

You can easily split multi-bore throttle bodies and use them with the slide if you like. I used one of the four bores from a YZF-R6. I removed the slide and ran alpha-N, so MAP pulses don't matter. The preferred technique is to select the lowest MAP value between two crank revolutions (or one cam rev) on a four stroke. It's easier on a two stroke, just select the lowest MAP value over one crank revolution. MS2 Extra now has support for this.

I decided to build my own circuit do this and so far it seems to work. Take a look at post #368 on page 37:

http://www.thumperta...=825575&page=37

I have yet to switch back to speed-density and use the smoothed out MAP value, but it's in the works!

Let me know if you are interested in the MAP filtering circuit.

[dmcca TT Silver Member]

msiddalingaiah said:

You can easily split multi-bore throttle bodies and use them with the slide if you like. I used one of the four bores from a YZF-R6. I removed the slide and ran alpha-N, so MAP pulses don't matter. The preferred technique is to select the lowest MAP value between two crank revolutions (or one cam rev) on a four stroke. It's easier on a two stroke, just select the lowest MAP value over one crank revolution. MS2 Extra now has support for this.

I decided to build my own circuit do this and so far it seems to work. Take a look at post #368 on page 37:

http://www.thumperta...=825575&page=37

I have yet to switch back to speed-density and use the smoothed out MAP value, but it's in the works!

Let me know if you are interested in the MAP filtering circuit.

Thanks for the info... I am still a little confused with the map issue, been doing lots of reading on the megasquirt site but obviously need to do a lot more to understand it properly. So basically I need a map sensor initially to determine the lowest pressure of one cycle then save that number and basically run alpha-n, which won't need the map input to operate?? Or if I use your unit I can just set it up to run from map signals the whole time??

I am also very interested in your fuel controller... I will probably only get 75w out of a stator rewind so need to save as much power as possible... Do you offer the complete kit wired ready to go including Bosch sensor and circuit board??? I can easily source the pump but I'd be interested to hear if you can supply the rest?

The issue I have with the throttle body is the usually the banked TB's have tps on one and throttle cable on another and I need to find a single one with cable and tps on it... I also need an injector around 300cc/min and most of the street bike TBs seem to be around 250. I still wonder if a TB off a big single atv will be better, something like a raptor 700 or ltr450 should provide what I need.

[HeadTrauma Get Help Now]

dmcca said:

1.) Thanks for the info... I am still a little confused with the map issue, been doing lots of reading on the megasquirt site but obviously need to do a lot more to understand it properly. So basically I need a map sensor initially to determine the lowest pressure of one cycle then save that number and basically run alpha-n, which won't need the map input to operate??


2.) The issue I have with the throttle body is the usually the banked TB's have tps on one and throttle cable on another and I need to find a single one with cable and tps on it... I also need an injector around 300cc/min and most of the street bike TBs seem to be around 250. I still wonder if a TB off a big single atv will be better, something like a raptor 700 or ltr450 should provide what I need.

1.) The Idea is that MS-II will sample MAP pressure at its lowest point in every engine cycle. From what I understand, this is a technique employed by OEMs when injecting thumpers. The older code and older systems like MS-I sample constantly and try to account for the high and low peaks, leading to low rpm problems with spark advance and accel/decel enrichment. Hybrid alpha-N and toothed trigger wheels are a way to get around most of the problems. Straight alpha-N doesn't use MAP at all...except maybe to get an atmospheric pressure reading to use for altitude correction.

2.) If you shop around, you should be able to find a variety of injector sizes for a given TB type, especially if you're willing to buy them new. Some Honda cars use the same injectors as the bikes, only bigger. Barring that, other squirters can be adapted. With some modification, I put an early '80s Ford/Bosch 370cc injector in a Suzuki SV650 TB. There are better options, though, as the Ford injector is huge.

[dmcca TT Silver Member]

HeadTrauma said:

1.) The Idea is that MS-II will sample MAP pressure at its lowest point in every engine cycle. From what I understand, this is a technique employed by OEMs when injecting thumpers. The older code and older systems like MS-I sample constantly and try to account for the high and low peaks, leading to low rpm problems with spark advance and accel/decel enrichment. Hybrid alpha-N and toothed trigger wheels are a way to get around most of the problems. Straight alpha-N doesn't use MAP at all...except maybe to get an atmospheric pressure reading to use for altitude correction.

2.) If you shop around, you should be able to find a variety of injector sizes for a given TB type, especially if you're willing to buy them new. Some Honda cars use the same injectors as the bikes, only bigger. Barring that, other squirters can be adapted. With some modification, I put an early '80s Ford/Bosch 370cc injector in a Suzuki SV650 TB. There are better options, though, as the Ford injector is huge.

Thanks, I've done a lot more reading today and it's getting much clearer, looks like I just need to set up the map sensor as normal and work out the rest when it's actually on the bike... It says in the megamanual that it all makes much more sense when you set it up and start tuning so that's what I'll do

I've found a good TB also... Off a KFX450 atv... 42mm diameter with tps attached, sealed cable, and a 330cc injector standard. Coupled with an electronic Fuel pressure regulator, wideband O2 and a microsquirt it shouldn't be too hard... in theory anyway.

[sprint4425 TT Silver Member]

i'll bite...

If i have a stock MX bike that came from the factory fuel injected,  how do i change/edit fuel maps on my bike to increase performance?

I understand the 2010 yz450 has a programming tool available for ECU and fuel maps, but lets say i dont want to use that and/or that tool is not available for my bike.

[msiddalingaiah TT Silver Member]

dmcca said:

I am also very interested in your fuel controller...

PM sent.

sprint4425 said:

If i have a stock MX bike that came from the factory fuel injected,  how do i change/edit fuel maps on my bike to increase performance?

I understand the 2010 yz450 has a programming tool available for ECU and fuel maps, but lets say i dont want to use that and/or that tool is not available for my bike.

I don't know much about the YZ450 or the tools available for it. What you could do is replace the OEM ECU with Megasquirt/Microsquirt and use the existing fuel system (TB, injector, fuel pump, maybe some sensors). There are plenty of good tools for Megasquirt. You can log all your data and tune your bike the way you want it.

[HeadTrauma Get Help Now]

HeadTrauma said:

1.) The Idea is that MS-II will sample MAP pressure at its lowest point in every engine cycle. From what I understand, this is a technique employed by OEMs when injecting thumpers. The older code and older systems like MS-I sample constantly and try to account for the high and low peaks, leading to low rpm problems with spark advance and accel/decel enrichment.

I stumbled on this old datalog of mine that illustrates the MAP fluctuation with a single-cylinder engine. This was from '07; I can't remember if I already had a canister in the MAP line to smooth out the signal or not.



What's interesting is that the baro pressure was 86kPa, yet the log shows a maximum of 99kPa when the engine was running.

[msiddalingaiah TT Silver Member]

HeadTrauma said:

What's interesting is that the baro pressure was 86kPa, yet the log shows a maximum of 99kPa when the engine was running.

Yes, I have noticed this also. From what I am told, there is probably a slight valve overlap between the exhaust and intake stroke. Some of the exhaust gasses don't leave the engine and show up as a brief spike in MAP. It also affects your VE table in strange ways. Even though MAP is high, VE entries at that RPM/MAP bin will be low(er) because the cylinder contains exhaust products that do not participate in combustion. In the end, the engine wants less fuel because it has less oxygen than MAP is telling you.

[grayracer513]

msiddalingaiah said:

Yes, I have noticed this also. From what I am told, there is probably a slight valve overlap between the exhaust and intake stroke. Some of the exhaust gasses don't leave the engine and show up as a brief spike in MAP. It also affects your VE table in strange ways. Even though MAP is high, VE entries at that RPM/MAP bin will be low(er) because the cylinder contains exhaust products that do not participate in combustion. In the end, the engine wants less fuel because it has less oxygen than MAP is telling you.

This is why the MAP sensor needs to be located well upstream from the intake port whenever possible.

[msiddalingaiah TT Silver Member]

grayracer513 said:

This is why the MAP sensor needs to be located well upstream from the intake port whenever possible.

You don't really have much choice here. MAP has to be sensed between the throttle plate and the intake valve. Most bike throttle bodies have MAP sensor ports very close to the throttle plate. Even if you have a 100' hose from that port back to your MAP sensor, you will get the same reading. People have actually tested that and determined the length of hose from the TB to the MAP sensor makes little (if any) difference. It makes sense since the MAP sensor does not allow any air to flow through it. It's really measuring sound waves, which move at the speed of sound.

You can read more about that here:

http://www.megamanua...l/mwire.htm#map

No matter how you measure pressure in the cylinder, it's hard to tell how much of that pressure is due to fresh air and how much is due to exhaust gasses. Mass Air Flow (MAF) sensors are one solution, but that's one more sensor to install, test, tune etc; not to mention restrictions to air flow.

[grayracer513]

You know what, I just got busted for multi-tasking again.

MAF sensors took the place of MAP sensors when I read that.  You're right, obviously. Manifold Air Pressure is that between the intake valve and the inside of the throttle.  On most motorcycles, that's a tube less than 6" long.

[dmcca TT Silver Member]

Need some advice on stator output... Thinking of efi'ing my cr250 using one of msiddalingaiah's fuel pump controllers (total draw 2.5amps, or possibly less if I can source an LTR450 pump), yet it's looking like I will struggle to get a stator output of much more than 50w... If I also use a small battery do you think I'll have enough to run a microsquirt for fuel only?? Will there be anything left for charging the battery at higher rpm?? For those with experience, what is the current draw from the rest of the system?? Remembering the cr has no lights and it generally run at high rpm.

Thanks Dave.

[msiddalingaiah TT Silver Member]

Here's a simple power estimate:

Pump - 2.5 Amps max
Injector - 1 Amp max (assuming high impedance injector)
MS - 100 mA max

At 12 Volts, that's about 43 Watts max. At idle, the current is lower for both the pump and injector, which is the majority of the power consumption. If you don't have any other power requirements, I think you should be able to do it.

Have you tested your stator output? Connect a 50 W headlamp and see if the voltage drops below charging voltage at idle.

[roger young TT Newbie]

I have a 2007 Honda Rancher 2 wheel drive 420 electric shift. 40 psi pressure from the fuel pump. The fuel injector is not opening when you crank the engine over.
I have put 12 volts directly to the injector and its has a strong spray. I checked the voltage at the connector unpluged and it is 12 volts with the key on. When you Plug it
in the voltage drop off to tenths of a volt and is insignificant when cranking. The resistance through the injector is 12.5 ohms and draws about .9 amps when connected to  12 volt power. The bike will run when you spray starting fluid in to the air cleaner but not stay running. Can you tell me how to proceed to troubleshoot this problem.
Thanks
Roger

[36MotoMarc TT Titanium Member]

It should have 12v key-on regardless, the ECM switches it to ground to pulse it.  I'd look at the fuse and upline connectors for corrosion.  You could patch it 12v from a solid supply to confirm that's the problem, it should run.  If it doesn't, it's probably a problem between the injector and ECM, or the ECM itself.

Are you getting a flashing fault code?