Microsquirt (was MS3) Questions for use in Two-Stroke

[PVO22]

I've done a lot of reading the last few days and think the MS3 may be the answer for converting my single cylinder 500cc two-stroke engine to fuel injection. I've got some specific questions that I haven't found any answers to yet that I would like to ask.

1. I'm thinking about separating the oil from the fuel in the injection process. Normally the engine is premixed but I think I could get better control out of it if I injected oil and fuel. Is there a way I could run an electronic oil pump and have its output correspond to RPM? Basically I'd like to be able to map the oil flow along with fuel. Could I get an oil pump to do this by connecting it in some way to a spare injection or ignition channel? Or is there another way that would be better to make this work?

2. I have no need to operate a power valve at the moment but in the future I may wish to so would it be possible to set up a servo motor to operate the powervalve and have its operation based on RPM?

3. I've not read enough about the MS3 yet but is it possible to tune off of an EGT sensor? I'm not entirely sure if a wideband O2 sensor will read entirely accurate since there is always more air in the exhaust than on a four stroke.

Those are the main questions remaining in my mind at the moment. I've got a pretty good idea how the other components of the system will work. Thanks for your time and help.

[billr]

Although I am no MS expert, I'm pretty certain the answer to all your questions is an enthusiastic "Yes!". You will find that MS has extensive capabilities and unlimited support and experience on this forum. You may need to build some fairly simple electronic circuits for custom output drivers or input signal-conditioning, but even the design for those will probably be provided here (if needed). The biggest "problem" you will have getting into MS is that so much capability, info, and help is available that you will initially be overwhelmed! Just take it "one bite at a time"... and enjoy.

[dontz125]

Hi, there! Always glad to see another rider in this sea of cages ... What bike are you squirting?

1. Given the small amount of oil under discussion, you may be borrowing trouble. Premix is used because it works. That said, for a street bike it can be a major pain in the butt. The MS3 does include a number of "generic PWM" outputs that could be used to control a high-speed valve. I would suspect that, rather than control an injector directly, you'd be better off controlling a valve (like a water-injection valve) feeding a fixed-orifice nozzle. See this link for the generic PWMs http://www.msextra.com/doc/ms3/genericpwms.html, and this link for a description of a water-injection rig http://www.msextra.com/doc/ms1extra/MS_Extra_Hardware_Manual.htm#waterinj - yes, it's MS1, but it's the best I could find! You may want to consider an actual 2-stroke oil injection pump with throttle cable from something like the RZ/RD series or similar.

2. What sort of servo are you talking about? Most of the older PVs used 'dumb' 5-wire servos, which were basically a 2-wire DC motor (+/- to open, -/+ to close) with a TPS-style pot close-coupled to the servo output shaft. Unfortunately, the code doesn't currently support this sort of beast, but I'm trying to bash together a control board that would take a PWM signal (from another of the generic outputs) and use that signal to drive the PV servo.

3. I'd just go with the EGO sensor. The problem with EGO and 2-strokes is the oil (and sometimes the lead, depending on the fuel) contaminating the sensor, but until it fails you'll get good reliable readings, and you should still be talking hundreds of hours. There have been a number of people, here and over at the RZ/RD500 site, who've added EFI of various brands to various bikes, and most have used WBO2 sensors. Note that the reading doesn't have to be ACCURATE so much as it has to be REPEATABLE.

[PVO22]

Hi, there! Always glad to see another rider in this sea of cages ... What bike are you squirting?

The engine is a Kawasaki KX500 dirt bike engine however the bike I've made from the frame up for road racing.

1. Given the small amount of oil under discussion, you may be borrowing trouble. Premix is used because it works. That said, for a street bike it can be a major pain in the butt. The MS3 does include a number of "generic PWM" outputs that could be used to control a high-speed valve. I would suspect that, rather than control an injector directly, you'd be better off controlling a valve (like a water-injection valve) feeding a fixed-orifice nozzle. See this link for the generic PWMs http://www.msextra.com/doc/ms3/genericpwms.html, and this link for a description of a water-injection rig http://www.msextra.com/doc/ms1extra/MS_Extra_Hardware_Manual.htm#waterinj - yes, it's MS1, but it's the best I could find! You may want to consider an actual 2-stroke oil injection pump with throttle cable from something like the RZ/RD series or similar.

You may be right there but I want to explore the possibility. How do the injectors handle the premix? That would be my only concern injecting premix. Thanks for the info on the water injection that may be doable for oil as well. I'll have to look into it more. I've thought about adding a mechanical pump but that would require a lot of work to get it to run off the crank gear or any gear for that matter. I figured by running it off the ECU I could get more control over the oil output as well.

2. What sort of servo are you talking about? Most of the older PVs used 'dumb' 5-wire servos, which were basically a 2-wire DC motor (+/- to open, -/+ to close) with a TPS-style pot close-coupled to the servo output shaft. Unfortunately, the code doesn't currently support this sort of beast, but I'm trying to bash together a control board that would take a PWM signal (from another of the generic outputs) and use that signal to drive the PV servo.

Honestly I don't know. This project is kind of a stepping stone on to bigger and better things so I'm already looking ahead and exploring the capabilities of MS. I'm working on designing an engine and some of the features would be direct injection, electronic oil pump, and some sort of power valve system so if MS could do all those things figuring out how to control it would be pretty easy. Would MS be able to control an electromagnetic valve?

3. I'd just go with the EGO sensor. The problem with EGO and 2-strokes is the oil (and sometimes the lead, depending on the fuel) contaminating the sensor, but until it fails you'll get good reliable readings, and you should still be talking hundreds of hours. There have been a number of people, here and over at the RZ/RD500 site, who've added EFI of various brands to various bikes, and most have used WBO2 sensors. Note that the reading doesn't have to be ACCURATE so much as it has to be REPEATABLE.

Great, I'll be running a leaded fuel as well. Hopefully the sensor will last awhile. Do you happen to have any links to the RZ/RD500 conversions?

I have a few more questions as well now.

4. If I've read correctly I should be able to go with the direct ignition coil control option for my application using a single spark plug. If I wanted to modify the head and run dual plugs I could still go direct coil control and run the other coil off the 2nd ignition output. I've never been electronic savy but I'm trying to learn quickly so some of my questions may seem a bit silly like the following. Can I run any old coil in this setup or are there specific specs that the coil should be rated for? I see from the diagram that it is a 12V coil and the driver limits it to 7amps so is that all I need to shop for or am I missing something? Or does the amperage not matter since the driver limits it anyway?

5. I've read about MegaView but I don't think it would work all that great as a display for a bike. Is there anyway I could take some of the info from the ECU like RPM, speed, egt, O2, water, temp, etc and get them to input into a more bike friendly display? For example I have a Digitron display that I'll use while running on carbs and it uses many of the same sensors so could I just run wires from the ECU sesnors to the Digitron to get my info?

6. In order to convert the KX500 so that I can control ignition (and have a decent charging system) I think the easiest thing to do will be to take an alternator from another bike and graft it onto the 500 along with a trigger wheel. I've done a bit of searching and haven't found any specs on the power requirements for the ECU or the actual size of an MS3 for that matter. I'm guessing it just needs 12 volts and it will work? As far as the alternator output goes what would be the best way to determine the amount of power I need it to produce? I think most bike alternators are 12 volt 35 watt so I'm guessing that would suffice? Would it be possible to take an alternator from a snowmobile which I believe produce a bit more power and use that and eliminate the need for a battery? Or probably more realistic I could just plug a battery in upon start up and then remove it. After the alternator I would have to wire in a voltage regulator which would then go to the ECU, correct? So far all I feel EFI is adding is a bunch of weight and complexity! Haha, all kidding aside I think the added weight will be worth the benefit as I should be able to get some decent power increases out of the motor with the progamable igntion not to mention I will never have to worry about jetting again!

[billr]

Did you have a typo, meaning 350W and not 35W? I changed out the stator on my son's GSXR750 a while back, and my recollection is it was 360W... At any rate, the MS itself doesn't draw much, but all the actuators, displays, injectors, sensors, etc. do have to be powered and can add up. I wouldn't try to get by without a battery, it helps to smooth out AC ripple from the alternator, which might affect some things. Also, it will help to clamp any voltage spikes if any large loads are suddenly turned off. I don't know about your voltage regulation, but the GSXR is a remarkably crude system. Lastly, a battery will allow some grace period before the engine craps out, in event of alternator/regulator failure. Well worth the size and weight trade-off, I think. The MS ECU unit is approximately 2.5" x 4" x 6.25", but you better figure to have another couple of inches at each end (i.e. the 6.25" turns into 9-10") for connecting cables and in/out of the SD card.

[PVO22]

Did you have a typo, meaning 350W and not 35W?

Yes it was. I found out today that the 08 Yamaha R6's stator puts out 420W and most of the other sport bikes fall between 300-450W so I have an idea now of how much they're making. As for what I'll be consuming I'm guessing it will be far less since I won't be running any lights, just the bare essentials really so I suspect a sport bike stator will work for me. I'll be looking at some other stators though as I would like something with a little smaller flywheel if possible.

[dontz125]

If you're going road-racing then I really would just stick with premix - I used an AccuMix jug for many years with my TZ125, and it's the matter of a minute to quickly and accurately make a fresh batch. The fuel injectors really won't care - just a SWAG, but I'd almost think they'd give longer life with less chance of corrosion and fouling between races.

Electromagnetic valve - do you mean a solenoid valve with PWM positioning? Yes, absolutely - that's one of the two common 'car' idle valves.

Spark control - the MS can certainly control spark. With the dual plugs, will they be fired together, or separately? If together, you're still talking a single coil driver and control channel for both coils / plugs. If you want one to fire after the other (2 deg? 5 deg?), then you would need a second driver, and you would set it up as an "odd-fire" V-twin (with a REALLY small included angle ... )

Gauges - This is a racebike, not a Wing-abego touring barge! RPM and water temp is all you need to worry about watching while screaming down the back straight; leave the rest for datalogging. There are a few options for that, too.

Power - there are race cars and bikes (some even running EFI! ) that run total-loss off a battery; LiFePO4 batteries are wonderful things. All things being equal, a single injector and single coil running sequential will have 1/4 the power draw of a comparable 4-cyl. The pump is a significant adder, but there's a fellow named Madhu (sorry - can't remember his user name) who has been doing great work on PWM control of the pump vs rail pressure, and has seen HUGE reductions in pump power consumption. By starting the bike while hooked up to a pit battery, the pre-heat of the WBO2 is removed as a drain. Or, as you say, you could graft on an alternator. Modern 3-phase voltage regulator / rectifiers are quite stable and reliable; note that you'd still need a small battery for start-up (to power the MS and prime the system) even if you decided to push-start the bike - it IS a 2-stroke! Also, non-battery alternator-regulator systems can be tricky to design; if you lay it out like a typical bike or car and just omit the battery, you'll fry your regulator and then everything else downstream.

I realise you stated this thread asking about the MS3, but depending on your wish list it really sounds like you could use an MS2 or MicroSquirt.

Last question, and this one's a biggy - What does your crank trigger pattern look like? Bike OEMs like to use long teeth, sometimes 30-60 crank angle degrees in span. The standard MS triggers have significant issues with anything much past 5 degrees; there has been a significant success by one fellow on a specific trigger wheel design, but this still remains as a potential show-stopper. Will you be using a VR or Hall sensor?

[billr]

While you are in there grafting on (to the crank) a rotor for an alternator, make sure it includes a nice trigger wheel.

[dontz125]

That's cheating!

[PVO22]

If you're going road-racing then I really would just stick with premix - I used an AccuMix jug for many years with my TZ125, and it's the matter of a minute to quickly and accurately make a fresh batch. The fuel injectors really won't care - just a SWAG, but I'd almost think they'd give longer life with less chance of corrosion and fouling between races.

I think I may end up using premix for the first version of the system. Once up and running I would like to try using a solenoid pump to control the oiling system using the PWM feature of MS. Along with this I'd like to try port injection too.

Electromagnetic valve - do you mean a solenoid valve with PWM positioning? Yes, absolutely - that's one of the two common 'car' idle valves.

Yes, thanks.

Spark control - the MS can certainly control spark. With the dual plugs, will they be fired together, or separately? If together, you're still talking a single coil driver and control channel for both coils / plugs. If you want one to fire after the other (2 deg? 5 deg?), then you would need a second driver, and you would set it up as an "odd-fire" V-twin (with a REALLY small included angle ... )

I'm thinking together initially however I have thought about staggering them later on. From what I've gathered with the coils I wont need an ignition driver correct? The coil can hook up directly to the ECU? What type of coil do I need to be shopping for? Most bike coils are CDI type so I would guess that these won't work? I would say an inductive type coil would be what I need but I'm still not really sure?

Gauges - This is a racebike, not a Wing-abego touring barge! RPM and water temp is all you need to worry about watching while screaming down the back straight; leave the rest for datalogging. There are a few options for that, too.

True, I'm curious as to why you're taking water temp over say an EGT gauge? Is there a tuning/engine monitoring method based off water temps that I should know about?

Power - there are race cars and bikes (some even running EFI! ) that run total-loss off a battery; LiFePO4 batteries are wonderful things. All things being equal, a single injector and single coil running sequential will have 1/4 the power draw of a comparable 4-cyl. The pump is a significant adder, but there's a fellow named Madhu (sorry - can't remember his user name) who has been doing great work on PWM control of the pump vs rail pressure, and has seen HUGE reductions in pump power consumption. By starting the bike while hooked up to a pit battery, the pre-heat of the WBO2 is removed as a drain. Or, as you say, you could graft on an alternator. Modern 3-phase voltage regulator / rectifiers are quite stable and reliable; note that you'd still need a small battery for start-up (to power the MS and prime the system) even if you decided to push-start the bike - it IS a 2-stroke! Also, non-battery alternator-regulator systems can be tricky to design; if you lay it out like a typical bike or car and just omit the battery, you'll fry your regulator and then everything else downstream.

I'm thinking the best option for me will be to go with the alternator and small battery combo.

I realise you stated this thread asking about the MS3, but depending on your wish list it really sounds like you could use an MS2 or MicroSquirt.

That may be true. If I could control fuel, ignition, and at least one PWM output with a MicroSquirt I may consider it which after reading about it I think I could. I initially chose MS3 as it has the most features and tunability. I'm still in the "what's best/what do I need/what's most feasible stage" so I haven't made any concrete decisions yet.

Last question, and this one's a biggy - What does your crank trigger pattern look like? Bike OEMs like to use long teeth, sometimes 30-60 crank angle degrees in span. The standard MS triggers have significant issues with anything much past 5 degrees; there has been a significant success by one fellow on a specific trigger wheel design, but this still remains as a potential show-stopper. Will you be using a VR or Hall sensor?

I think I'll be going with a VR trigger and adding the trigger wheel to the flywheel.

[dontz125]

I'm thinking together initially however I have thought about staggering them later on. From what I've gathered with the coils I wont need an ignition driver correct? The coil can hook up directly to the ECU? What type of coil do I need to be shopping for? Most bike coils are CDI type so I would guess that these won't work? I would say an inductive type coil would be what I need but I'm still not really sure?

Most 2-strokes and off-road 4-strokes are CDI. Most road bikes are TCI. Mixing and matching is a fast way to damage to damage the coils and other components. Both types of coils need some type of 'driver'; some of the newer car TCI COP and CNP designs have built-in drivers and only need a 5v logic signal from the ECU. A TCI coil needs a high-energy transistor, whether an IGBT or a Darlington. The current MS "best practice" is the BIP373 3-stage Darlington available from DIYAutotune; it's practically indestructible, and will shut down if the dwell settings threaten to fry it (and your coil!). CDI obviously needs a charge/discharge circuit; if your alternator is typical low-volt (30-60vac), you'll also need an inverter to generate the 250+ DC volts that the firing capacitor needs.

True, I'm curious as to why you're taking water temp over say an EGT gauge? Is there a tuning/engine monitoring method based off water temps that I should know about?

For a two-stroke race bike, tach and water temp are the traditional gauges. EGT is definitely a tuning datum, which you don't care about in the middle of a race - leave it for the logger. Water temp is important for engine life, especially in a stroker. Run it too hot, and you lose power and then seize. Hammer it too hard before you're up to temperature, and you seize. ("Cold" seizes - gotta love 'em!)

If I could control fuel, ignition, and at least one PWM output with a MicroSquirt I may consider it which after reading about it I think I could. I initially chose MS3 as it has the most features and tunability. I'm still in the "what's best/what do I need/what's most feasible stage" so I haven't made any concrete decisions yet.

Fair enough. One thing to consider is the size of the ECU - the MS3 is BIG, while the MicroSquirt is quite small. One point to ponder (tying back to the ignition concerns) is that the latest version of the uS has deleted the BIP coil drivers, and is 'logic' only. I'm in the process of starting my own business, and one of my products will be a 2-channel coil driver to address this.

I think I'll be going with a VR trigger and adding the trigger wheel to the flywheel.

Good plan; just make sure the trigger pattern is something the MS can recognise.

[PVO22]

Also, non-battery alternator-regulator systems can be tricky to design; if you lay it out like a typical bike or car and just omit the battery, you'll fry your regulator and then everything else downstream.

I don't intend to go this route but I'm curious as to what would have to be done differently to make it work without a battery?

[dontz125]

The problem with running an alternator with 99% of the systems out there is that the harness is set up so that the regulator measures the voltage across the battery; as the voltage across the terminals fluctuates, so too does the regulator.

The problem comes when you remove the battery, or disconnect the ground or positive cable. There is now NO voltage across the terminals, and the regulator goes WFO and stays there. This will quickly slag the VR and / or alternator field winding (acvtive / variable field alternator) or the VRR and / or alternator stator winding(s) (permanent magnet alternator).

The VR for a battery-less system has to be designed with some sort of limiter to prevent this run-away self-destruction. I THINK that a small resistor (10k, 1/4W) wired in parallel across the +ve and -ve terminals would suffice - but I haven't worked up the courage to try it yet!

[bojane-1]

"I'm thinking together initially however I have thought about staggering them later on. From what I've gathered with the coils I wont need an ignition driver correct? The coil can hook up directly to the ECU? What type of coil do I need to be shopping for? Most bike coils are CDI type so I would guess that these won't work? I would say an inductive type coil would be what I need but I'm still not really sure?"

That is how I run my 3 cyl 2-stroke. However it seems like my inductive coils have problems with keeping my sparkplugs alive. When I ran MSD capacitive ignitions I never fouled a plug.

http://www.diyautotune.com/catalog/ign1 ... p-394.html

-Anders

[dontz125]

Anders - Most 2-strokes are CDI; the sharp voltage rise helps burn through fouling due to mis-jetting or over-oiling. A TCI coil SHOULD be able to run a 2-stroke, but you may need to use a hotter plug. You may simply be running rich, especially at idle - what mix ratio are you running, and is the oil injected or pre-mix? The RZ350s, with their throttle-cable-operated oil injector pump, varied from (IIRC) about 30-1 at WFO to 100-1 at idle.

[bojane-1]

dontz125

premix 3-3.5%
fuel-map is far from optimal, would probaly burn much less plugs with better fuelmap.

I'm thinking of convert the output to capacitive ignition.

http://www.msdignition.com/Products/Ign ... utput.aspx

Sorry for the sidestep from the thread, I'll start a new one

-Anders

[PVO22]

The VR for a battery-less system has to be designed with some sort of limiter to prevent this run-away self-destruction. I THINK that a small resistor (10k, 1/4W) wired in parallel across the +ve and -ve terminals would suffice - but I haven't worked up the courage to try it yet!

So would a simple solution be to install a VR from a system that already doesn't use a battery? For example use the VR off a snowmobile or ATV that has a batteryless electrical system?

[billr]

You still are going to have the problem of how to start it. With no electrical power before you try kick or push starting, the MS will have to go through whatever power-on-self-test that it does, and all the sensor inputs will have to stabilize, and then sync established, and then fuel pressure, ignition, and injector pulses established. I doubt all that could be done during the one or two strokes of a kick-start; a push start or external battery jump will probably be required. Both can be kind of "unhandy" if it kills out on the road or track...

[dontz125]

You still are going to have the problem of how to start it.

Agreed.

So would a simple solution be to install a VR from a system that already doesn't use a battery? For example use the VR off a snowmobile or ATV that has a batteryless electrical system?

Most batteryless 2-stroke dirt bikes and sleds are AC-only, with two coils in the stator. The charging / firing coil for the CDI puts out 250-300vac, while the lighting coil is closer to 30vac. I'm really not sure how the VR works, but there is no rectifier on most of these - even the dash lamps are AC.

The more I think about it, the more I'm convinced to retract any comments on a batteryless system and just recommend that you go with a battery, whether total-loss, semi-loss, or full-charge. With no lights, fan, or horn, and only a single injector and spark (maybe two plugs), your power budget is going to be pretty small compared to a 4-cyl streetbike. If a current R6 superbike can run total-loss, charging / swapping every other session, you could probably get away with one charge for a track day, recharging overnight. Look into the current LiFePO4 batteries - they are immensely cheaper, easier, and safer vs the older Li-ion designs, and a fraction of the size and weight of a lead-acid battery.

[PVO22]

The more I think about it, the more I'm convinced to retract any comments on a batteryless system and just recommend that you go with a battery, whether total-loss, semi-loss, or full-charge.

Like I said before, I do intend to run a battery. I don't have a lot of experience with electrical stuff and I've been enjoying learning about it so I thought I'd ask. Thanks for the explanation!