Help with EFI project

[450exc115]

Hi all,

I’d like a little help from my friends here in the off road motorcycle community. I started my masters in system engineering and my first class is about project management. I thought it would be cool to do a project on generating a business study on the need for company based on the new technology of fuel injection. If the project goes well (including feedback from you all) I may actually start the business. I’ve generated some questions for some market research. Please answer them as best as you can.

1)How many years have you been off-road motorcycle riding? How would you describe your skill level (recreational, novice, amateur, expert, or pro)?

2)Are you aware that there is an upcoming change in technology from carburetion to electronic fuel injection?

3)Do you see yourself purchasing an off-road motorcycle with EFI? How much does / did your ability to tune the EFI bike play into this decision? Are you excited for the possibilities EFI may hold in performance

4)How much would you be willing to pay for a custom EFI tune:[/font]

• Down load custom maps on line? (25, 50, 75, 100, 150?)
  
• At a store/shop with a dynamometer? (50, 75, 100, 150, 200?)
  
• Track side, allowing for the rider to have the bike tuned for the conditions? ((50, 75, 100, 150, 200?)
  

5)Which of the three tuning methods would suit your preferences best?

6)How often would you buy a custom EFI tune? 1-5, 5-10, 10-20, 20+ a year?

Thanks to all that participate!

Tom

Note to Admin, I email you a little bit ago to see if this was OK but haven't gotten a response so hopefully this is OK..

[Smacaroni]

1) 2

2) yes

3) eventually (perhaps 20 years!), not at all, not much

4) $0 (this has serious home brew potential, however, the manufacturers will make it extremely difficult just like cars), $100, unlikely

5) first one, then second

6) hopefully once, perhaps once a year

Good luck with your project. It has lots of potential. A friend of mine does stuff like this with his car. However, the amount of work involved to re-engineer what the manufacturers could have done but didn't is tremendous.

[450exc115]

Come on guys.... Give me a little feedback.

[tenis24]

1) about 40. Amateur

2) Yes

3). Yes, Not having to tune as much with EFI for the average rider is what I would want.

4) $100 For a one time deal.

5) online of course.

Dyna Jet has little franchises all over the country that sell their Power Commanders that can than take different downloads.

[450exc115]

I've heard of power commanders but I thought they were indepent programers that currently are selling to the ATV and street bike crowd. I'll look into what they do!

More feed back please.......

[Smacaroni]

I'm sending this thread to the friend of mine who does home brew ECM and such, see if he's got any input.

I really think the money is in home brew, not necessarily in selling fuel/timing maps, but in selling equipment which allows easy user downloads of maps. Probably USB or bluetooth.

[Dr Honda]

1) riding for 30 years, use to race SCORE, and local SoCal MX.

2) Up coming? it's been around since the 50s, but on dirt bikes... Yes.

3) Yes, a lot, yes

4) depends on what I paid for the system. (I'll add a note to this)

5) None

6)Depends on bike usage. (note on this)

OK... this is a very cool thing... and since I'm a small business owner... I like to tell anyone thinking of doing their own thing to GO FOR IT!!!

Now... a few words on your Idea from some one who has their own line of product.

The idea of a bolt on, aftermarket EFI is great... and it is needed for the off road comunity... but you have a few flaws that need to be addressed. First of which is the "Custom Tunes." In the dirt bike world, you have racers, and riders. The racer will want a way to tune their own system... so they will not be willing to pay for custom tunes. They will want a base map specific to their bike... probably one that will emulate the stock jetting... and then a hand held device that will let them do there own maps. (just as an FYI Yamaha has a cheap handheld programmer for the new YZ450, so you will have OEM competition)

A "rider" will want one tune EVER, and they will not want to pay much for it. They will want to buy a kit... bolt it on, and have it work in all conditions. That means you will have to make it a closed loop system that can adapt to temp, and altitude. Also... this group isn't going to be willing to pay more for a EFI system than they would for a replacement carb, or exhaust system. I know it's more complex, but if you have a retail price point at $900 the recreational guy isn't going to buy it if their bike is running well... that is, unless you can give them a 15% boost in HP. Also, in turn, the racer may not buy it either... they may just buy a new bike with factory tunable EFI.

Now... pricing a custom map isn't like selling a product. To do it right, you will have to go to a dyno, or have someone who knows how to tune do the map for the bike. That guy is going to want to make money for his time. If some one has to give you $100 for the map... and $100 more for the 3 dyno pulls... no one will buy it. so what you should think about is you need to set up dealers who will buy a tuning kit (one time purchase... like any tool) and then they can tune for their normal labor price.

Last thing to think about... Product liability. If you have a system where the end user has total control over the fuel... and they smoke the engine... it may come back on you. That's why a system like Micro-squirt works. They sell a box and all the info... but not a "Bolt on" kit. so they have no liability for problems with the engine, or people getting hurt.

I have a background in Physics, and I started down the path of an EE. I have built chip programmers, and reverse engineered products for the last company I worked for. Also, I was part of the original group who dissected the OEM Honda code at the machine level to be able to program the OBD1 ECU's. Right now I own a small shop that builds custom parts. If you need throttle bodies machined, or advice on the circuts... shoot me a PM.

Edited October 27, 2009 by Dr Honda

[William1]

1 - 40 years, retired expert

2 - Yes, for about 30+ years

3 - Of course, none at all as EFI is easier to tune than a carb, Excited? No, pleased yes.

4 - I'll buy my own tuner. My experience is many 'mechanics' do not know the difference between an axe and and an Allen wrench.

5 - See my answer to #4

6 - Not at all. Custom EFI mapping is air density dependent. Depending on the sophistication of the EFI system, it may never need to be e-mapped or (just like with a carb) mapped almost daily for maximum performance. However, once you have the base parameters for a particular engine, air density changes typically can be plugged into a formula and accurate changes made without testing.

I think you need to do a bit more research. Your questions make a lot of assumptions that are a stretch. For example, the very first tuning must be done on a dyno or at least with a data logger and wideband O2 sensor. If done with a logger terrain is a factor that must be accounted for when analyzing the data. Blind trackside tweaks becomes pure guess work and the only advantage then becomes not getting gas on your fingers. You must have a baseline done in tightly controlled circumstances. Then, based on air density, work from there. You need to increase you breadth of tuning knowledge before you try to create a tuning business. Do not put the cart before the horse. Please understand I am not trying to be harsh but rather provide good criticism.

[Smacaroni]

here's what he said:

1) 0

2) yes

3) No, I don't ride

4) I wouldn't pay anything for a custom tune. I'd expect it to tune itself via O2 sensor feedback, or allow editing.

5) Closed loop

6) Never

More about tuning... If I were to go about designing such a device I'd go with the AVR family of microcontrollers. A T2313 can support a 10x10 map and a Mega168 will give you up to a 22x22 map. You don't want to fuss with a delicate mass air flow sensor on a bike, so using a manifold air pressure sensor in a speed density configuration would be better. You'd have to fit a narrowband oxygen sensor somewhere in the exhaust, a cheap single wire sensor would work fine.

The microcontroller would maintain a table of fuel values based on the O2 sensor, MAP sensor and tach signal. They would be accurate for an air to fuel ratio of 14.7:1. Once this table is established you can achieve any ratio you want by scaling the values. For maximum power in most engines you want around 12.5:1.

One of the key points of EFI is the fact that it tunes itself, correcting for temperature, elevation, fuel quality, air filter restriction, etc. I really don't think someone is going to go through the effort and expense of installing such a device while being held hostage to $100 tunes which are basically the equivalent of a screw turn on the carbs they already have. A good EFI system shouldn't move the tuning process to a laptop, it should eliminate it altogether and let the rider spend more time on his bike, not in front of his computer.

[William1]

SM, the issue most bikes have is they have a minimal amout of sensors. Takes electrical power to run them (more stator, battery....) and adds weight. Have your friend look at a RMZ450. The only bikes that have the full boat of sensors are the big touring bikes like Goldwings and BMW RS1200's. So for now, most bikes will need remapping when any change of significance is done (engine mod or air density). On the plus side, for most riders, once the bike is tuned, they will be fine with it and the variations of air density will be a non-issue as they are not that critical.

[Dr Honda]

SM...

you have some good points... but a system with a narrowband O2 would never be able to tune itself, and adding a wide band system would be to expensive and fragile on a dirt bike.

Also... there are almost NO EFI system that are self tunable. They can adjust in a small window around their maps. If you don't believe me... bolt a turbo on a stock EFI car and see how long the engine lasts.

[Smacaroni]

He's been watching the thread:

Oxygen sensors don't consume any power and MAP sensors use a fraction of a watt. They weigh less than a pound combined. The biggest electrical loads in an EFI system are going to be the pump and injectors. The control unit can easily be the size and weight of a pack of playing cards.

From a tuning standpoint, an open loop factory EFI system is inferior to a carb as there's little you can do with it. I can't think of any reason someone would install such a system on their bike, even for free. I'm sure bikes that come with such a system are very reliable and consistent, but like you said, remapping is required after any significant upgrades. Without some kind of oxygen sensor, any remapping you do is still a shot in the dark.

I see several advantages in fitting a closed loop system to a previously carbed bike:

1. Maximum fuel economy and power

2. No remapping needed

3. Can quickly adapt to any mod or atmospheric condition

None of these advantages are present in open loop systems, so why would anyone bother retrofitting a carbed bike with one?

[Dr Honda]

SM... What have you been smokin'?

Most modern O2 sensors have a heater in them. (and all widebands do) They pull a bunch of power until they get enough heat from the engine.

also... ALL factory automotive EFI systems are closed loop. They have to be since the EPA, CARRB, and other smog Nazi's want control over them.

The type of system you are talking about would be fine in a big street bike... but if I was building a race bike... I wouldn't want a computer changing my base maps because I took a few corners easy. I would only want a slight change in fuel based off of air density, and temp.

This will be my last rebuttal.

[Smacaroni]

I'm just the messenger:

Fitted a turbo to a car with a stock ECU? Smacaroni can tell you I've done just that. A fifth injector was needed, but I built the custom controller to drive it. It holds 12.5:1 all day at 10 psi with no check engine lights and I didn't have to dyno tune it at all. It reads the wideband and tunes itself.

A system with a narrowband can indeed tune itself, OBD cars do it all the time. You're right about the wideband adding too much complexity and weight. It would accelerate learning, but it just isn't necessary. You can build an entire map solely from a narrowband signal. The base map on OBD cars just serves as a starting point. The manufacturers never expected the trim values to vary much since factory engines are more or less the same, but custom ECUs don't need to be limited in this way. The base map is there to keep the car from falling on its face each time you disconnect the battery.

Once you build your stoich table, you can lock it, use O2 sensor feedback to scale the entire thing and in closed loop, and achieve any air to fuel ratio you wish in open loop mode. You can run lean at light throttle for maximum fuel economy and rich at wide open throttle, occasionally stopping back at 14.7:1 to adjust the trim.

Yes, newer cars do have heated sensors, but sensors without heaters are still readily available. Even those with heaters don't require the element to be energized to produce a signal. Simply waiting another 30 seconds accomplishes the same thing.

[450exc115]

I appreciate all the comments guys. To get it back on track though I will say the buiness model is working with current EFI machines and the current tuners/programmers available for them. To be able to build a efi system or come up with the brain to work with current efi systm would be a big task and take a lot of time to reverse engineer.

The questions are general to not lead anyone to a specific answer and get general information from the experienced to the not so experienced.

Some of the research is to find if there is a market for it. The cost should be driven from profit margin and operating costs including inventory which without know what people are willing to pay makes it tough to know if the buisness should be started in the first place.

As for me and my talents.. 12 years as an engineer 10 years of racing dirtbikes and 5 years of tuning EFI cars (turbo ones) has given me some experience. ?

I am also of the opinion that our current state of EFI is equal to what GM and Ford produced in the late 80s early 90s with OBD1 cars.

[Smacaroni]

12 years as an engineer, a few weeks with a decent modern circuitry book, a reasonable grasp on logic and a stoichiometric ratio is all you need to build a EFI system. No need to reverse engineer.

Now if you want that to work hand in hand with a specific bike's CDI/TCI etc. that would require some reverse engineering, lots depending on what you want to do. I would just use a oscilloscope and timing light to figure out the timing curve and go with that.

Or don't worry about it at all. The guy who's comments I posted, is the one who helped me with the psuedo CDI unit on my XT550, it's five components, one is a programmable chip. It's way over built.

Forget everything Ford and GM did, it was then and is now, crap. Dream bigger.

If you want an example of what could and should be done, look at Flying Miata and Lynx. At least I think it's called Lynx. It might be Lynks or Links.

[Etown]

this is a very interesting thread. i am mechanical engineer looking to start my own motorcycle related business and with all the new efi bikes there is definitely potential to provide a product or service. just need to find a niche in the market where people would pay for the service/product.

[450exc115]

?

Yeah but don't steal my business idea... LOL

Smacacroni,

I apreaciate the comments but the cost to make boards, chips, package, program, etc is not a small investiment. Without endless capital some things are tough to start and finish. But start small and grow into the greater nich would be the plan..

I look at F2 as an example. Great products, great idea, but they couldn't deliver and had manufacturing issues that led to them going out of business.

Thanks all for the comments and give me more!!!

[Etown]

?

Yeah but don't steal my business idea... LOL

haha dont worry dont really have the resources to, i wish i was still in school so i could use their labs and professors and what not.

[Smacaroni]

Since you're still looking for comments...

Why not get a good handle on a few of the EFI machines that are typically raced at your local events.

This could even be... (gasp) sprint cars, those guys love the Kawi 636 motors. But build them and race them in a way that you'd never want in a street or race bike.

Anyway, once you've got two or three current models that you know inside and out, offer your services to a few local racers in exchange for some sticker space.

If your product is good, you'll have people banging on the door of your van with cash.

If you pull it off in the sprint car circuit, wads of cash. I know, there's lots of guys who race on budgets too small to finish a season, but there's also guys who run them with deep, deep pockets.

Oh, and the psuedo CDI circuit was $20 + a harley coil... hey, I had it in my shed so I call it "free". However, the knowledge to build the circuit is something I don't have.