Ski-Doo Snowmobiles Forum banner
1 - 13 of 13 Posts

·
Registered
Joined
·
359 Posts
That's odd. Most modern EFI systems do monitor fuel pressure to control fuel pressure via pump speed (PWM Fuel pump) or to at least derive a basic injector pulse width. Fuel injectors are rated at how much fuel they flow for a given amount of time at a defined fuel pressure. If these factors are known an injector pulse with is calculated. This allows for basic or open loop fuel control and other things like calculating fuel economy. And that's just the start. Older EFI systems used a mechanical regulator and a fixed speed pump. Injector pressure at the rails was assumed. This proved to be problematic in high performance systems. If the real pressure wasn't the same as the assumed the ECU may not be able to compensate for pressure deviations depending on a few other conditions. So even when some later systems still used fixed speed pumps and regulators monitoring fuel pressure to modify fuel injector pulse width is a good idea. Generally higher performance engines require more accurate fuel control to prevent engine failure. More accurate fuel control is a one of the big reasons we are getting more power out of engines these days. Can you tell I used to really be into this at one point?

But I admit, that while I have a good amount of experience with automotive ECUs, I've never been in my doo's computer. I am curious to find out more.

Ed
 

·
Registered
Joined
·
976 Posts
This is from the shop manual:
"Fuel is used to maintain proper fuel injector operating
temperature.
The flow starts from the fuel pump, through the
ECM, then around the voice coil inside the fuel
injector housings to cool down the fuel injector
components.
Fuel enters the inlet port located at the bottom
of the fuel injector housing and exits through the
outlet port on top of the fuel injector."
 

·
Registered
Joined
·
3 Posts
That's odd. Most modern EFI systems do monitor fuel pressure to control fuel pressure via pump speed (PWM Fuel pump) or to at least derive a basic injector pulse width. Fuel injectors are rated at how much fuel they flow for a given amount of time at a defined fuel pressure. If these factors are known an injector pulse with is calculated. This allows for basic or open loop fuel control and other things like calculating fuel economy. And that's just the start. Older EFI systems used a mechanical regulator and a fixed speed pump. Injector pressure at the rails was assumed. This proved to be problematic in high performance systems. If the real pressure wasn't the same as the assumed the ECU may not be able to compensate for pressure deviations depending on a few other conditions. So even when some later systems still used fixed speed pumps and regulators monitoring fuel pressure to modify fuel injector pulse width is a good idea. Generally higher performance engines require more accurate fuel control to prevent engine failure. More accurate fuel control is a one of the big reasons we are getting more power out of engines these days. Can you tell I used to really be into this at one point?

But I admit, that while I have a good amount of experience with automotive ECUs, I've never been in my doo's computer. I am curious to find out more.

Ed
Oh my God! That's why I'll never be a mechanic-it takes so much digging into technical details that I just don't have the patience for it. I like speed, but I don't like digging deep.
Question: is it possible to control fuel accurately without an on-board computer? I'm trying to figure out how much influence the on-board computer has on the fuel supply.
 

·
Registered
Joined
·
976 Posts
Oh my God! That's why I'll never be a mechanic-it takes so much digging into technical details that I just don't have the patience for it. I like speed, but I don't like digging deep.
Question: is it possible to control fuel accurately without an on-board computer? I'm trying to figure out how much influence the on-board computer has on the fuel supply.
What computers are doing for us is it is giving us over 360hp 2l engines that are available to the mass market, that run on pump gas and they comes with a 4y warranty.

In the carb days a well tuned 2l would give you at most 125hp if you where lucky at twice the fuel consumption.
 

·
Registered
Joined
·
359 Posts
Oh my God! That's why I'll never be a mechanic-it takes so much digging into technical details that I just don't have the patience for it. I like speed, but I don't like digging deep.
Question: is it possible to control fuel accurately without an on-board computer? I'm trying to figure out how much influence the on-board computer has on the fuel supply.
Most mechanics are never near this stuff but should have a basic understanding. I was SCTs first customer/tuner. My backgrounds of drag racing/engine building/avionics tech/system engineering kinda made me a good choice to help them getting their company off the ground. I’ve received training from the experts. I then spent about 13 years custom tuning most any domestic car brought to me. I retired from tuning when working two full-time plus jobs contributed to health issues. I then sold my truck/trailer/drag car. I recently sold my last built street car. I play with old sleds and an occasional ATV for entertainment now

As for your question, it all depends on what you are trying to do. More computing power allows for monitoring more sensors and reading them faster. You can then do more complex calculations using that higher resolution data input. There are DIY EFI solutions that allow you to use and scale to just want you want to use. There has been a trend over the last 10 years or so to try to simplify some of these controls and sensor by constantly controlling fuel 100% closed loop and more adaptive strategies. But that’s a big topic.

So no on board computer means carbs or mechanical fuel injection. Maybe I’m missing something. I would think it’s quite possible to build a simple EFI system from some form of Arduino. Very minimal. Remember EFI also controls the ignition system as well.

Thanks, Ed
 

·
Registered
Joined
·
248 Posts
Most mechanics are never near this stuff but should have a basic understanding. I was SCTs first customer/tuner. My backgrounds of drag racing/engine building/avionics tech/system engineering kinda made me a good choice to help them getting their company off the ground. I’ve received training from the experts. I then spent about 13 years custom tuning most any domestic car brought to me. I retired from tuning when working two full-time plus jobs contributed to health issues. I then sold my truck/trailer/drag car. I recently sold my last built street car. I play with old sleds and an occasional ATV for entertainment now

As for your question, it all depends on what you are trying to do. More computing power allows for monitoring more sensors and reading them faster. You can then do more complex calculations using that higher resolution data input. There are DIY EFI solutions that allow you to use and scale to just want you want to use. There has been a trend over the last 10 years or so to try to simplify some of these controls and sensor by constantly controlling fuel 100% closed loop and more adaptive strategies. But that’s a big topic.

So no on board computer means carbs or mechanical fuel injection. Maybe I’m missing something. I would think it’s quite possible to build a simple EFI system from some form of Arduino. Very minimal. Remember EFI also controls the ignition system as well.

Thanks, Ed
Nice post - well done
I too got into tuning fuel injection systems (personal use)
At first timing tables , fuel tables and pwc etc. hurt my head - and then it all makes sense.

The snowmobile (xm 2012 to 2016) systems ~appear to use a modified chevy ish system - At first i figured *someone would have posted a prom online by now or at least the definition files

Now they seem to have gone their own way for the newer 850 systems

I *think (imo) that the clutch system on these sleds is key to keeping the computer in the sweet spot.
make the sled rev to the "sweet spot" on the curve (table) - based on, YOUR SLED, YOUR WEIGHT, YOUR AREA and RIDING type.
 

·
Registered
Joined
·
359 Posts
Nice post - well done
I too got into tuning fuel injection systems (personal use)
At first timing tables , fuel tables and pwc etc. hurt my head - and then it all makes sense.

The snowmobile (xm 2012 to 2016) systems ~appear to use a modified chevy ish system - At first i figured *someone would have posted a prom online by now or at least the definition files

Now they seem to have gone their own way for the newer 850 systems

I *think (imo) that the clutch system on these sleds is key to keeping the computer in the sweet spot.
make the sled rev to the "sweet spot" on the curve (table) - based on, YOUR SLED, YOUR WEIGHT, YOUR AREA and RIDING type.
Learning on the job was daunting at first. But after a while it just didn't matter. NA, N20, Turbo, blown whatever. I did mostly Fords. Just about all models. A LOT OF 2v and 4v modulars. Next would have been Windsor motors. Stock to radical. Street to road racing and drag racing. The same car could have 5 different strategies for that model year. Some were speed density and most Mass air. Just just tuned it! The biggest problem was getting builders/customers to put the right combination of parts/sensors on the vehicle for what they are trying to do. I would spend hours and even days figuring out all of the issues just so I could tune it.

I had a handful of articles and mentions in the Mustang rags. But best was part of an article in Hot Rod magazine. And I did a sponsored tune for an NMRA Factor Stock 4v. First 281 CID to win a season championship. I still have my SCT dealership and all of their software. But with recent EPA crackdowns I have no intention of doing anything as a business.

Once I decided to get out of it, instead of tuning the cars, I would teach shops how to do their own. The troubleshooting the combinations I really couldn't help them with. There's a lot of intuition troubleshooting a custom mess you didn't build. :)

I really would like to learn more about sled EFI. But I'm assuming it's a small, tight crowd. Learning could take some time on your own. And probably a few mistakes. I did have to replace a few ECUs experimenting early on. And somewhat expensive?? Other than my speedo being 10% high all of the time and how the grip heaters work I'm pretty happy with my current sled. I've built a few bread boarded devices to fix both of those problems. Even wrote the firmware. But I always get slowed up by creating a finalized product. It's when you have to actually spend money on a project.

Thanks,

Ed
 
1 - 13 of 13 Posts
Top