lifted club cars - lifted ezgo
Home FAQDonate Who's Online
Go Back   Buggies Gone Wild Golf Cart Forum > Golf Cart Repair and Troubleshooting > Gas EZGO
Gas EZGO Gas EZGO Marathon, Medalist, TXT and RXV.



Post New Thread  Reply
 
Thread Tools Display Modes
Old 11-10-2015, 12:14 PM   #21
02TXT
Not Yet Wild
 
Join Date: Apr 2013
Posts: 30
Default Re: Fuel Injected TXT 295cc w/ LS2 Coils

Quote:
Originally Posted by RibbedGiraffe View Post
I want to have beers with you while you teach me! Cheers!
Hey, anything I can help with I look forward to I love beer so we are in good company
02TXT is offline   Reply With Quote
Alt Today
BGW

Golf car forum Sponsored Links

__________________
This advertising will not be shown in this way to registered members.
Register your free account today and become a member on Buggies Gone Wild Golf Cart Forum
   
Old 11-10-2015, 12:48 PM   #22
1989Marathon
Gone Wild
 
Join Date: Oct 2010
Posts: 6,730
Default Re: Fuel Injected TXT 295cc w/ LS2 Coils

Can you tell me what it is about the fuel delivery portion of the Speeduino project that is "far more advanced"? You don't need to add complexity for the sake of fuel delivery. I have MAP, IAT, TPS, AFR sensors, multiple fuel maps, startup enrichment, acceleration enrichment, min/max protection, etc...

I am not running parallel processors. I have 2 independent processors, with the sensors wired in parallel to each of them. This was one of the steps I took to be able to achieve deterministic timing.

As for the missing tooth/hall setup, I can only imagine that the firing of the injector is based off of a specific tooth position, not all of them. That makes the only important tooth, the last one before injection starts.
Why is it specifically that you think a single trigger can't be used? The only reason I can come up with is because of a variation in processing time. The engine speed can't change fast enough within one revolution to have appreciable effect.

I'm not suggesting that we should avoid any additional information as it would burden the processor. The additional information(MAP, TPS, IAT, AFR) should be input to the controller during the main program loop. ~1msec.

I have already achieved deterministic timing with both controllers. It is less a function of the controller, and more a function of the speed/accuracy/repeatability of the Schmitt trigger, and the efficiency of the code I have written.

I'm also not trying to say that automakers are just too lazy or not smart enough to engineer "deterministic timing". I'm just saying that I don't believe it is necessary to have a toothed wheel. That will reduce complexity and cost for the end user.
1989Marathon is offline   Reply With Quote
Old 11-10-2015, 02:39 PM   #23
RibbedGiraffe
Not Yet Wild
E-Z-GO
 
Join Date: Jun 2009
Location: Washington
Posts: 57
Default Re: Fuel Injected TXT 295cc w/ LS2 Coils

I've been over on the Speeduino sight all day - I don't know if I missed it or not, but I'm wondering if there is a way to incorporate a knock sensor into the mix. I know nothing of writing code etc.. but now I want to learn.
RibbedGiraffe is offline   Reply With Quote
Old 11-10-2015, 02:57 PM   #24
1989Marathon
Gone Wild
 
Join Date: Oct 2010
Posts: 6,730
Default Re: Fuel Injected TXT 295cc w/ LS2 Coils

Quote:
Originally Posted by RibbedGiraffe View Post
I've been over on the Speeduino sight all day - I don't know if I missed it or not, but I'm wondering if there is a way to incorporate a knock sensor into the mix. I know nothing of writing code etc.. but now I want to learn.
Yes. A knock sensor can be hooked up to the controller analog input and code written to retard the timing if the engine is detonating.
1989Marathon is offline   Reply With Quote
Old 11-10-2015, 03:57 PM   #25
02TXT
Not Yet Wild
 
Join Date: Apr 2013
Posts: 30
Default Re: Fuel Injected TXT 295cc w/ LS2 Coils

Quote:
Originally Posted by 1989Marathon View Post
Can you tell me what it is about the fuel delivery portion of the Speeduino project that is "far more advanced"? You don't need to add complexity for the sake of fuel delivery. I have MAP, IAT, TPS, AFR sensors, multiple fuel maps, acceleration enrichment, etc...

I am not running parallel processors. I have 2 independent processors, with the sensors wired in parallel to each of them. This was one of the steps I took to be able to achieve deterministic timing.

As for the missing tooth/hall setup, I can only imagine that the firing of the injector is based off of a specific tooth position, not all of them. That makes the only important tooth, the last one before injection starts.
Why is it specifically that you think a single trigger can't be used? The only reason I can come up with is because of a variation in processing time. The engine speed can't change fast enough within one revolution to have appreciable effect.

I'm not suggesting that we should avoid any additional information as it would burden the processor. The additional information(MAP, TPS, IAT, AFR) should be input to the controller during the main program loop. ~1msec.

I have already achieved deterministic timing with both controllers. It is less a function of the controller, and more a function of the speed/accuracy/repeatability of the Schmitt trigger, and the efficiency of the code I have written.

I'm also not trying to say that automakers are just too lazy or not smart enough to engineer "deterministic timing". I'm just saying that I don't believe it is necessary to have a toothed wheel. That will reduce complexity and cost for the end user.
When I use the word timing, I am speaking of ignition timing, not fuel timing. As I stated, one tooth setup like yours is fine for fuel only. That was how I ran my cart to start with... but if you ever want to start advancing your ignition timing to get more power, you will have to add a missing tooth wheel.

As to why the Speeduino is more advanced than your setup, it is a long list of functionality, but to sum it up simply, it does fuel and ignition timing.

When do you think you will have your cart back up and running?
02TXT is offline   Reply With Quote
Old 11-10-2015, 04:17 PM   #26
1989Marathon
Gone Wild
 
Join Date: Oct 2010
Posts: 6,730
Default Re: Fuel Injected TXT 295cc w/ LS2 Coils

Quote:
Originally Posted by 02TXT View Post
When I use the word timing, I am speaking of ignition timing, not fuel timing. As I stated, one tooth setup like yours is fine for fuel only. That was how I ran my cart to start with... but if you ever want to start advancing your ignition timing to get more power, you will have to add a missing tooth wheel.

As to why the Speeduino is more advanced than your setup, it is a long list of functionality, but to sum it up simply, it does fuel and ignition timing.

When do you think you will have your cart back up and running?
I've taken a quick look at the Speeduino code and unless I am missing something, it appears to be doing the same thing with the ignition timing as I am proposing. The difference is this:

My sensor will trigger at about 10 BTDC of cylinder number 2.(This leaves room for timing advance) It will calculate how long it needs to wait until it reaches the dwell angle, then start the dwell, and time out. At the same time it will do the same thing for cylinder number 1, 90 further.

Your sensor is sensing every tooth and running an interrupt every occurrence. It looks to see if the crankangle is less than the programmed ignition timing value, calculates the amount of time until it will reach the dwell angle, waits that amount of time, then it starts the dwell.

The main difference is that you need to update the ignitionstarttime on every tooth because of program latency. I do not have to do that. That's the difference between trying to execute several thousand lines of code, with numerous function calls and many interrupts, and executing 75 lines of code with one single interrupt.

Is that about how it works, or am I missing something?

The motor is currently on a test stand and will probably not be back in the cart until spring.

You will find that you didn't happen to stumble on a forum full of dummies. There is probably an expert on just about any subject, lurking in the shadows. These guys may play with golf carts for fun, but they are extremely smart in their area of expertise.
1989Marathon is offline   Reply With Quote
Old 11-11-2015, 11:15 AM   #27
02TXT
Not Yet Wild
 
Join Date: Apr 2013
Posts: 30
Default Re: Fuel Injected TXT 295cc w/ LS2 Coils

Let's star from the bottom, shall we:

Quote:
Originally Posted by 1989Marathon View Post
The main difference is that you need to update the ignitionstarttime on every tooth because of program latency. I do not have to do that. That's the difference between trying to execute several thousand lines of code, with numerous function calls and many interrupts, and executing 75 lines of code with one single interrupt.
This statement is just flat wrong. Go have a look in decoders.ino and review the following functions: triggerSetup_missingTooth(), triggerPri_missingTooth() and getCrankAngle_missingTooth().

Every time the toothed wheel triggers the interrupt, it updates the current crank angle. This crank angle is more accurate the more teeth there are on the missing tooth wheel. The ignition and fuel timing are constantly updated using this information.

That's what you don't seem to get - engines under load are not linear in their response like they are on your test stand. You may be able to achieve some semblance of ignition timing on your test stand with a single trigger, but in your cart you will never match the performance of missing tooth wheel setup. Period. Remember the yard stick metaphor I used in my last post?

As to code latency, you still don't seem to grasp the power of a 16,000,000Hz processor - even 10,000 lines of code consuming 5 instructions per line is only consuming 1/32 of the processor in a second - still lot's of processor capacity left.

Even if you use a 4Ghz Intel i7 (4,000,000,000hz btw) and did your timing in 75 lines along with your single trigger setup, you still would never match the ignition timing accuracy of a missing tooth setup and a little ol' 16Mhz Arduino.

Quote:
Originally Posted by 1989Marathon View Post
I've taken a quick look at the Speeduino code and unless I am missing something, it appears to be doing the same thing with the ignition timing as I am proposing. The difference is this:

My sensor will trigger at about 10 BTDC of cylinder number 2.(This leaves room for timing advance) It will calculate how long it needs to wait until it reaches the dwell angle, then start the dwell, and time out. At the same time it will do the same thing for cylinder number 1, 90 further.

Your sensor is sensing every tooth and running an interrupt every occurrence. It looks to see if the crankangle is less than the programmed ignition timing value, calculates the amount of time until it will reach the dwell angle, waits that amount of time, then it starts the dwell.
This again, goes to my argument: With a single trigger on the crank, how in the world do you know the crank is at 10 before TDC and then at 90 (or any other position)? Sure, if your single trigger is mounted at 10 BTDC you can get that one, but beyond that you are guessing based on the time it took for the last prior full revolution. Again, it can never be as accurate as using a missing tooth wheel to know the position of the crank during the current revolution.

Quote:
Originally Posted by 1989Marathon View Post
You will find that you didn't happen to stumble on a forum full of dummies. There is probably an expert on just about any subject, lurking in the shadows. These guys may play with golf carts for fun, but they are extremely smart in their area of expertise.
I have not implied that anyone was a "dummy", and I have been on the forum for 2 1/2 years.

Bottom line, my cart is up and running great with both fuel and ignition timing and you are still in test phase. My thread was about my completed cart, so I ask that going forward we continue our discussion on your thread, as it seems to pertain more to your build than my running cart.

Thank you

Louis
02TXT is offline   Reply With Quote
Old 11-11-2015, 09:22 PM   #28
1989Marathon
Gone Wild
 
Join Date: Oct 2010
Posts: 6,730
Default Re: Fuel Injected TXT 295cc w/ LS2 Coils

Quote:
Originally Posted by 02TXT View Post
As to code latency, you still don't seem to grasp the power of a 16,000,000Hz processor - even 10,000 lines of code consuming 5 instructions per line is only consuming 1/32 of the processor in a second - still lot's of processor capacity left.

Even if you use a 4Ghz Intel i7 (4,000,000,000hz btw) and did your timing in 75 lines along with your single trigger setup, you still would never match the ignition timing accuracy of a missing tooth setup and a little ol' 16Mhz Arduino.

This again, goes to my argument: With a single trigger on the crank, how in the world do you know the crank is at 10 before TDC and then at 90 (or any other position)? Sure, if your single trigger is mounted at 10 BTDC you can get that one, but beyond that you are guessing based on the time it took for the last prior full revolution. Again, it can never be as accurate as using a missing tooth wheel to know the position of the crank during the current revolution.
These statements tell me a lot about who I am dealing with. It sounds like someone who is butt-hurt that someone else figured out how to do the same thing much more simply. Eventually, I'm sure someone will figure out how to do it simpler than I did. I will be butt-hurt also, but I will ask questions about how they did it, instead of telling them it can't be done. If I want to learn something, I find it is better to ask questions and listen, than to keep talking and not hear. Apparently you think your setup is the only way to do it. That's just not true, and it won't be true no matter how many times you repeat yourself. I am done here.
1989Marathon is offline   Reply With Quote
Old 11-11-2015, 10:10 PM   #29
02TXT
Not Yet Wild
 
Join Date: Apr 2013
Posts: 30
Default Re: Fuel Injected TXT 295cc w/ LS2 Coils

Quote:
Originally Posted by 1989Marathon View Post
These statements tell me a lot about who I am dealing with. It sounds like someone who is butt-hurt that someone else figured out how to do the same thing much more simply. Eventually, I'm sure someone will figure out how to do it simpler than I did. I will be butt-hurt also, but I will ask questions about how they did it, instead of telling them it can't be done. If I want to learn something, I find it is better to ask questions and listen, than to keep talking and not hear. Apparently you think your setup is the only way to do it. That's just not true, and it won't be true no matter how many times you repeat yourself. I am done here.
I literally laughed out loud after reading this... have fun with your test stand, buddy.

Louis
02TXT is offline   Reply With Quote
Old 11-11-2015, 10:36 PM   #30
1989Marathon
Gone Wild
 
Join Date: Oct 2010
Posts: 6,730
Default Re: Fuel Injected TXT 295cc w/ LS2 Coils

Quote:
Originally Posted by 02TXT View Post
I literally laughed out loud after reading this... have fun with your test stand, buddy.

Louis
Apparently you have little knowledge of higher-order math, the value of efficient coding, or the benefit of using the processor clock to your advantage.
Efficient coding has nothing to do with processor speed. Even the fastest processors can't compensate for inefficient coding.
If you'd like to learn something, I can help.

1989Marathon is offline   Reply With Quote
Reply
Go Back   Buggies Gone Wild Golf Cart Forum > Golf Cart Repair and Troubleshooting > Gas EZGO


Thread Tools
Display Modes


Similar Threads
Thread Forum
My 2005 yamaha pro hauler 700 with a fuel injected 07 gsxr 750 Bike & ATV Engine Swaps
Fuel injected clone??? Clone Engine Swaps
fuel injected yamaha drive Gas Yamaha
CARBS vs. FUEL INJECTED ???? Bike & ATV Engine Swaps
88 EZGO Workhorse Oil Injected Help Gas EZGO


All times are GMT -5. The time now is 04:03 PM.


Club Car Electric | EZGO Electric | Lifted Golf Carts | Gas EZGO | Used Golf Carts and Parts

Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2021, vBulletin Solutions, Inc.
This Website and forum is the property of Buggiesgonewild.com. No material may be taken or duplicated in part or full without prior written consent of the owners of buggiesgonewild.com. 2006-2017 Buggiesgonewild.com. All rights reserved.