Sniper EFI for boat.

[Bigbore500r]

Right? Just ask Alex!

(Did I type that out loud? Damit!)

He does know how to go fast though. . . his full interior 4100lb SS ran 7.67, still running IRS in the rear and small tire! That's haulin

 

[mjc]

If i could justify the cost i would like to go to EFI university here in havasu. He teaches from no map up to completely done runnning. Mist people start with some form of running map and modify from there. Cost is to much for me to just do my own stuff.

 

[rush1]

Ohhh Noooooo! You're a Data Scientist? I'm assuming LOF is "Local Outlying Factor" in which case I will say adamantly "I RESEMBLE THAT REMARK!!!" (and take it as a compliment). If "LOF" means something else, I'm not sure I want to know... ;-)

Lube
Oil
Filter

 

[DaveH]

Back on the Knock sensor topic, just trying to square what i'm reading with what I thought I understood.

According to Ford, The 2024 Ford Bronco's 2.3-liter EcoBoost® 4-cylinder engine produces 300 horsepower and 325 pound-feet of torque with premium fuel. With regular gas, the engine produces 275 horsepower and 315 pound-feet of torque.

The explanation for how this is done: Ford's EcoBoost 2.3 engine uses a calculation called the Octane Adjust Ratio (OAR) to detect the octane of the fuel being used. The OAR is adjusted based on whether the engine detects detonation, and over time the computer learns the octane level of the fuel. The OAR is fully adjusted for higher power at -1.0, and fully retarded at 1.0.

while its easy to trend what a knock sensor is doing over time, i bet the OEM is doing a LOT more then just pullin' a couple degree's. id wager they are pulling boost, changing mixture, all sorts of stuff to meet their targets that only the OEM would really entirely know about.

 

[Racey]

Using this type of spark advance / knock retard logic that your stock Ecoboost uses is fine for OEM stock vehicles. The tune is very conservative, and they need to be able to cope with shit gas that you might encounter in the middle of nowhere. When it detects a pattern of detonation, or unacceptable levels of detonation, it reverts to a low octane spark table that is more conservative, so its not just “bouncing off” the knock sensors and advancing / retarding against detonation.

I'm not sure how the new Ford ecm tables are structured on their late model stuff, but GM stuff uses 2 different spark tables, depending on octane. If it detects knock above a certain threshold / consistency, it will revert to the low octane table and stay there and operate. It will attempt to advance timing and "test" against the table every so often, to see if it can get away with more timing. If it is successful in reverting back to the high octane spark values, it will revert to running from the high-octane table after the parameters are met (how many times was it successful in running from the high-octane table, how long has it been since it detected detonation using high octane table, etc . . .)

It's different than just running too much timing, and consistently relying on the motor running "on" the knock sensor's command to retard timing. all the time.

A "ping" here and there with a stock motor is one thing, but a performance motor hammering away against the knock sensors and going "ping PING PING - Retard. . . . ping PING....retard" is super hard on stuff and like Racey said in another post, is real hard on both pistons, and exhaust valves.

Knock sensors in a performance application are best used as a tuning aid, and a last ditch safety net for when shit goes wrong - not something you want to rely on for constant spark control under load, especially WOT..........even if it is in "CONTROLLED IN REAL TIME BY MEFI”

When the GM stuff is tuned for a performance application, here's the basic steps done to keep the motor from relying on the knock sensors, and to get timing in a safe range across the map:

• Low octane spark table is changed to match the high octane spark table (basically, eliminating this variable for tuning)
• The vehicle is ran thru different driving conditions, and the knock retard is logged via a histogram that mimics the layout of the main spark table
• Once enough reliable data has been collected under different conditions, the histogram will show the cells that experienced knock retard, and the average amount of timing that was pulled to eliminate detonation. Example - Histogram log will look like this:
• View attachment 1375133
• The histogram table then gets "copy / pasted" over the main spark tables and applied as a modifier to the existing values, in the cells (cylinder air mass and RPM) that were an issue. For example - At 4000rpm, under a Cylinder air mass of 0.54 g/cyl, timing would be reduced by 0.4 degrees.
• The car is ran and logged again, until the knock retard histogram is acceptable. It's normal to pick up a small ping here and there, and the table I pasted above is very mild. But when you see knock all over the map, and its having to pull alot of timing - that needs to be fixed.
• When starting fresh with a performance engine, the spark tables are set conservative, and then you "creep up" on the sweet spot. You don't want to data log the car, see a shitload of knock, and then work backwards and reduce timing. Sometimes, the Grim Reaper only knocks once . . .

This is information for everyone that wants to know. Thanks for taking the time to write all this out

 

[rivermobster]

while its easy to trend what a knock sensor is doing over time, i bet the OEM is doing a LOT more then just pullin' a couple degree's. id wager they are pulling boost, changing mixture, all sorts of stuff to meet their targets that only the OEM would really entirely know about.

Absolutely.

You have variable valve timing, and BMW even has variable lift! You also have variable length intake runners and per cylinder timing and injector pulse width.

And all this tech is from over a decade ago? I left the dealership a Long time ago now.

Who could have imagined 700hp engines, from the factory, With a warranty, that is capable of 20+ mpg?

Not me, that's for damm sure.

And are also 100 percent right about what the factory will put out there. Many times, not enough info is supplied.

 

[Bigbore500r]

while its easy to trend what a knock sensor is doing over time, i bet the OEM is doing a LOT more then just pullin' a couple degree's. id wager they are pulling boost, changing mixture, all sorts of stuff to meet their targets that only the OEM would really entirely know about.

I’m sure the new forced induction stuff from OEM’s goes much deeper these days, the Chrysler stuff has some very diffferent ways it goes about things compared to GM. I haven’t played with the Ford stuff, but given the complexity of the motors I can only imagine

 

[n2otoofast4u]

GRADS AT THE NEXT LEVEL!!! He’s gotten all ya’ll this time!

 

[Bigbore500r]

GRADS AT THE NEXT LEVEL!!! He’s gotten all ya’ll this time!

Finkel is Einhorn????

 

[ElectroNautical]

So when is the CAN BUS network switching to silicon optronics to save weight from copper wiring?

Who said "there are no stupid questions."? LoL. Ok, so you're either a syncophant, a minion or a parrot. Let's see which one...

I'd guess you are just being a dick and trying to amuse people, but your smart-ass question shows ignorance in at least four areas and stupidity in two others, so you are as amusing as a clown. Nice Clown and Clown-car photos OBTW ;-)

1) The Controller Area Network (CAN) bus you heard about in VocTech (but apparently didn't understand) shares signal multiplexing methods with the Marine equivalent, NMEA 2000, but that's all the similarity their is.
2) Haven't you heard? Optical fiber can't carry electrical current, and NMEA 2000 devices get their electrical power directly from the NMEA 2000 cable.
3) Ruggedized optical fiber and the waterproofed connectors suitable for use in Boats, Aircraft and Military vehicles actually weighs more per unit of length than an NMEA cable, and weighs twice as much as the copper CAN bus wiring in your car/truck.
4) The ruggedized optical-to-copper signal converters you would need at every optical termination point would weigh more and cost five times more than all the cabling combined.
5) The five copper wires in an NMEA 2000 cable are very thin gauge wire, so neither cost nor weight is ever a factor.
6) There is no such thing as "silicon optronics", which is trademarked name of a company that makes IMAGE SENSORS. You take a lot of pictures of clowns and clown cars so I'd have thought you'd have known this.

Of course you would not have learned any of this riding the short bus to your VocTech class. You should have stayed in regular Jr. High School classes and paid attention in your science classes.

So...which are you?

A syncophant repeats what others say, understands context and meaning, but does not use it's OWN brain to evaluate what it hears before repeating it.
A minion understands context but not meaning in what it hears, takes orders and repeats on command.
A parrot hears and repeats without any understanding at all, and really only does this when it needs attantion.

Parrot...for sure.

"When there's a conversation going on about something you don't understand, you can keep quiet and let everyone think you're an idiot, or you can open your mouth and prove it."

 

[ElectroNautical]

If i could justify the cost i would like to go to EFI university here in havasu. He teaches from no map up to completely done runnning. Most people start with some form of running map and modify from there. Cost is to much for me to just do my own stuff.

Re: "most people start with some form of running map and modify from there..."

Right...and now it is usually the EFI computer doing the 'modify from there' part.

In the beginning of EFI (1970's) there was no map at all until you calculated it manually and burned it to a PROM, and then it became the 'running map' (accessed as a LUT or Lookup Table). Every performance engine build would have a custom map.

Later (1980's), you would upload a pre-calculated "Base Map" to the fancy new ERASEABLE EPROM...tune that manually on a dyno, save it to floppy disk, erase the 'Base Map' from your EPROM (using a UV light, in early EPROMs), then upload the dyno tune back from floppy to your EPROM. It was still accessed as a LUT.

Today, (since mid-1990s) a 'Base Map' is installed at the factory into EEPROM (electrically eraseable PROM) with enough data to start the engine, but then the computer takes over and modifies everything in real-time after the engine is started and all the sensors are online. Once the engine is running everything is calculated and recalculated in real time, runtime values are in DRAM and logged to EEPROM (or FPGA) so that the 'tune' can 'persist' after the engine is shut off, ready to go back into DRAM next time you start the engine. The original OEM 'Base Map' gets replaced by the computer-generated runtime values, and that will be where you start from the next time you start the engine.

Over time as these transitions made everything more and more automated, EFI tuners who could once put their kids through college doing this stuff have struggled making money. I've never met one who won't offer to "tune" a marine MEFI for you for $500-$1,000, but they really aren't doing much (or anything) the engine/MEFI wouldn't do by itself.

The honest ones are probably the ones who start up a school to teach EFI. Have you got a link for that "EFI University" you mentioned?

 

[MonkeyButt70]

Who said "there are no stupid questions."? LoL. Ok, so you're either a syncophant, a minion or a parrot. Let's see which one...

I'd guess you are just being a dick and trying to amuse people, but your smart-ass question shows ignorance in at least four areas and stupidity in two others, so you are as amusing as a clown. Nice Clown and Clown-car photos OBTW ;-)

1) The Controller Area Network (CAN) bus you heard about in VocTech (but apparently didn't understand) shares signal multiplexing methods with the Marine equivalent, NMEA 2000, but that's all the similarity their is.
2) Haven't you heard? Optical fiber can't carry electrical current, and NMEA 2000 devices get their electrical power directly from the NMEA 2000 cable.
3) Ruggedized optical fiber and the waterproofed connectors suitable for use in Boats, Aircraft and Military vehicles actually weighs more per unit of length than an NMEA cable, and weighs twice as much as the copper CAN bus wiring in your car/truck.
4) The ruggedized optical-to-copper signal converters you would need at every optical termination point would weigh more and cost five times more than all the cabling combined.
5) The five copper wires in an NMEA 2000 cable are very thin gauge wire, so neither cost nor weight is ever a factor.
6) There is no such thing as "silicon optronics", which is trademarked name of a company that makes IMAGE SENSORS. You take a lot of pictures of clowns and clown cars so I'd have thought you'd have known this.

Of course you would not have learned any of this riding the short bus to your VocTech class. You should have stayed in regular Jr. High School classes and paid attention in your science classes.

So...which are you?

A syncophant repeats what others say, understands context and meaning, but does not use it's OWN brain to evaluate what it hears before repeating it.
A minion understands context but not meaning in what it hears, takes orders and repeats on command.
A parrot hears and repeats without any understanding at all, and really only does this when it needs attantion.

Parrot...for sure.

"When there's a conversation going on about something you don't understand, you can keep quiet and let everyone think you're an idiot, or you can open your mouth and prove it."

You are exhausting on every level.

 

[Outdrive1]

You are exhausting on every level.

I’ve seen enough. I put the call in. Bye Felicia

 

[rivermobster]

So when is the CAN BUS network switching to silicon optronics to save weight from copper wiring?

BMW did this back when I was working for them, and that was well over a decade ago.

The wiring will be at a bare minimum today.

Weight saving is only one advantage.

The Speed of light is pretty quick...

Everything in the air bags was on fiber optics back then, as well as everything in the multi media systems.

No clue what they are doing these days.

 

[4Waters]

Re: "most people start with some form of running map and modify from there..."

Right...and now it is usually the EFI computer doing the 'modify from there' part.

In the beginning of EFI (1970's) there was no map at all until you calculated it manually and burned it to a PROM, and then it became the 'running map' (accessed as a LUT or Lookup Table). Every performance engine build would have a custom map.

Later (1980's), you would upload a pre-calculated "Base Map" to the fancy new ERASEABLE EPROM...tune that manually on a dyno, save it to floppy disk, erase the 'Base Map' from your EPROM (using a UV light, in early EPROMs), then upload the dyno tune back from floppy to your EPROM. It was still accessed as a LUT.

Today, (since mid-1990s) a 'Base Map' is installed at the factory into EEPROM (electrically eraseable PROM) with enough data to start the engine, but then the computer takes over and modifies everything in real-time after the engine is started and all the sensors are online. Once the engine is running everything is calculated and recalculated in real time, runtime values are in DRAM and logged to EEPROM (or FPGA) so that the 'tune' can 'persist' after the engine is shut off, ready to go back into DRAM next time you start the engine. The original OEM 'Base Map' gets replaced by the computer-generated runtime values, and that will be where you start from the next time you start the engine.

Over time as these transitions made everything more and more automated, EFI tuners who could once put their kids through college doing this stuff have struggled making money. I've never met one who won't offer to "tune" a marine MEFI for you for $500-$1,000, but they really aren't doing much (or anything) the engine/MEFI wouldn't do by itself.

The honest ones are probably the ones who start up a school to teach EFI. Have you got a link for that "EFI University" you mentioned?

Well, you haven't found the thread so you should continue to shut the fuck up until you find it

 

[Bigbore500r]

Re: "most people start with some form of running map and modify from there..."

Right...and now it is usually the EFI computer doing the 'modify from there' part.

In the beginning of EFI (1970's) there was no map at all until you calculated it manually and burned it to a PROM, and then it became the 'running map' (accessed as a LUT or Lookup Table). Every performance engine build would have a custom map.

Later (1980's), you would upload a pre-calculated "Base Map" to the fancy new ERASEABLE EPROM...tune that manually on a dyno, save it to floppy disk, erase the 'Base Map' from your EPROM (using a UV light, in early EPROMs), then upload the dyno tune back from floppy to your EPROM. It was still accessed as a LUT.

Today, (since mid-1990s) a 'Base Map' is installed at the factory into EEPROM (electrically eraseable PROM) with enough data to start the engine, but then the computer takes over and modifies everything in real-time after the engine is started and all the sensors are online. Once the engine is running everything is calculated and recalculated in real time, runtime values are in DRAM and logged to EEPROM (or FPGA) so that the 'tune' can 'persist' after the engine is shut off, ready to go back into DRAM next time you start the engine. The original OEM 'Base Map' gets replaced by the computer-generated runtime values, and that will be where you start from the next time you start the engine.

Over time as these transitions made everything more and more automated, EFI tuners who could once put their kids through college doing this stuff have struggled making money. I've never met one who won't offer to "tune" a marine MEFI for you for $500-$1,000, but they really aren't doing much (or anything) the engine/MEFI wouldn't do by itself.

The honest ones are probably the ones who start up a school to teach EFI. Have you got a link for that "EFI University" you mentioned?

 

[DarkHorseRacing]

You are exhausting on every level.

The best part was I didn’t even reply to the tool, it just piped up for no reason and got nothing right. But I agree, LOF needs to go.

 

[Bigbore500r]

The best part was I didn’t even reply to the tool, it just piped up for no reason and got nothing right. But I agree, LOF needs to go.

It’s a bot! Lol

 

[rivermobster]

It’s a bot! Lol

As JBB would say...

"Satan's spawn"

 

[Racey]

View attachment 1375667

Lol it sounds to me like his distinction is that since the look up tables are all loaded out of the non volatile memory and into RAM on startup they cease to be look up tables any more.

 

[Cray Paper]

This thread has been so informative in so many ways! I had no idea AI was this advanced. There may be a human directing EN's responses, but NFW it is a human being responding.
Scary none the less

 

[Bigbore500r]

Lol it sounds to me like his distinction is that since the look up tables are all loaded out of the non volatile memory and into RAM on startup they cease to be look up tables any more.

A picture is worth a thousand words…

 

[DaveH]

I’m wondering if this timing control offered by mefi is smart enough to keep soot off the transom if you run 91 octane in your merc engine. Perhaps @ElectroNautical can shed some light on this.

 

[4Waters]

I’m wondering if this timing control offered by mefi is smart enough to keep soot off the transom if you run 91 octane in your merc engine. Perhaps @ElectroNautical can shed some light on this.

Turn the fuel pressure down

 

[Racey]

A picture is worth a thousand words…

View attachment 1375737

Omg dude

 

[Outdrive1]

A picture is worth a thousand words…

View attachment 1375737

I actually laughed out loud at that