How many times, 2,20,100, anyone know?
Get Longevity.V2.45 for the Cummins ISV Now!
281 - 300 of 348 Posts
It’s been hit or miss for me. I’ve had to reset my batteries at three installed tunes and north of ten. I have been installing so often, I just gave up for now and use the remote unlock and hold the key to the start button.
This is correct.
After every install the ecm and tcm go into a more aggressive learning mode. In theory it needs 400km to relearn everything.
The other thing is that every single time you flash your ecm there is a risk of bricking it. If your batteries die mid flash, your kid trips and pulls the unlock cable hanging from the side of the truck, or any number of other possibilities. The installation process is volatile as any interruption can brick the ecm.
The main goal of the longevity tune is to be a well rounded enough tune that it’s the last one ever installed. It’s going to take a good year of collecting feedback to get longevity to that point. I can’t get it there on my truck alone.
Editor
The editor has become a critical project, I need it to access the low pressure turbine maps.
Back on the interface. Bit of a slog. Trying to put it together as a react.js application. I've had bad luck with it before. I'm approaching it with much more experience in javascript, so I'm making a bit better progress this time. The css isn't beating me up as badly either.
Longevity Branching
It's not going to happen until I get my editor working but I've decided to branch Longevity into two versions. Hypermiller and Leadfoot. They both will still be 0hp and softened torque. Leadfoot will has less economy but will respond very aggressively to throttle inputs. Hypermiller will be a bit more lathargic but strive for economy.
Any longevity branch will be available to permanent longevity owners.
Longevity.V2.37 - Current
So far the feedback is still coming back positive, so I'm gaining confidence v2.37 will continue being the current release for the foreseeable future.
The editor has become a critical project, I need it to access the low pressure turbine maps.
Back on the interface. Bit of a slog. Trying to put it together as a react.js application. I've had bad luck with it before. I'm approaching it with much more experience in javascript, so I'm making a bit better progress this time. The css isn't beating me up as badly either.
Longevity Branching
It's not going to happen until I get my editor working but I've decided to branch Longevity into two versions. Hypermiller and Leadfoot. They both will still be 0hp and softened torque. Leadfoot will has less economy but will respond very aggressively to throttle inputs. Hypermiller will be a bit more lathargic but strive for economy.
Any longevity branch will be available to permanent longevity owners.
Longevity.V2.37 - Current
So far the feedback is still coming back positive, so I'm gaining confidence v2.37 will continue being the current release for the foreseeable future.
Editor
I've made huge progress over the last two days on the editor. A fair amount of the react.js stuff is coming back to me. This time I don't seem to be struggling as badly with the css and the react components as I was last time.
I have all the single value parameters working nicely, highlighted when they've been editied, normal when not. I've got tables and y axis tables forming correctly. There is some big in the air questions about how to impliement copy and paste like a spreadsheet. I want to get all the parameters working first before worrying about smaller things like that, because with this current editor I can actually edit each of the cells as I so desire.
There is still an absolute ton of development work to do. This is all just the interface, but I'm following rawzes recommendation to start from the interface side because no matter how many times I modelled the intel hex and the ecfg, bring them together often over whelmed me.
I've made huge progress over the last two days on the editor. A fair amount of the react.js stuff is coming back to me. This time I don't seem to be struggling as badly with the css and the react components as I was last time.
I have all the single value parameters working nicely, highlighted when they've been editied, normal when not. I've got tables and y axis tables forming correctly. There is some big in the air questions about how to impliement copy and paste like a spreadsheet. I want to get all the parameters working first before worrying about smaller things like that, because with this current editor I can actually edit each of the cells as I so desire.
There is still an absolute ton of development work to do. This is all just the interface, but I'm following rawzes recommendation to start from the interface side because no matter how many times I modelled the intel hex and the ecfg, bring them together often over whelmed me.
Paused 
Well it never ever pays to be honest it seems. Due to me being on disability I'm required to report each sale as income. So, despite me still being in red ink in this endeavour, I lose the amount totalling my entire portion of the sale. I'm expected to take every cent of profit from the company and then I lose that amount in my disability cheque. So over night I've gone from slowly building the company towards black ink to losing money on the sale of every tune.
So for the time being I can't sell any more tunes, or I have to massively increase my prices to an unreasonable level. I have no idea how the forum owners will react if/when I drop my vender status. I don't have a time frame as to when I'll be able to sell the tune again. It is possible some of you will end up with a unicorn tune.
I won't be leaving my current customers in the dark. I have no idea what will happen to my account here, or this thread when I drop vender status. If anything happens to my account here, I'll still have my email address.
Well it never ever pays to be honest it seems. Due to me being on disability I'm required to report each sale as income. So, despite me still being in red ink in this endeavour, I lose the amount totalling my entire portion of the sale. I'm expected to take every cent of profit from the company and then I lose that amount in my disability cheque. So over night I've gone from slowly building the company towards black ink to losing money on the sale of every tune.
So for the time being I can't sell any more tunes, or I have to massively increase my prices to an unreasonable level. I have no idea how the forum owners will react if/when I drop my vender status. I don't have a time frame as to when I'll be able to sell the tune again. It is possible some of you will end up with a unicorn tune.
I won't be leaving my current customers in the dark. I have no idea what will happen to my account here, or this thread when I drop vender status. If anything happens to my account here, I'll still have my email address.
Man that is terrible! Utterly ridiculous! How do they expect anyone to live on disability alone?Paused
Well it never ever pays to be honest it seems. Due to me being on disability I'm required to report each sale as income. So, despite me still being in red ink in this endeavour, I lose the amount totalling my entire portion of the sale. I'm expected to take every cent of profit from the company and then I lose that amount in my disability cheque. So over night I've gone from slowly building the company towards black ink to losing money on the sale of every tune.
So for the time being I can't sell any more tunes, or I have to massively increase my prices to an unreasonable level. I have no idea how the forum owners will react if/when I drop my vender status. I don't have a time frame as to when I'll be able to sell the tune again. It is possible some of you will end up with a unicorn tune.
I won't be leaving my current customers in the dark. I have no idea what will happen to my account here, or this thread when I drop vender status. If anything happens to my account here, I'll still have my email address.
It’s not like I would mind, but I assumed they would treat it the same way Revenue Canada does. Revenue Canada doesn’t make a business pay taxes until they do $30,000 in sales. I’ve done $1800. If I did $30k in sales, the business could afford to pay me a little bit. They are only willing to accept that I have to pay ez lynk each time, but I even lose out on the exchange rate.
I mean I have to make customers wait until the money clears… now this. It’s just too much. I hope they back down, but it’s the government and they do whatever they want.
Perhaps it's better to adondon ship with the company amd just sell under the table. Only the government can somehow make selling something produce a loss. Not sure what the best method is for payment processing, but I hate to see you giving in to the government's schenanigans.
If I was giving in, I would completely pull the pin. I'm only pausing for now. I got approved for my account with Nissan and I have components for product development on the way that were purchased with what I thought was company money...
I can't sell to Americans under the table, only Canadians. Now that worksafe knows I sell the tune, I'm going to have to prove to them I'm not selling it anymore. So I shot myself in the foot. Again.
Editor
I've been slowly plugging away on rewriting my interface again and again. I get a little bit further ahead each time, but I still get a bit bound up when it comes to handling all the different types of data and events. onClickHandler, onChangeHandler, this needs useRef, this needs a useState... Sometimes it feels like it would be easier to manually edit the binary.
Pause continued.
Until I get a firm yes or no on the business plan I submitted to worksafe, I'm stuck in limbo. If they deny my business plan, then I won't be able to sell any more tunes until I'm on my own two feet again. That could be years. I'll still continue developing the tune and updating everyone that has it, because that's what I'm going to do for my own truck anyways.
Development Path
Again, I need my editor to work. I don't have access to the low pressure turbine tables in ez lynk and I don't want to spend the time to build a custom configuration for ez lynk when, once my editor works I'll have a harmonized editor for every parameter.
As much as I want to get to timing, the turbine maps need to be tuned better so that I have complete control over the high pressure turbo. I don't like that there are still a few places in v2.37 where the high pressure turbo can get away from the charge air manager and force snap position changes on the bypass valve.
I've been slowly plugging away on rewriting my interface again and again. I get a little bit further ahead each time, but I still get a bit bound up when it comes to handling all the different types of data and events. onClickHandler, onChangeHandler, this needs useRef, this needs a useState... Sometimes it feels like it would be easier to manually edit the binary.
Pause continued.
Until I get a firm yes or no on the business plan I submitted to worksafe, I'm stuck in limbo. If they deny my business plan, then I won't be able to sell any more tunes until I'm on my own two feet again. That could be years. I'll still continue developing the tune and updating everyone that has it, because that's what I'm going to do for my own truck anyways.
Development Path
Again, I need my editor to work. I don't have access to the low pressure turbine tables in ez lynk and I don't want to spend the time to build a custom configuration for ez lynk when, once my editor works I'll have a harmonized editor for every parameter.
As much as I want to get to timing, the turbine maps need to be tuned better so that I have complete control over the high pressure turbo. I don't like that there are still a few places in v2.37 where the high pressure turbo can get away from the charge air manager and force snap position changes on the bypass valve.
Editor
I can't believe how many hours I actually got into the editor today. The vast majority of it was spent on the input field for each parameter. The input field needs to handle to primary data types and a few subsets of them.
The data coming out of the calibration is human readable as:
An integer, which is a whole number.
A fixed point number, which is a number with a decimal and a specific number of places
A hexadecimal, which is a specific string representation of binary in base 16.
A string, which is any combination of utf-8 characters made from a byte array
Html allows you to set different data types for an input field. However managing the styling for numbers and strings is more of a pain than writing my own data validation and treating everything as text. This means my styling is always consistent.
I had to learn something called regular expressions. It's a powerful, yet extremely confusing way to tell a computer how to search through a string and decide if it matches what you're searching for, or not. Once I got that figured out, I could write a regular expression for each of the datatypes so that every keystroke can be tested to ensure compliance with the data type specified. This is important because if the configuration says that the unit and data type is a hexadecimal, the user input will only work if the correct characters are inserted. At first it seems counter intuative because it provides no visual feedback beyond refusing to accept an incorrect character. I'm hoping that the units being displayed directly beside the value will help clarify, but if not, I suppose I could just as easily highlight the units if I enter an invalid character.
My brain is mush, but this has been a good distraction from all the other problems I have no power over.
I can't believe how many hours I actually got into the editor today. The vast majority of it was spent on the input field for each parameter. The input field needs to handle to primary data types and a few subsets of them.
The data coming out of the calibration is human readable as:
An integer, which is a whole number.
A fixed point number, which is a number with a decimal and a specific number of places
A hexadecimal, which is a specific string representation of binary in base 16.
A string, which is any combination of utf-8 characters made from a byte array
Html allows you to set different data types for an input field. However managing the styling for numbers and strings is more of a pain than writing my own data validation and treating everything as text. This means my styling is always consistent.
I had to learn something called regular expressions. It's a powerful, yet extremely confusing way to tell a computer how to search through a string and decide if it matches what you're searching for, or not. Once I got that figured out, I could write a regular expression for each of the datatypes so that every keystroke can be tested to ensure compliance with the data type specified. This is important because if the configuration says that the unit and data type is a hexadecimal, the user input will only work if the correct characters are inserted. At first it seems counter intuative because it provides no visual feedback beyond refusing to accept an incorrect character. I'm hoping that the units being displayed directly beside the value will help clarify, but if not, I suppose I could just as easily highlight the units if I enter an invalid character.
My brain is mush, but this has been a good distraction from all the other problems I have no power over.
Your posts about tuning these trucks reinforces the idea of how modern vehicles are basically desktop computers that we drive around. Tell this to the dudes at Linus Media Group, and they'll watercool, overclock, and install RGB on an ECM and release a video by next week.
The ecm is dumb though. Like as dumb as a distributor on a small block chevy.
I've been getting quite a few positive reports on the throttle response with v2.37. Even my wife is saying positive things about it. However, it is showing at the pump for us. Having such quick access to the turbos does come with a fuel consumption penalty.
So, for you, is the throttle response worth it, or do you want better economy?
So, for you, is the throttle response worth it, or do you want better economy?
I prefer efficiency, but I also haven't noticed a change in that with V2.37. That is most likely because I accelerate like a 1950's tractor when I drive the truck. The fun driving is saved for the Mazda.
Editor
I've again, made more progress on the editor. I'll be getting back on it shortly but some things arrived in the mail that have me all excited!
CAN Bus Sniffer
Well I'm glad I showed a bit of restraint when I pressed the submit order button on the adafruit website. I was going to order two of the can m4 feathers, but instead only got one and decided against getting a bunch of other parts. I'm only just figuring out how to use the library, but the idea here is that I will passively observe the trucks can bus or the engine can bus. If I had the second board I could do both at the same time.
The idea is that I will learn the addresses of the different components then use that to build something that lets me watch their messages. For example, knowing what the commanded rotary valve position, fuel rate and rpm, could let me map the actual behaviour and response of the rotary valve. It's possible that I'll have to take in raw data then post process it.
Now that I have the display library working, have the basic sketch setup, I have to go back to working on the editor.
![Circuit component Passive circuit component Writing implement Hardware programmer Breadboard]()
I've again, made more progress on the editor. I'll be getting back on it shortly but some things arrived in the mail that have me all excited!
CAN Bus Sniffer
Well I'm glad I showed a bit of restraint when I pressed the submit order button on the adafruit website. I was going to order two of the can m4 feathers, but instead only got one and decided against getting a bunch of other parts. I'm only just figuring out how to use the library, but the idea here is that I will passively observe the trucks can bus or the engine can bus. If I had the second board I could do both at the same time.
The idea is that I will learn the addresses of the different components then use that to build something that lets me watch their messages. For example, knowing what the commanded rotary valve position, fuel rate and rpm, could let me map the actual behaviour and response of the rotary valve. It's possible that I'll have to take in raw data then post process it.
Now that I have the display library working, have the basic sketch setup, I have to go back to working on the editor.
My fuel economy is better than my previois tune. Throttle response is about the same, but I have a pedal commander. It does release throttle really sharply though. To the point it jerks when I let off the the gas pedal or cancel the cruise control. Other than that I have no issues. Does this tune tell the EGR to remaim closed? Because on my EZlynk im seeing different percentage commands to the EGR so should I unplug it?
Egr should be unplugged yes. Ideally removed.
I need to look into that a bit more, but the egr manager isn’t shut off, just neutered.
281 - 300 of 348 Posts
