Get Longevity.V2.45 for the Cummins ISV Now!
301 - 320 of 348 Posts
Editor
Oh my goodness... my efforts on data validation seem like they're taking forever. As one thing works, another breaks... It doesn't make sense!
Very strict data validation is extremely important and I will have layers of it, just to be on the safe side. Let's take one of our very simplest data types, an integer.
An integer is stored in the calibration as either one, two or four bytes in length. A one byte, unsigned integer can only have a value of 0 to 255. It can't have any decimal places, it can't be negative. It can only be a whole number. A two byte integer would have the range of 0-65535, but still be a whole number with no decimal place. So when using the editor, if I'm editing a parameter that is a one-byte unsigned integer, it must keep me within that range. On top of that, there are some cases where we know the minimum and maximum allowed values and in those cases I also want to ensure that the value does not cross this boundary.
Another issue is, that even computers with microprocessors are flawed with the same problem... the person operating it. If I don't add layers of data validation and test to make sure errors don't slip through the cracks, something as simple as putting 256 as a value where 255 was the maximum would be very bad. A computer is stupid and also very helpful. So if the parameter was supposed to be a one byte integer, and I erroneously put in 256, well, the computer would helpfully add an extra byte to the result so that 256 would fit. This in turn would shift every following byte off by one. The effects of which would be catastrophic at best.
I have data validation working for strings, signed and unsigned integers, hexadecimals from one to four bytes, and floating point numbers. I still need to add min max validation for the values that have it, then address actually storing the changes and submitting them to the server.
Oh my goodness... my efforts on data validation seem like they're taking forever. As one thing works, another breaks... It doesn't make sense!
Very strict data validation is extremely important and I will have layers of it, just to be on the safe side. Let's take one of our very simplest data types, an integer.
An integer is stored in the calibration as either one, two or four bytes in length. A one byte, unsigned integer can only have a value of 0 to 255. It can't have any decimal places, it can't be negative. It can only be a whole number. A two byte integer would have the range of 0-65535, but still be a whole number with no decimal place. So when using the editor, if I'm editing a parameter that is a one-byte unsigned integer, it must keep me within that range. On top of that, there are some cases where we know the minimum and maximum allowed values and in those cases I also want to ensure that the value does not cross this boundary.
Another issue is, that even computers with microprocessors are flawed with the same problem... the person operating it. If I don't add layers of data validation and test to make sure errors don't slip through the cracks, something as simple as putting 256 as a value where 255 was the maximum would be very bad. A computer is stupid and also very helpful. So if the parameter was supposed to be a one byte integer, and I erroneously put in 256, well, the computer would helpfully add an extra byte to the result so that 256 would fit. This in turn would shift every following byte off by one. The effects of which would be catastrophic at best.
I have data validation working for strings, signed and unsigned integers, hexadecimals from one to four bytes, and floating point numbers. I still need to add min max validation for the values that have it, then address actually storing the changes and submitting them to the server.
Editor
Progress has ground to a halt. I've had a number of other things to deal with and I haven't had time to touch the editor in a couple days.
Planning the next versions
I'm getting a number of km on my truck and some helpful datalogs from customers (thank you! I genuinely appreciate you taking the time to send them to me!). I want to make some changes to the charge air map and add a bit more math to how I calculate the charge air maps.
My primary goals here are to tighten up the fuel/boost relationships. The second, although almost as important, is to better calculate the boost curve that results from gear changes. Ideally, I want the engine rpm and fuel load to drop back into a place where there is the least amount of change in charge air pressure and rotary valve position.
That's going to take a bit of learning, as I'm just eyeballing gradients right now, so... time to learn some more maths so I can parametrically general gradients.
Planning V2.38
As close to stoichiometric as possible at all points on the map. Tuned for economy. It will be a bit lethargic on the throttle response side and likely will also result in more transmission hunting at low rpm.
Planning V2.39
This will be my first attempt at a high performance tune. Tuned to shred tires and wallets. It will not have the hp or torque turned up. It's going to run a bit on the lean side to keep everything prespooled, the high pressure turbo is going to be bouncing around a bit. This will be longevity for people who don't care about fuel costs.
Long Term Planning
The businessman in me is realizing that I'm eventually going to need to find ways to make more money. I am looking into assembling complete diesel diet kits, the can bus sniffer will eventually result in a monitoring system. I'm also eventually going to build a headache rack controller, so it can have nice led lighting, strip lighting, etc, and have it listen to the can bus and behave like a component of the truck instead of having to splice into different sub systems. I have many plans for our truck!
Once I'm confident in my ability to restrain the turbos, I've decided I'm going to go back on something I said... I will eventually have a series of towing specific tunes with increased torque. I'll be aiming for around 700ft.lbs at the crank. This is not going to happen until I have my editor built, and it's also going to be expensive. Longevity will always be the entry point for my tunes.
If someone decides they want to turn their engines up, they're usually willing to spend more money on performance parts and equipment. Since a person will need to have longevity in order to get the towing tune, I can make more money on the towing tune without having to give 2/3rds of it to ez lynk.
The Pause
My wife has nearly convinced me to just eat the losses in disability benifits and get my tune on more trucks. Her arguments are compelling. I'm going to give it a few more days while I'm dealing with my case manager.
Progress has ground to a halt. I've had a number of other things to deal with and I haven't had time to touch the editor in a couple days.
Planning the next versions
I'm getting a number of km on my truck and some helpful datalogs from customers (thank you! I genuinely appreciate you taking the time to send them to me!). I want to make some changes to the charge air map and add a bit more math to how I calculate the charge air maps.
My primary goals here are to tighten up the fuel/boost relationships. The second, although almost as important, is to better calculate the boost curve that results from gear changes. Ideally, I want the engine rpm and fuel load to drop back into a place where there is the least amount of change in charge air pressure and rotary valve position.
That's going to take a bit of learning, as I'm just eyeballing gradients right now, so... time to learn some more maths so I can parametrically general gradients.
Planning V2.38
As close to stoichiometric as possible at all points on the map. Tuned for economy. It will be a bit lethargic on the throttle response side and likely will also result in more transmission hunting at low rpm.
Planning V2.39
This will be my first attempt at a high performance tune. Tuned to shred tires and wallets. It will not have the hp or torque turned up. It's going to run a bit on the lean side to keep everything prespooled, the high pressure turbo is going to be bouncing around a bit. This will be longevity for people who don't care about fuel costs.
Long Term Planning
The businessman in me is realizing that I'm eventually going to need to find ways to make more money. I am looking into assembling complete diesel diet kits, the can bus sniffer will eventually result in a monitoring system. I'm also eventually going to build a headache rack controller, so it can have nice led lighting, strip lighting, etc, and have it listen to the can bus and behave like a component of the truck instead of having to splice into different sub systems. I have many plans for our truck!
Once I'm confident in my ability to restrain the turbos, I've decided I'm going to go back on something I said... I will eventually have a series of towing specific tunes with increased torque. I'll be aiming for around 700ft.lbs at the crank. This is not going to happen until I have my editor built, and it's also going to be expensive. Longevity will always be the entry point for my tunes.
If someone decides they want to turn their engines up, they're usually willing to spend more money on performance parts and equipment. Since a person will need to have longevity in order to get the towing tune, I can make more money on the towing tune without having to give 2/3rds of it to ez lynk.
The Pause
My wife has nearly convinced me to just eat the losses in disability benifits and get my tune on more trucks. Her arguments are compelling. I'm going to give it a few more days while I'm dealing with my case manager.
CAN Bus sniffer
Today I took a break from everything else and assembled the can bus sniffer! I still have to write a program for it… but it has a working volt meter! The rj45 connector goes to the obd2 port. I don’t know how many of the pins nissan actually uses so it’ll be a bit of trial and error. I've also included the cummins tri pin connector so I can start to watch the ecms can bus. I've left it optional that I can power it by usb and watch the serial bus directly, so eventually I can datalog with my computer in the truck. Or it can be powered by the obd2 connector. This will be useful when I put gauges onto it, like a rotary valve position gauge.
![Passive circuit component Circuit component Hardware programmer Electrical wiring Resistor]()
Today I took a break from everything else and assembled the can bus sniffer! I still have to write a program for it… but it has a working volt meter! The rj45 connector goes to the obd2 port. I don’t know how many of the pins nissan actually uses so it’ll be a bit of trial and error. I've also included the cummins tri pin connector so I can start to watch the ecms can bus. I've left it optional that I can power it by usb and watch the serial bus directly, so eventually I can datalog with my computer in the truck. Or it can be powered by the obd2 connector. This will be useful when I put gauges onto it, like a rotary valve position gauge.
Business Status
Well in order to deal with the large suppliers so I can get proper warehouse prices on things like the ez lynk, exhaust components, etc, etc. I need to register for tax collection both Federally and Provincially. Which also means I have to deal with registering my business at a provincial level. I haven't done that because federal was enough to satisfy stripe. I'm completely tapped out. I can't keep putting money into the system like this.
Well in order to deal with the large suppliers so I can get proper warehouse prices on things like the ez lynk, exhaust components, etc, etc. I need to register for tax collection both Federally and Provincially. Which also means I have to deal with registering my business at a provincial level. I haven't done that because federal was enough to satisfy stripe. I'm completely tapped out. I can't keep putting money into the system like this.
Editor
I've been building a simple spreadsheet interface in order to display the maps and tables. Getting something as simple as clicking, dragging and selecting a range of cells is quite a bit of work! Progress is progress and I'm getting closer to a working editor. It's quite a fascinating process writing an interface.
I would like to add the ability to have calculated tables, but for now my primary goal is to be able to cleanly cut and and paste cells from excel to my editor and vice versa. My previous attempt only allowed changing one cell at a time. That can be quite tedious and error prone.
I've been building a simple spreadsheet interface in order to display the maps and tables. Getting something as simple as clicking, dragging and selecting a range of cells is quite a bit of work! Progress is progress and I'm getting closer to a working editor. It's quite a fascinating process writing an interface.
I would like to add the ability to have calculated tables, but for now my primary goal is to be able to cleanly cut and and paste cells from excel to my editor and vice versa. My previous attempt only allowed changing one cell at a time. That can be quite tedious and error prone.
Available!!!
Well I can't keep doing nothing at all. I'm not getting clear answers from the worksafe people. So hopefully I can sell enough tunes to make a go of it, regardless what they decide.
Experimental.V2.39
Quite a large increase in boost, this should make sure the engine is always pre-spooled. Rotary valve positions have been tightened up, along the shape of the charge air map. I have also added a minimum soft limit to the rotary valve. In theory this should altogether prevent the rotary valve from trying to build boost through excessive clamping. This might also be sufficient to help mitigate the shaft rpm spikes on slack throttle.
Horsepower and torque have not been increased. This version of longevity is aimed towards very strong throttle response. I did stare at the torque curve and thought very hard about turning it up a bit. I intend to prove that the turbocharger system on this engine can't be beat and that it doesn't need excessive fuel to be an absolute monster.
This revision will use more fuel, no question. I will install it on my truck tomorrow and validate it. Any existing customer that would like access to this experimental build, let me know and I'll push it to you once I know it's safe enough for extended trials.
Well I can't keep doing nothing at all. I'm not getting clear answers from the worksafe people. So hopefully I can sell enough tunes to make a go of it, regardless what they decide.
Experimental.V2.39
Quite a large increase in boost, this should make sure the engine is always pre-spooled. Rotary valve positions have been tightened up, along the shape of the charge air map. I have also added a minimum soft limit to the rotary valve. In theory this should altogether prevent the rotary valve from trying to build boost through excessive clamping. This might also be sufficient to help mitigate the shaft rpm spikes on slack throttle.
Horsepower and torque have not been increased. This version of longevity is aimed towards very strong throttle response. I did stare at the torque curve and thought very hard about turning it up a bit. I intend to prove that the turbocharger system on this engine can't be beat and that it doesn't need excessive fuel to be an absolute monster.
This revision will use more fuel, no question. I will install it on my truck tomorrow and validate it. Any existing customer that would like access to this experimental build, let me know and I'll push it to you once I know it's safe enough for extended trials.
Editor
Made huge progress on selecting cells. I can now select all by keystroke, or I can select a rectangular range with click and hold. That’s about the most important functions so now I have to work on how things are actually copied to the clip board and how I can manage copy/pasting.
Made huge progress on selecting cells. I can now select all by keystroke, or I can select a rectangular range with click and hold. That’s about the most important functions so now I have to work on how things are actually copied to the clip board and how I can manage copy/pasting.
Experimental.V2.39 +30km
Well that was completely not the results I was expecting. Good results but back to the drawing board. Mapping the lower limit has worked wonders but I was so aggressive in my mapping that the high pressure turbo is non-existant in the idle zone. It has also taken away the cold idle clamping, so I now know the lower limit table overrides everything. Tonight I’ll prepare v2.40.
Well that was completely not the results I was expecting. Good results but back to the drawing board. Mapping the lower limit has worked wonders but I was so aggressive in my mapping that the high pressure turbo is non-existant in the idle zone. It has also taken away the cold idle clamping, so I now know the lower limit table overrides everything. Tonight I’ll prepare v2.40.
Preparing V2.40
Studying the datalogs for V2.39, it completely eliminates exhaust back pressure and boost spikes. It's not responsive enough right now, that it actually causes the ecm to go rich to build boost. That's no good. That tells me I have the lower limit too high. I'm going to slowly work on bringing the responsiveness back up until I reach a reasonable balance. The one thing with having the turbos be very responsive, it really makes the truck way more comfortable to drive and really makes the transmission be civilized.
I've lowered peak boost, because it really isn't needed since I'm not turning up the HP. Instead I'm working on improving the responsiveness. It's very likely I'll stumble on a good economy configuration as I'm trail and erroring a few different map setups.
Studying the datalogs for V2.39, it completely eliminates exhaust back pressure and boost spikes. It's not responsive enough right now, that it actually causes the ecm to go rich to build boost. That's no good. That tells me I have the lower limit too high. I'm going to slowly work on bringing the responsiveness back up until I reach a reasonable balance. The one thing with having the turbos be very responsive, it really makes the truck way more comfortable to drive and really makes the transmission be civilized.
I've lowered peak boost, because it really isn't needed since I'm not turning up the HP. Instead I'm working on improving the responsiveness. It's very likely I'll stumble on a good economy configuration as I'm trail and erroring a few different map setups.
Engine Zones
I took one of my fresh airflow maps, did a quick copy and paste and a bit of editing to roughly show where the different power zones are. You'll frequently see me mention the idle zone. I haven't properly shown what I've meant by this, until now.
![Colorfulness Rectangle Product Font Line]()
I took one of my fresh airflow maps, did a quick copy and paste and a bit of editing to roughly show where the different power zones are. You'll frequently see me mention the idle zone. I haven't properly shown what I've meant by this, until now.
V2.40 +60km
After reviewing the datalogs I can come to only one conclusion. I have boost and shaft rpm spikes completely under control. I saw no more than 130,000rpm on the high pressure turbo, max allowable is 165,000. I saw a maximum exhaust back pressure of 44.2psig and the maximum allowable is 65psig.
This is unbelievably good news, but also shoots me in the foot. The closer I get to the most efficient charge air, fresh air and turbine mapping, the closer I get to the most performant version. I can't separate efficiency and performance for longevity, because they're the same. So until I release turned up high performance and towing tunes, longevity will remain the only tune.
V2.40 is too aggressive in the prespooling which is actually holding the engine back. So far it appears v2.37 has a nearly perfect charge air map. The other issue I'm having with so much additional prespooling. The ecm is willing to produce significant boost below 1000rpm and that's actually confusing the transmission. The engine doesn't seem to lug down there, but it doesn't sound happy to be putting out that much torque at 950rpm. v2.37 is almost a perfect balance of available bottom end torque, with a sharp enough lead in from idle that the transmission has a clear distinction line where there isn't enough power and it needs to gear down.
Preparing V2.41
Use v2.37 charge air map, recalculate high pressure turbine bypass lower limit to v2.37 charge air. Allow max clamping 600-1000 rpm to ~24mm3 so cold warm up cycle can function correctly. Recalculate commanded tbv position and fresh air tables.
Longevity.V2.37 - Current
Will be superseded and withdrawn from service. The next released version of longevity will be a recommended update.
The Transmission
While I have zero access to the tcm, it's clear that the tcm and the ecm communicate. The way the engine is tuned directly effects the behaviour of the transmission. My excessive prespooling also introduced an interesting artifact... it started to shift hard again, despite me driving like aunt martha.
After reviewing the datalogs I can come to only one conclusion. I have boost and shaft rpm spikes completely under control. I saw no more than 130,000rpm on the high pressure turbo, max allowable is 165,000. I saw a maximum exhaust back pressure of 44.2psig and the maximum allowable is 65psig.
This is unbelievably good news, but also shoots me in the foot. The closer I get to the most efficient charge air, fresh air and turbine mapping, the closer I get to the most performant version. I can't separate efficiency and performance for longevity, because they're the same. So until I release turned up high performance and towing tunes, longevity will remain the only tune.
V2.40 is too aggressive in the prespooling which is actually holding the engine back. So far it appears v2.37 has a nearly perfect charge air map. The other issue I'm having with so much additional prespooling. The ecm is willing to produce significant boost below 1000rpm and that's actually confusing the transmission. The engine doesn't seem to lug down there, but it doesn't sound happy to be putting out that much torque at 950rpm. v2.37 is almost a perfect balance of available bottom end torque, with a sharp enough lead in from idle that the transmission has a clear distinction line where there isn't enough power and it needs to gear down.
Preparing V2.41
Use v2.37 charge air map, recalculate high pressure turbine bypass lower limit to v2.37 charge air. Allow max clamping 600-1000 rpm to ~24mm3 so cold warm up cycle can function correctly. Recalculate commanded tbv position and fresh air tables.
Longevity.V2.37 - Current
Will be superseded and withdrawn from service. The next released version of longevity will be a recommended update.
The Transmission
While I have zero access to the tcm, it's clear that the tcm and the ecm communicate. The way the engine is tuned directly effects the behaviour of the transmission. My excessive prespooling also introduced an interesting artifact... it started to shift hard again, despite me driving like aunt martha.
Longevity.V2.37 - Deprecated
Has now been superseded, and remains available to help me see who has what installed.
Longevity.V2.41 - Current +40km
Longevity has been updated. This is a recommended update.
Shaft rpm spikes, boost spikes are now completely under control. V2.40, with its more aggressive map worked a bit better with the handoff to the low pressure turbo, so with V2.41 there is a small blip that you can notice, if you're looking for it, during the hand off at around 2500rpm. I need more data to solve that.
Driving experience is about the same as v2.37, but the turbocharger response is just a little bit softer which makes the transmission gear down just a little bit sooner. Overall I'd say transmission behaviour is better with v2.41.
In theory V2.41 should solve the majority of oscillation issues when pulling weight, but I need some data to verify.
The cold warm up cycle is properly enabled again and should help warm up the engine faster. It is important to note, that the engine is going to always behave erratically when it's cold as it jumps back and forth between fuel/exhaust throttling. That's the nature of the beast and it's not worth trying to chase it down when any diesel engine should be allowed to warm up before getting any real throttle input.
Has now been superseded, and remains available to help me see who has what installed.
Longevity.V2.41 - Current +40km
Longevity has been updated. This is a recommended update.
Shaft rpm spikes, boost spikes are now completely under control. V2.40, with its more aggressive map worked a bit better with the handoff to the low pressure turbo, so with V2.41 there is a small blip that you can notice, if you're looking for it, during the hand off at around 2500rpm. I need more data to solve that.
Driving experience is about the same as v2.37, but the turbocharger response is just a little bit softer which makes the transmission gear down just a little bit sooner. Overall I'd say transmission behaviour is better with v2.41.
In theory V2.41 should solve the majority of oscillation issues when pulling weight, but I need some data to verify.
The cold warm up cycle is properly enabled again and should help warm up the engine faster. It is important to note, that the engine is going to always behave erratically when it's cold as it jumps back and forth between fuel/exhaust throttling. That's the nature of the beast and it's not worth trying to chase it down when any diesel engine should be allowed to warm up before getting any real throttle input.
Quick question, I have my old AFE scorcher piggy back that has 0, 15, 30, and 60hp boost. How would that workl with your tune. Also, With 2.37 my transmission shifts mucn better but it does downshift earlier when I want tp use thw low rpm torque, so with 2.41 i noticed you said it will downshifr sooner. One of the big problems I had with my GDP tubes is that the trans always short shifted from 3rd to 4th even under boost and moderate throttle, which I hated because it wpuld shift hard and you would hear a chuffing sound from the intake. It had such a delay between me pushing the throttle amd it responding. If I was in 3rd and wanted more power id hit the gas, it would upsbift too early then downshift amd take off. Lkke I said it was like it had a big lag. Your tune has really done qonders for my transmission shifting but I dont want it to downshift too soon under light throttle, although that would help with towing. I am going to try the new tune and see how it behaves. I'll report back. I have noticed that it seems lkke the active torque management really cuts power in the lower gears and it really doesnt wake up yill 3rd or 3th gear, so i have to really watch pulling i to traffic and merging. Thats really the only drawback I can think of. When my lrevious GDP tunes were at 0hp setting it still pulled good in the lower gears. Do you have a way to disable the torque management so the truck will not be so slow in 1st and 2nd gear at full throttle? Because other than that the tune is great! I absolutely love how much better my transmission shifts since installing your tune!. One more thing, when the truck is off does the egr fully close? I guess I could just loom at my ezlynk yo see the egr command, that way I don't unplug it with it open or partially open. Thank you again for all of the worl you have put into the tune. My fuel economy is better but once i. Up to 75 its about the same maybe a little better which coincides with what you said about better fuel economy ay 60mph and below. When accelerating from a stop I do notice the fuel economy bar does seem dip lower than I remember, even under light throttle, but I can't be too sure about that one. That being said I love what you've done. Just the better transmission shifting alone makes this tine worth it. It no longer is reluctant to downshift and doesnt short shift from 3rd to 4th when im trying to accelerate like my old tune did. Again, I cant express how much I appreciate ehat youre doing for us.
Piggy backs, or other components that are added inline with the ecm, I have no idea what they're doing or what effect it would have on my tune. I would imagine those devices are designed around a stock calibration, so I cannot recommend using them. Those type of devices tend to lie to the ecm, that's not a good idea in my opinion.
EGR system should be unplugged or ideally removed. I personally have not removed my egr, I don't even have the spare cash for @Wheeler 's egr kit and I have to admit I've been thinking about it pretty hard. The EGR system should be removed on a deleted engine. There is no other recommendation.
Technically speaking the egr manager is still working in my calibration. Even in the non-egr mode directions are sent to the egr valve, as simple of things as cycling it to maintain function. In this case, do as I say, not as I do.... get rid of the egr cooler. The egr parked position should be fully closed.
In terms of downshift, with v2.37 I found there was a few places where I was adding more throttle than I was comfortable with, especially around 1100rpm when it would happily give 15+psig of boost. v2.40 was even more aggressive, and started getting into lugging territory. V2.41 feels right to me. It will build up to 15psig of boost at 1100rpm. If you roll onto the throttle it will build boost instead of gearing down. If it's a rapid throttle change and a large enough change it will gear down but it feels much more seamless. If you did a rapid throttle change in v.2.37 it would sometimes still just add boost instead of gearing down in places where it should have geared down. In v2.40 it would hold the gear much too long and struggle to recover once it decided 950rpm was too low to be in gear.
The fuel economy bar is dipping lower on throttle input in the low rpm because the engine is able to burn more fuel down there, so it does.
I can't express enough how much I appreciate you and every other people who has purchased and installed this tune. I have the chance to live the dream and it's because of people like you. So thank you. I'm glad you enjoy my efforts as much as I have getting to this point.
You just nailed it on the rolling into the throttle. I wasn't aware of the egr kit you mentioned. I will message him. Thanks again!
Had a quick look at the logs, your engine appears very happy! Glad you enjoy it! Thank you so much for sending me the logs, they're very helpful!
301 - 320 of 348 Posts
