When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Here is a plot off the same motor but running Motec. On the left is the Target AFR verse on the right is the Actual AFR. This motor was running totally open loop in this log. In theory, both sides should look identical on a perfectly tuned motor.
Thanks guys. I'll give Megalog Viewer a shot. Are there any other types of graphs that you guys use regularly when tuning? The better I get at interpreting data the better I can be at tuning.
Post up about 20 min of data off your car and let me take a look at the data. There are some tricks that I have not covered yet in the interest of keeping it simple to start. It really helps to have MAP, TPS, RPM, I am spending most of my spare time at night trying to get the B18R up and running.
It really helps to have....
MAP (manifold air pressure)
TPS
RPM
Pulse Width or Duty Cycle
Wide Band AFR/Lambda or Narrow Band Voltage
VSS (speed)
CTL (engine temp
IAT (air temp)
Log Time
MAF (mass air flow) if equipped
Long Term Fuel Trims
Short Term Fuel Trims
System Volts
Ignition Timing
Knock (in some form)
Knock Timing Adjust
With this list, you can figure out "almost" anything.......
I got my setup running cheap..but it goes on and on..
(These prices are in CAD)
$20 for chip burner
$50 for chipped ecu
$20 for spare chipped ecu
$0 modifying chipped ecu
Started with $0 crome then crome pro
$240 for wb02 and gauge
$??? (Can't say) for eCtune Tuner Edition
Don't have an emulator yet but have plans to build one, may get a demon or ostrich but I want hondata. Neptune/demon is a great option too.
eCtune is about as good as it gets for burnt chips tho..find it way better than crome tbh
Here is where DIY tuning takes an entire new turn. First we must look at the fueling equation that is the basis of almost every EFI system in the world. The megasquirt name for all of these pieces is in red. The names change from one system to the next but the concepts are all the same.
You can see that each piece of the formula is multiplied (or added at the end) to make up the Injector Pulse Width.
The two biggest and most dynamic pieces of this are the RPM piece and the MAP piece. Everything else in the formula is relatively constant.
Once everyone starts to get their head around this formula, we can take this DIY tuning up to the next level where it really get fun.
This is something I put together about a year ago. Feel free to ask any questions about this stuff. I promise it gets more clear as you start to catch on.
Very cool. Those parameters in Megasquirt look to be just about the same as those found in ECU Flash. I wish this was posted sooner. It would have made my learning curve a lot less steep. Thanks for sharing!
On a side note, I notice many tuners don't review data via scatter graphs or even using some of the other line graphs listed in your tutorial. They simply tune on the dyno and rely on AFR, EGT, boost (for turbo applications), HP, and torque graphs. I understand reviewing all the data is time consuming, but it leaves me to wonder just how much power is being left on the table.
Last edited by pogeeboy27; 08-19-2016 at 08:57 AM.
Very cool. Those parameters in Megasquirt look to be just about the same as those found in ECU Flash. I wish this was posted sooner. It would have made my learning curve a lot less steep. Thanks for sharing!
On a side note, I notice many tuners don't review data via scatter graphs or even using some of the other line graphs listed in your tutorial. They simply tune on the dyno and rely on AFR, EGT, boost (for turbo applications), HP, and torque graphs. I understand reviewing all the data is time consuming, but it leaves me to wonder just how much power is being left on the table.
Tuning with a dyno is great for things like designing headers or cams where you do back to back testing under very controlled and comparable conditions. They are bad at duplicating the real world conditions of changing altitude, air temperature, corning force, launch loads, constantly changing throttle position, shifts......
The Megasquirt world is made up of a mix of diehard gear heads mixed in with big time geek computer programmers. It is amazing when you see the final results of those two groups. I spend 3 hours a day helping people tune stuff with graphing software developed to tune motors. Most people do not have free access to a dyno like I happen to have. Even with that free access, I do 95% of my own tuning with data logs. If it used to take me 3 sec to sweep from 2000 rpm to 8000 RPM and now I do it in 2.3 sec, I am making more power. I am a racer and I simply don't care what the new HP is. If a throttle stab instantly results in a steady climb in RPM with no delay, the AE setting are really close. If the RPM falls first, you have AE issues. I strive to have my racecar to have street like manners even in the pits. It can only be done if I have a handle on the AFR thru the entire operating range. Not just the WOT part of the power band where tuning is easy. The devil is in the transitions and driveability.
This is the sort of thing that I love. When the owner of the car calls or emils back that his motor simply rips off corners now. I am shooting to get my new B18C motor running before the end of the season. I will post screen shots as I work thru the issues that always show up with a new install.
Regarding settings in your tuning SW, I work with lots of different SW packages. They really are all very similar. It is all the same basic math. The names simply change. If the owner can get data logs, I can teach the owner to tune it. For me, the fun part of racing is tuning. When my kid climbs out with a smile on his face and claims it was perfect and did everything he requested the car to do.......
As a side note, I have 3 axis accelerators tied into my Megasquirt so I can also do geez overlays. This is what is known as the traction circle. That car is just touching 1.5 geez cornering and braking. About .56 geez coming of corners. That is on a typical asphalt parking lot on a 95 degree day.
It reads almost any CSV data file. There unregistered version will read the first 500 or so records. You need the HD version. Google Megalogviewer HD download.
From the time I nailed the throttle till the time the car was at full acceleration was .385 sec as seen at the bottom of the screen in white. The max RPM per sec was at .161 sec after the throttle was rolled on.
Feel free to PM me far an email address and I will be happy to take a look at your data.
I have been playing around this weekend with data that people reading this thread have sent me. We will be able to DIY tune at least Megasquirt, eCtune and FlashPro logs.
Here is a eCtune histogram. It is showing the average AFR in each VE table MAP vs RPM. I changed the RPM and MAP break points to even numbers like the tuning software may have. These numbers are totally adjustable so it is best to have them the same to make tuning your VE table way easier. I like to have the WOT AFR set to 13.0 and idle and low power AFR targets set to about 14 AFR to start. Later we can get into getting settings better for fuel milage.
The eCtune system is logging data at about 5 samples per sec. That is fairly slow. When tuning this way, more data is always better so driving for a couple of hours would be better at getting good sampling.
I've been using a relatively old/slow laptop for logging since it has a LPT port I can use for chip burning.
I may start using my Macbook Pro for logging instead, as the old laptop overheats sometimes when driving around, though having two laptops in the car is a PITA.
I'm wondering if that'll help the log rate, the baud rate is 36400bps so I should be able to get a maximum of 4.5kB/s from the ECU, I have no idea what format the logs come through but assuming a binary format and 32 256bit parameters it should be able to do almost 1000 samples per second, but that's just looking at theoretical bandwidth and not actually knowing eCtunes protocol for communication or factoring in how much resources are available on the actual ECU dedicated for logging output?
Hopefully I can get a bit better sample rate, I hadn't realized it was so low.
It does, but it's just a usb with a ferrite bead. Nothing special.
theres nothing great about the logging rates on these old obd1 setups, the ecus are all AT LEAST 21 years old now and the boards are piggy backs on top of those ecus I would love to have kHz+ logging and be able to pick out individual misfires and crank position, but the hardware is pretty limited.
08-18-2016, 03:39 AM
Re: DIY tuning, a discussion
