My Megasquirt adventures over the years
Alright, I guess it's time for the next tab, fuel settings. This is probably the potatoes of the meat and potatoes of tuning. There are some fairly important settings in here and this is largely where I found the difference between a tune that runs "good" and one that runs "smoothly".
First up, we need to characterize our injectors. When dealing with a speed density tune, these are some of the more important numbers you can find. To reiterate something I heard Greg Banish say in an interview, any errors that you make at this point are just going to get baked in and compound things down the road. Don't let that scare you, it's not like your engine won't run or will immediate blow up if you get things wrong. It's more that you might end up chasing your tail trying to figure out why your tune feels like it changes day to day if you get things way off here. In reality, this is the "fine tuning" portion that I found makes my engine feel much smoother and more OEM like. The biggest problem is finding good data for your injectors. OEM stuff can be hard to find, aftermarket maybe a little easier, though you may need to be wary of any values you get. If you're really interested, the Minnoxide interview with Greg Banish on YouTube is a really interesting watch.
So anyway, we're going to start off with injector dead time settings. Injectors are ultimately just little solenoids. They use electricity to produce a magnetic field to move something out of the way to allow fuel to flow. "Dead time" as Megasquirt calls it, is the time between sending a signal to an injector and when fuel actually starts flowing. This can also be called offset or a number of other things. It's effectively the amount of time it takes the pintle/disk/whatever inside in the injector to move out of the way. The voltage you run them at will affect time timing as lower voltages will produce lower magnetic fields, which causes the injectors to take longer to open. I pulled these values out of my HP Tuners file, but had to do a lot of math and head scratching to end up with what I think are the "correct" numbers for my injectors.
You have the option of setting up to four curves. Generally speaking, I think you would probably only do this if you were running staged injection (so for instance, one set of injectors that runs at low rpm and another that kicks on at high rpm). Or, if you are running dual fuel and have one set that runs your pump gas stuff and another that runs your methanol or something like that. Generally speaking, most people will only need to set one. I'm using the MS3X, so I have full sequential and one injector per cylinder. You can also set these up to vary with fuel pressure (I just learned that clicking through options), but that would most likely be used for a boosted setup where your fuel pressure might increase with boost. I don't know enough about it, so I'm sticking with voltage control. The curve you see is how many milliseconds it takes an injector to start flowing fuel after being commanded to open at any given voltage. This is where the pain/fun begins.
I believe the Megasquirt characterizes injectors in a way similar to GM. Dodge, on the other hand, seems to have a different way of characterizing them. In GM land, 0ms is when the command to open is given. In Dodge land, 0ms is when the injector first start flowing fuel. So the values in HP Tuners can't be directly entered into the table/curve here. Here is the HP Tuners page with offset (dead time) values:
If you compare those numbers to GM ones of a similar size, you'll see that they are quite large, like almost double what you might normally see (note, the MS also takes values in microseconds, not milliseconds, so take all the numbers here are multiply by 1000). The other table you need to make sense of this is the injector PW vs fuel mass table:
This is the really important one that characterizes how much fuel the injector flows based on the commanded pulse width. Graphed out, it looks something like this:
Pretty straight line right? Well...kinda. The slope of this apparent straight line is typically what I'd call the injector size. In my case, this works out to be ~29.94 lbs/hr. However, if we zoom waaaaay in at the very start, we see the interesting part:
See that kink at the start? This is what gave me several hours of confusion trying to understand. The problem is, the MS code wants to assume that injectors are always linear, so it starts at 0 and draws a straight at the slope we established in the last graph:
That's a problem, because that means we're actually injecting more fuel than the MS thinks we are. What we need to do is include the offset value as well to get a more complete picture of a full injection event from the start of the command to the end of the injection event. Let's assume we're running at 14v and pick that offset off the table. That gives us something that looks more like this:
Now we've gone the other way and the MS thinks we're injecting more fuel than we actually are. What we need to do now is add offset to the red line so that it lines up as close as we can get it with the blue line:
Where this red line crosses the X axis is the offset value that we need to use for the MS, which is quite a bit different than where the blue line crosses the X axis, which is the offset value that Dodge uses in HP Tuners. In this case, the offset is ~0.75ms, or 750 on the MS table/curve, compared to the 1.272 or 1272 value that HP Tuners has in the table. I learned the hard way that just because two things might be called the same thing in tunes, doesn't mean they are measured the same way. I tried using the values straight from HP Tuners on a tune and I could barely even get my car to start, let alone idle.
So it's also worth noting that you could have similar problems even if you buy your injectors aftermarket and they provide numbers. It's probably worth talking to your supplier and asking what system or computer those numbers were measured for. Some will say injector drivers inside the computer itself will also matter (the actual transistors switching the current), though I haven't found a lot of information either way to back up that theory. Fortunately, the rest of the settings are a little more straightforward.
Alright, I guess it's time for the next tab, fuel settings. This is probably the potatoes of the meat and potatoes of tuning. There are some fairly important settings in here and this is largely where I found the difference between a tune that runs "good" and one that runs "smoothly".
First up, we need to characterize our injectors. When dealing with a speed density tune, these are some of the more important numbers you can find. To reiterate something I heard Greg Banish say in an interview, any errors that you make at this point are just going to get baked in and compound things down the road. Don't let that scare you, it's not like your engine won't run or will immediate blow up if you get things wrong. It's more that you might end up chasing your tail trying to figure out why your tune feels like it changes day to day if you get things way off here. In reality, this is the "fine tuning" portion that I found makes my engine feel much smoother and more OEM like. The biggest problem is finding good data for your injectors. OEM stuff can be hard to find, aftermarket maybe a little easier, though you may need to be wary of any values you get. If you're really interested, the Minnoxide interview with Greg Banish on YouTube is a really interesting watch.
So anyway, we're going to start off with injector dead time settings. Injectors are ultimately just little solenoids. They use electricity to produce a magnetic field to move something out of the way to allow fuel to flow. "Dead time" as Megasquirt calls it, is the time between sending a signal to an injector and when fuel actually starts flowing. This can also be called offset or a number of other things. It's effectively the amount of time it takes the pintle/disk/whatever inside in the injector to move out of the way. The voltage you run them at will affect time timing as lower voltages will produce lower magnetic fields, which causes the injectors to take longer to open. I pulled these values out of my HP Tuners file, but had to do a lot of math and head scratching to end up with what I think are the "correct" numbers for my injectors.
You have the option of setting up to four curves. Generally speaking, I think you would probably only do this if you were running staged injection (so for instance, one set of injectors that runs at low rpm and another that kicks on at high rpm). Or, if you are running dual fuel and have one set that runs your pump gas stuff and another that runs your methanol or something like that. Generally speaking, most people will only need to set one. I'm using the MS3X, so I have full sequential and one injector per cylinder. You can also set these up to vary with fuel pressure (I just learned that clicking through options), but that would most likely be used for a boosted setup where your fuel pressure might increase with boost. I don't know enough about it, so I'm sticking with voltage control. The curve you see is how many milliseconds it takes an injector to start flowing fuel after being commanded to open at any given voltage. This is where the pain/fun begins.
I believe the Megasquirt characterizes injectors in a way similar to GM. Dodge, on the other hand, seems to have a different way of characterizing them. In GM land, 0ms is when the command to open is given. In Dodge land, 0ms is when the injector first start flowing fuel. So the values in HP Tuners can't be directly entered into the table/curve here. Here is the HP Tuners page with offset (dead time) values:
If you compare those numbers to GM ones of a similar size, you'll see that they are quite large, like almost double what you might normally see (note, the MS also takes values in microseconds, not milliseconds, so take all the numbers here are multiply by 1000). The other table you need to make sense of this is the injector PW vs fuel mass table:
This is the really important one that characterizes how much fuel the injector flows based on the commanded pulse width. Graphed out, it looks something like this:
Pretty straight line right? Well...kinda. The slope of this apparent straight line is typically what I'd call the injector size. In my case, this works out to be ~29.94 lbs/hr. However, if we zoom waaaaay in at the very start, we see the interesting part:
See that kink at the start? This is what gave me several hours of confusion trying to understand. The problem is, the MS code wants to assume that injectors are always linear, so it starts at 0 and draws a straight at the slope we established in the last graph:
That's a problem, because that means we're actually injecting more fuel than the MS thinks we are. What we need to do is include the offset value as well to get a more complete picture of a full injection event from the start of the command to the end of the injection event. Let's assume we're running at 14v and pick that offset off the table. That gives us something that looks more like this:
Now we've gone the other way and the MS thinks we're injecting more fuel than we actually are. What we need to do now is add offset to the red line so that it lines up as close as we can get it with the blue line:
Where this red line crosses the X axis is the offset value that we need to use for the MS, which is quite a bit different than where the blue line crosses the X axis, which is the offset value that Dodge uses in HP Tuners. In this case, the offset is ~0.75ms, or 750 on the MS table/curve, compared to the 1.272 or 1272 value that HP Tuners has in the table. I learned the hard way that just because two things might be called the same thing in tunes, doesn't mean they are measured the same way. I tried using the values straight from HP Tuners on a tune and I could barely even get my car to start, let alone idle.
So it's also worth noting that you could have similar problems even if you buy your injectors aftermarket and they provide numbers. It's probably worth talking to your supplier and asking what system or computer those numbers were measured for. Some will say injector drivers inside the computer itself will also matter (the actual transistors switching the current), though I haven't found a lot of information either way to back up that theory. Fortunately, the rest of the settings are a little more straightforward.