[miker]

Have I missed something here, if you could run hardware injection test at a preset 1 of 5 predefined pulse widths (eg 20%, 40, 60 ,80 &100%) with the battery voltage displayed and measure the fuel flow. What else would you need to calculate dead time?

I'm reading this thread thinking most of what is needed already exists, it's just a slight change to hardware test to generate the numbers.

[jonbill]

I'd vote for it.

[moonie223]

Run your injectors wide open for a precise and known value, measure the result and use math to figure your max wide open fuel flow rate. Depending on the size of the injector and the size of your graduated cylinder, run it wide open as long as you can. Lets just say it's 30 seconds.

Next, create a set of pulses that together equal exactly the time you ran the injectors for the first test, make note of how many pulses that is. In this case let's say it's 5ms, so 30 seconds is 6000 pulses. Measure the fluid, subtract the volume from the first test from the volume in the second test, then divide by number of pulses.

In essence you are running with no deadtime by leaving the injector wide open and not closing it. Next you are running the injector for the same exact value of time, except you are putting a deadtime in there.

You don't need anything but something to reliably trigger the injector. Stupid simple.

Please for the love of god do not burn your garage down. If you are doing this with regular old gasoline, this means you.

[jonbill]

Isn't the proposal that Speeduino HW test mode be the thing that opens the injectors for a precise and known amount of time?

[moonie223]

What exactly is the point of making the speeduino the tester?

It already is a tester, I could write code that would run on a mega that does so stupid easily. Putting the code to do so in the firmware is pointless.

If the question is can the test be done on the car, then the answer is no it cannot safely be done. If you are going to go through the trouble of setting it up on the car, then you're right back to "please don't use gasoline" as well as setting yourself up for massive failure trying to shove cylinders filled with fuel and vapor under the hood.

If we aren't getting into ETB for liability, then I suppose we should also forgo in-situ injector flow benches for the same exact reason.

Someone post the video of the kid burning his miata/garage down fucking with open cans of gasoline that aren't even being injected at 40+ PSI.

[theonewithin]

There are plenty of other options freely available you can use instead of adding to Speedy no code.

If you need an injector test bench, build one. Or grab someone else's code that does injector testing on the Mega and set it's output pins to use the same as your chosen Speeduino board.

Adding extra single use stuff that isn't actually part.of running or controlling the engine isn't really what Speeduino's is about.

[theonewithin]

Hell. Use Ardustim and a Speeduino and set it up anyway you want so you can have a precise pw.

So much overthinking when what you want can't be done right now with no modifications at all.

[jonbill]

The point of making speeduino the tester would be to save others the effort of thinking about it, writing some stupid easy code and assembling a stupid easy circuit to fire an injector.
Its sort of the point of open source projects in general. its sort of the point of commercial products.
knowing you can make something is not the same as having a reliable tested product.

I'd agree it's a pollution of Speeduinos purpose, so a separate doobie might be better, but if someone is proposing to add it to speeduino hw test mode and TS, I'd vote for it.

[theonewithin]

If it's not required to run the engine it should not be in the code.

The memory of the Mega is already getting pushed.

There are other projects out there.

Use them


Adding it to Speeduino is a waste of developer time.

[miker]

As we have hardware test, I was just asking if it was as simple as a change to that (which it seems to be). Variable memory is tight on speeduino, however this is a small code change so less memory critical.

Reality is we're suggesting having the existing hardware test for fuel injector run for a fixed period instead of until you click stop. So the pollution of the code is already there. This is just an idea that would potentially make the existing code pollution help more people who struggle to find the information.