Injector opening delay
 

Hi,

I've built a similar gauge for my '99 Nissan Pulsar (which I think is Almera or Altima or something like that in the USA) after being impressed by the MPGuino when it got mentioned on hackaday. My hardware is similar but totally different software.

My question applies to the OpenGauge/MPGuino as well as my thing. Has anyone actually measured the opening and closing delay of a fuel injector? How do we know it's ~500uS? If the opening time is because of the physical movement of the valve, wouldn't there also be a similar closing delay?

Thanks. -Simon

It is an approximation, the opening takes longer than the closing, and is also dependent on vehicle voltage. But it is measured in terms of consistency between refills at the pump (factoring in refill technique of course). My saturn comes in within 1% consistently, your mileage may vary.

There was a proposal, assuming that your tank refill is accurate, that by keeping track of the number of injector pulses for the tank in addition to the open time , and keeping track of the actual refill amounts. That over several tanks you can find a value for injdelay and microsecondspergallon that best fit all the refills.

Of course if your refills are already within 1%, or off by a consistent percentage, then you might be like "why bother" :) But if not then this is something to investigate.

That's a great idea! I haven't arrived at a value that works consistently for my car yet, although I haven't been keeping data for long yet. I did notice that my measured flow rate is higher than the published rate for my car - at least from what I can find out online. I should have 185cc/min injectors but my current setting is around 210cc/min. That was one of the things that prompted me to wonder about the opening delay.

I doubt my tank refill is very accurate (being a small car with a small tank). I think more information is the answer here. I am actually keeping track of the number of injector pulses in my software anyway to measure the RPM, so I just need to display that number on a separate screen. I wish it was the weekend already...

Here's my setup, for interest's sake. Everything looks better in black.
http://smonson.com/photos/fuelcount2.jpg

That looks like it belongs there, smonson, very nice!

VPV-Motorracing Electronics- Bosch Injectors

I was looking for the delays from Google one year ago. I have found only this link.

SIEMENS-DECA 006 injectors

Flow
2.5 bar 550ml/min
6 bar 860ml/min
Opening delay: 1.14ms at 14V
Closing delay: 0.85ms at 6 bar

Total delay would be 0.29 ms. 6 bar is quite high fuel pressure. Closing delay must depend of fuel pressure.

So then the question is,,,

If I lower the current but double the voltage will the injector open faster?

Hmm. I suddenly have a new experiment for the racecar.... Bwahahahahaha

Dave

That's interesting - a difference of about 0.21ms from what dcb measured on his car. But these ones look like high-performance parts to me. Perhaps that is why. I wish car manufacturers weren't so secretive about everything, it really makes projects like this difficult.

I didn't "measure" it directly, I added it in and the pump numbers corroborated it. I'm sure it could be refined, but to what end? I would suggest a review of the calibration instructions for consistent refills, a lot easier than setting up a test rig to measure it for your specific vehicle, and you might just be overthinkin it. Or you might have peak and hold injectors too, dunno.

Mpguino calibration - EcoModder

The accuracy of my setup varies depending on whether I'm commuting or travelling on the highway. I am just trying for more accuracy here. I'm not saying 500uS is wrong for your car.

Even though your system is quite accurate is all situations, mine isn't. There's a reason why the accuracy changes depending on the driving conditions and I'm just hoping to find out what that is. I don't have peak and hold injectors, though.

I'll be doing a 700km trip over Christmas and have added the ability to display an injector pulse counter on the screen, so that will provide some extra information to narrow down the source of the discrepancy, even if it's just to verify that the fuel injector opening time isn't a big deal.

I understand, maybe if you provided some info about indicated and actual refill amounts and related driving style and mileage, something might click.

Will do :)

measuring injector opening
 

scan data does not measure
ACTUAL injector on time

you can use a high resolution scope .
it is easiest to measure injector open with injector current waveform and it is easiest to measure injector close with injector signal voltage

the open time is about 70% of signal time but varies with different systems

http://farm3.static.flickr.com/2507/...ed33cebe_b.jpg

above the Blue trace is Injector current , the red trace is Injector signal voltage - BUT
this waveform was captured using an injector tester pulsing 4ms signals , it is NOT the same as what you will see on a running car / system
BUT
you can see the "gull" in the current waveform which is caused when the magnetic field changes because the pintle has moved opening or closing the injector's opening

an operating system looks like this
pintle movement is clearer and easier to see , this what an actual system waveforms will look like , black is voltage green is current
http://farm3.static.flickr.com/2563/...1860904b_b.jpg

=================================

That's excellent information. So for that Volvo the total delay is around 0.8ms - quite a difference from the Bosch one. That definitely looks like the best way to measure the open/close delay for a given injector.

hmm yes it is but
i think you are looking for total Open time , the delay itself is only part of the equation as
there is a delay in injector pintle closing as well.

you would need to subtract the delay in closing from the delay in opening and then

use software to subtract that total from total injector pulse width
so the software would know
total injector open time

imho
that is BANK injection , where all injectors squirt at the same time
sequential injection will be similar IN OBD1 systems
but
ThrottleBodyInjection waveforms are very different


that value is going to be different on different system s and
if you are not accounting for
asynchronous pulse , extra fuel at the pleasure of the ECM
you probably should

Right. So, if I read the graph axis correctly, 1.6ms opening delay minus 0.8ms closing delay is 0.8ms total delay, subtracted from the pulse width to get the injector operating time.

I'm working on the mathematical approach suggested by dcb now as I don't have a DC current clamp to measure my injector directly.

1 ohm 10 watt resistor in series to view current
 

Thanks, I know how electricity works. I don't want to go disconnecting everything in the middle of collecting some mileage data, but if I can't get an accurate result this way I will try a current measurement.

Here are the results of my observations :)

The following maths treat all cylinders as a single measurement. Figures will be 4 times higher than the real value for a single injector.

WARNING: PEDANTIC GUY AHEAD

Tank 1 (country):
- predicted 28405mL
- actually 28360mL (error: 0.15%)
- PWM fired 332468 times for 2195.598 seconds
- average pulse width: 6603uS
- average fuel quantity injected: 85.301uL

Tank 2 (city):
- predicted 30954mL
- actually 30950mL (error: close enough)
- PWM fired 385078 times for 2403.540 seconds
- average pulse width: 6241uS
- average fuel quantity injected: 80.373uL

The difference in pulse widths is 362uS and the difference in fuel injected is 4.928uL. From this we can infer that the fuel pumping rate is 0.01361325 uL per uS. Converting units, that's 204cc per minute which is in the expected ballpark for a small car.

The total time to transmit the average injected fuel amount into the cylinder is then 6266uS (tank 1) or 5904uS (tank 2). This predicts the injector time delay at 337uS for both, depending on how inaccurate those particular fuel pumps were.

Thanks dcb and everyone else for all the advice!

put your car make and model somewhere so folks can use that info pls.

It's a Nissan Pulsar/Almera N15 1999 (Pulsar Plus Series II) 5-door hatchback automatic.

I have setup my MPGuino using calculated fuel flow rate from the injector spec, and the standardish 500us delay. This seemed to give pretty sensible figures and typical averages from normal driving, followed by much improved figures after figuring out what the engine likes to do.

I've just done my first tank full and this calculated as 4% more estimated used over pump figure. After plugging in the correction, I am now getting stupid figures for flat MPG, like 60-70 along the flat.

I'm thinking the low duty cycle of travel along the flat and tick over is more sensitive to delay.

I assume the standard MPGuino software does not allow pulse counting so we can use simultaneous equations (eg long periods of tickover against full rally)

Anyway I have done the current measurement suggested at tickover and about 2500rpm and gotten delay figures of 632us and 643us.

Obviously my delay is bigger, but how accuracte is this method. Does it account for fuel momentum etc. Will 40us make much difference?

I think I'll go back to the calculated flow figure and change the delay to 645. Opinions?

deleted