Measuring Timing using injector pulses

Ask for and share advice on using the PicoScope kit to fix vehicles here.
Post Reply
Mpylypchuk
Newbie
Posts: 5
Joined: Thu Jan 05, 2023 9:20 pm

Measuring Timing using injector pulses

Post by Mpylypchuk »

Hi everyone! I work on diesel engines and was wondering about using injector pulses to determine engine timing or at least a change in timing.
Is there any posts related to this already or is there someone who has some info on this?

I also understand this may be difficult as today's engines using multiple injector firing events for different engine modes. I'm curious if this is still somehow possible.

picowight
TwoWaves
TwoWaves
Posts: 140
Joined: Sat Oct 26, 2019 6:45 pm

Re: Measuring Timing using injector pulses

Post by picowight »

If you just want to see timing change relatively over the rev range/load, scope an injector and trigger off any fixed point on a cam waveform using an advanced trigger.

Mpylypchuk
Newbie
Posts: 5
Joined: Thu Jan 05, 2023 9:20 pm

Re: Measuring Timing using injector pulses

Post by Mpylypchuk »

Thanks so much I will try that. Haven't played with advanced triggers yet but i'll give it a shot!

picowight
TwoWaves
TwoWaves
Posts: 140
Joined: Sat Oct 26, 2019 6:45 pm

Re: Measuring Timing using injector pulses

Post by picowight »

This video might help


Use the same concept but measure the section of cam signal that you want so you get it spot on (rather than the crank shown here.)

Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 1404
Joined: Sun Aug 25, 2013 7:22 am

Re: Measuring Timing using injector pulses

Post by Steve Smith »

Hello and thank you for the posts

Thank you also picowight for the tip regarding “triggering off any fixed point on a Cam waveform”

As described in the above video, triggering off the missing teeth of the crankshaft sensor using the “Pulse width trigger” & time base settings described, works fine at fixed engine speeds (typically idle) but will tend to miss-trigger with an increase in engine speed.

There will be occasions where the trigger captures the missing teeth exactly as required, and other events where the trigger fires upon the missing teeth 360 degrees later as both pulses (at 360-degree intervals) meet the specified trigger conditions

In the example below the trigger captures the missing teeth as required given the “positive” pulse width is greater than 900 µs (thanks to the missing teeth)
Image 1
Image 1
Below the trigger conditions have not changed (only the RPM) yet we are now triggering on the missing teeth 360 degrees prior to the missing teeth in image 1
Image 2
Image 2
Utilizing the camshaft (often with a unique lug arrangement) instead of the crankshaft presents improved options for triggering

Below we have the camshaft lug arrangement of a BMW 320 d (Engine code N47) which includes 3 positive pulses (small, medium, and large)
Image 3
Image 3
Note above how the trigger condition have changed to a specified time range (10.75 ms & 3 ms) rather than just a time interval used in images 1 & 2 (“Greater than” 900 µs)

So, what does this mean?

To summarize the trigger conditions in image 3 above, we are asking PicoScope to trigger on a Positive pulse width spanning a time range inside 3 ms to 10.75 ms.

This means PicoScope will trigger on any positive pulse width between 3 ms and 10.75 ms; in other words, if our positive pulse measures anywhere between 3.000001 and 10.749999 ms in width, it will meet the specified trigger conditions and fire accordingly

So, why is this useful?

The time range feature of the “Pulse width trigger” provides a means to trigger on pulses that change in width, one typical example is our camshaft signal at various engine speeds

Our chosen pulse (upon which to trigger) will decrease in width with an increase in engine speed (frequency)

How do I find the time range values to enter?

Starting with image 3 above (engine at idle) I have chosen to trigger off the small positive pulse of the camshaft which measures 10.27 ms. In the example below (image 4) I have increased the engine speed to approx. 2000 rpm and measured the same “small” positive pulse at 4.3 ms.
Image 4
Image 4
Armed with this information I have chosen to enter a time range of 3 and 10.75 ms into my trigger dialog box and select “Inside range”. This is because I now know our chosen small positive pulse of the camshaft will fall inside this selected time range when increasing the engine speed from idle to 2000 rpm, so providing a stable trigger upon which to monitor injector timing.

However, as the engine speed continues to increase, not only will our chosen small positive pulse decrease in width, so will the medium and large pulses.

Note below how we have miss triggered on the medium positive pulse instead of the small pulse ! Why?
Image 5
Image 5
Remember, pulse width decreases with an increase in engine speed (frequency) and so our medium positive pulse now meets the trigger conditions specified (inside a time range of 3 to 10.75 ms) as it measures 10.5 ms! (This is expected behavior as the scope is doing exactly as instructed)

As a result of the miss trigger above, our waveform will appear to jitter making it difficult to keep focus on shifts in our injector timing.

Here then lies the limitation of pulse width triggering and the challenges faced when trying to stabilize engine cycles across the entire rpm range at such a time frame (20 ms/div)

I have experimented with our Rapid Trigger feature (1 ms/div and faster) which you can read about here [https://www.picotech.com/download/manua ... oScope.pdf] using a single pulse generated by reflective tape attached to the crankshaft pulley in conjunction with our BNC+ Optical sensor https://www.picoauto.com/products/nvh-a ... sensor-kit

Here the trigger and time base are so fast we end up with one buffer per engine rotation, meaning our cylinder 1 injection event is only present on every other buffer! (See below)
Rapid
Rapid

There are several options at present in which to monitor injector timing across varied engine speeds.........

One of which is choosing a slower time frame (200 ms/div) whilst carrying out a wide-open throttle (WOT) test.

This will require zooming into points of interest within the capture and measuring the start of injection in relation to degrees of crankshaft rotation using both the phase and time rulers

The following forum post topic22201.html looks at using the frequency of the crank as our X axis which was very interesting indeed and I must pick up again at some point in the future. The initial post is very similar to what we wish to achieve above (i.e., stabilize capture across varying engine speeds)

Persistence mode (not yet available in PS7) but not far away (said too much already) This could pay dividends here but may well display the same characteristics when using the pulse width trigger!

A rotation domain would be the ultimate goal where our X axis is liked to degrees of crankshaft rotation rather than “time”. Rest assured work is always in progress with regards to new features.

In the meantime, I will look at all the options and discuss with the team how we can overcome the hurdles described above

I hope this helps, take care…..Steve

picowight
TwoWaves
TwoWaves
Posts: 140
Joined: Sat Oct 26, 2019 6:45 pm

Re: Measuring Timing using injector pulses

Post by picowight »

Fantastic observations and info Steve, much appreciated.
Of course degrees of rotation would be the perfect solution :D

Post Reply