Multiple ways to find misfires!

So I recently put together an accelerometer and infrared switch for driveshaft balancing, I had the idea of using it for misfire detection for some of those pesky engines (Looking at you GM LS) that use the ugliest frequency CKP that is virtually impossible to detect misfires with.... During my testing I decided to check alternator voltage as well, as this is what pico diagnostics use (but never works for me) and it worked wayy better than I would have imagined! I don't think I will be backprobing CKP for misfire anytime soon! Here's a snapshot of the test I saved the file if anyone wants it.

Sent from my SM-G781V using Tapatalk

Image Attached

Sent from my SM-G781V using Tapatalk

1 thing that's curious is this jeep I was testing with dropped to 300 rpms when the injector kill test was performed I will find another vehicle to try this on that doesn't cut rpm in half and see if it is still easy to identify.

Sent from my SM-G781V using Tapatalk

That looks interesting, for sure. Thank you for sharing! Did you work with the pico NVH kit stuff?

With pesky ones I often go with exhaust pulses as I find this test pretty conclusive. Depending on the severity I also like the CSS script within the autoscope software.

This is a home built accelerometer using an adxl335 this particular setup is only using x axis, im not sure I will use it enough to justify the cost of pico nvh kit so I started out cheap! I don't seem to have much luck with exhaust pulse sensor for misfire although it could be my sensor design....

Sent from my SM-G781V using Tapatalk

Unfortunately you can't use picos nvh software without unlocking your scope with a 1 time use code from there nvh kit...

Sent from my SM-G781V using Tapatalk

Here is same test on another vehicle this vehicle is a transverse engine and cylinder 1 is not accessible so I am synced to cylinder 2 it is significantly harder to spot in this waveform with the vibration sensor than it is in the voltage test I am very shocked how well it shows up in the voltage on both vehicles

Sent from my SM-G781V using Tapatalk

I love your captures. :woohoo: There's been a few vehicles in the past that I could have really used an alternate misfire testing method like this.

Can you give us more details about your accelerometer build? I love a good excuse to go shopping on DigiKey.

This build is super simple it requires an accelerometer, 5v power connection, bnc or banana connector, a box to enclose it in and a magnet to attach the box to whatever your testing and some hot glue to hold it all together I used off the shelf boxes but you could easily 3d print them and I went with bnc for this because I have a 25ft bnc cable here's the link for the accelerometer on Amazon I can link the box,bnc,power connect if needed.(also I use a 5v USB cable to power this from my laptop or battery power supply) (for this build I only used X axis )

AITRIP 3PCS ADXL335 3-Axis Accelerometer Angular Transducer Sensor GY-61 Module Analog Output 3-5V a.co/d/4lYnZRW

Sent from my SM-G781V using Tapatalk

Ben wrote:

Image Attached

So that's how Pico get AC voltage to work. Very heavy low pass filtering. That ain't gonna work with the Hacktek but it gave me an idea...........

Channel 1 is an RPM pickup on #1 plug wire. Channel 2/6 is volt dropping the alternator power cable. Lights and blower on in the first two captures.

Straight DC on channel 2 in this one. It's a bit rough but appears to show the 4 power stroke accelerations.


Moved to the hardware low pass filter on channel 6 and unplugged #2 injector.


Turned on the rear defrost too for this last one. Maybe more load ain't better.......

This is a pretty cool device you made. I have the stuff ordered as we speak. I was curious though; if I wanted all 3 axis’
Would I just use 3 BNC connectors then?

Thank you sir