One waveform One picture

Ask for and share advice on using the PicoScope kit to fix vehicles here.
ben.martins
Pico Staff Member
Pico Staff Member
Posts: 539
Joined: Tue May 16, 2017 1:02 pm

Re: One waveform One picture

Post by ben.martins »

Sometimes direction is all you need when it comes to diagnosis and whilst DTC's can point you in the right direction they can also lead you down a rabbit hole. When faced with the challenges of a crank, no start we must establish some basic essentials for the internal combustion engine to start.

For the diesel engine it's slightly easier than the petrol variety as we don't need to rely on spark igniting our air and fuel mixture but ultimately they both need 3 main things - compression, air and fuel. Making sure you have the basics is always vital. This is where a relative compression check comes in handy as it's quick, easy to perform and non intrusive. Adding in a fuel pressure sensor will let you see if fuel is getting to the right places and if you include a pressure transducer/first look sensor, you can get an idea on whether the engine is pulling air into the cylinder.

What happens though if all these come back OK though? Where to next? This one picture one waveform follows on from the following study - https://www.picoauto.com/library/case-s ... ic-process where we look at just that and how including math channels can help to reveal more about what the engine is doing.

Graphing the crankshaft sensor isn’t anything new and we have a number of forum posts surrounding this technique of observing the crankshaft speed as it increases and decreases during the 4 stroke cycle. These can be seen here - viewtopic.php?p=37861#p37861. Recently we have made some changes to how the crank math function works, in order to make it a little easier on the eye when it comes to analysis. However we do need to be aware of what’s happening to ensure we know when to use this math channel. The old crank math used to leave a gap where the missing teeth were present and we would then have a very long spike as the signal returned and the math continued. With the new crank math channel in PS7A, this spike has been removed and we have joined the points between where the missing teeth are. This makes things a lot easier to view as seen below.
Volvo Crank non start.png
As you can see, the crank math channel in blue clearly shows a slowing down in the crankshaft on four of the 6 injectors. Using the firing order we can also see that cylinders 3 and 6 are the only ones increasing the crankshaft speed. I hope you agree that the new math channel makes it easier to see this slow down in the crankshaft when compared to Freq math which is below in Red. Here we see the dropout in the signal due to the missing teeth.

However, there are some advantages to seeing these dropouts, especially if there was a glitch on the crankshaft sensor. As I’ve mentioned the crank math will now join the gap between the missing teeth which means the crank math channel could mask this glitch. As both the crank and frequency math channels are built into PS7A, when evaluating the crankshaft sensor it may well be beneficial to include both math channels to ensure no stone is left unturned during diagnosis.

The above vehicle was fitted with 4 injectors which cured the crank no start and the lorry was back out on the road again without having to replace all 6. Some may argue that all 6 should have just been done and it solves any problems in the future. However, in a world where prices continue to rise, being able to save replacing parts where not needed urgently will surely be appreciated by the customer.

I hope this helps and thanks to Lee Sharp, L&D Commercials, for supply the captures and ultimately doing the job! Keep an eye out for a case study with crank math back in action but not all is as expected!

If you have any feedback on the crank math channel please could I ask you to post on the PicoScope 7 Automotive Software page - viewtopic.php?p=102798#p102798.

Kind regards

Ben

Sharpy
TwoWaves
TwoWaves
Posts: 55
Joined: Sun Dec 23, 2018 9:59 am

Re: One waveform One picture

Post by Sharpy »

Great write up Ben, no thanks needed as without my Pico I would have struggled to prove the problem as with no codes to give us a clue we would have been guessing.
I would have loved to replace all 6 injectors but I struggled to convince the customer to replace 4 he felt that doing 1 or 2 would have given it enough to start as he felt once it was running even as it was it drove just fine but now it’s repaired he has noticed a massive difference in performance and at 32 tonnes that helps.

Thanks to Ben and Pico

muttnjeff
TwoWaves
TwoWaves
Posts: 72
Joined: Tue Jul 21, 2020 7:19 pm

Re: One waveform One picture

Post by muttnjeff »

Spent some time this weekend investigating a P0304 on a 1996 Toyota 4Runner 3.4L

Fairly constant misfire, in all rpm ranges, used scan tool to check misfire counts, cyls 1-3 had none and cyls 5-6 would accumulate some while cyl 4 was going crazy

Hooked scope up
Channel 1 on cyl 4 secondary Kv
Channel 2 on cyl 2 secondary Kv
Channel 3 on CMP
Channel 4 on coil 1 primary current

After collecting a good capture I found that the cause could not be in the coil primary as it was affecting only cyl 4, this system uses a waste spark ignition, 1 coil for 2 cylinders, being only cylinder 4 was misfiring quickly rulled out the coil.

I set out removing the plug wire that feeds Kv to cyl 4 and found that the insulation around the plug boot was deteriorated allowing the spark to release outside the combustion chamber.

New plug wires fitted, and misfire was resolved.
Attachments
1996 Toyota 4Runner P0304.png

martinr
TwoWaves
TwoWaves
Posts: 53
Joined: Tue Mar 06, 2018 4:06 pm

Re: One waveform One picture

Post by martinr »

ben.martins wrote:
Fri Feb 11, 2022 4:51 pm
Sometimes direction is all you need when it comes to diagnosis and whilst DTC's can point you in the right direction they can also lead you down a rabbit hole. When faced with the challenges of a crank, no start we must establish some basic essentials for the internal combustion engine to start.

For the diesel engine it's slightly easier than the petrol variety as we don't need to rely on spark igniting our air and fuel mixture but ultimately they both need 3 main things - compression, air and fuel. Making sure you have the basics is always vital. This is where a relative compression check comes in handy as it's quick, easy to perform and non intrusive. Adding in a fuel pressure sensor will let you see if fuel is getting to the right places and if you include a pressure transducer/first look sensor, you can get an idea on whether the engine is pulling air into the cylinder.

What happens though if all these come back OK though? Where to next? This one picture one waveform follows on from the following study - https://www.picoauto.com/library/case-s ... ic-process where we look at just that and how including math channels can help to reveal more about what the engine is doing.

Graphing the crankshaft sensor isn’t anything new and we have a number of forum posts surrounding this technique of observing the crankshaft speed as it increases and decreases during the 4 stroke cycle. These can be seen here - viewtopic.php?p=37861#p37861. Recently we have made some changes to how the crank math function works, in order to make it a little easier on the eye when it comes to analysis. However we do need to be aware of what’s happening to ensure we know when to use this math channel. The old crank math used to leave a gap where the missing teeth were present and we would then have a very long spike as the signal returned and the math continued. With the new crank math channel in PS7A, this spike has been removed and we have joined the points between where the missing teeth are. This makes things a lot easier to view as seen below.

Volvo Crank non start.png

As you can see, the crank math channel in blue clearly shows a slowing down in the crankshaft on four of the 6 injectors. Using the firing order we can also see that cylinders 3 and 6 are the only ones increasing the crankshaft speed. I hope you agree that the new math channel makes it easier to see this slow down in the crankshaft when compared to Freq math which is below in Red. Here we see the dropout in the signal due to the missing teeth.

However, there are some advantages to seeing these dropouts, especially if there was a glitch on the crankshaft sensor. As I’ve mentioned the crank math will now join the gap between the missing teeth which means the crank math channel could mask this glitch. As both the crank and frequency math channels are built into PS7A, when evaluating the crankshaft sensor it may well be beneficial to include both math channels to ensure no stone is left unturned during diagnosis.

The above vehicle was fitted with 4 injectors which cured the crank no start and the lorry was back out on the road again without having to replace all 6. Some may argue that all 6 should have just been done and it solves any problems in the future. However, in a world where prices continue to rise, being able to save replacing parts where not needed urgently will surely be appreciated by the customer.

I hope this helps and thanks to Lee Sharp, L&D Commercials, for supply the captures and ultimately doing the job! Keep an eye out for a case study with crank math back in action but not all is as expected!

If you have any feedback on the crank math channel please could I ask you to post on the PicoScope 7 Automotive Software page - viewtopic.php?p=102798#p102798.

Kind regards

Ben
Ben, another very interesting write-up. Thankfully, the software team had the good sense in PS7 to retain the original crank maths channel, with its raw data, alongside the new “airbrushed” crank maths channel. As you imply, one should always have access to the raw data, as well as being aware of any processing that might have been applied to a signal.

ben.martins
Pico Staff Member
Pico Staff Member
Posts: 539
Joined: Tue May 16, 2017 1:02 pm

Re: One waveform One picture

Post by ben.martins »

I take no credit from the following post as it is from one of our users on our Facebook Automotive Group. Edward Boers was kind enough to share his experience of working on a Pierce Ariel Ladder. Should you not know what this is please look it up! Thanks Edward for allowing me to post here as it's great to see Pico being used in application that may not be obvious!

"Hello all,

Just wanted to share a screen capture of a Picoscope being used to diagnose an issue not of a traditional automotive application.

I am a EVT technician and our city owns nearly 200 fire apparatus and approximately 120 ambulances.

This screen capture was for the purpose of diagnosing a Pierce Aerial ladder which had complaint of the ladder would continue to either both raise and lower for a period of time after the operator would move the control lever to its neutral position.
Pierce Aerial ladder EB.png
The blue trace is the joystick control signal. 2.5 volts would equal the lever being in the neutral position and close to 0 volts would indicate that the lever was moved to the full raise position and 5 volts would indicate the lever was moved to the lower position.

The green markings on the screen are the pulse width modulated output from what is known as an AIM (aerial interface module) module to move a solenoid operated directional control valve to raise the ladder and the red would be the same except to lower the ladder.

I added a math channel which is the black trace so that one could see the duty cycle on the raise pulse width modulation function coming out of the AIM module.

What was initially challenging was to determine if the problem was an electrical issue or a hydraulic issue. One could clearly see that after the control lever is moved from the raised position or the lower position and moved back to the neutral position the module could take up to 3 seconds to bring the pulse modulation back to zero and stop the hydraulic control of the ladder.

I didn't have the forethought to take another capture but this pulse width modulated signal leaves the AIM module and then enters a digital to analog converter and then it moves on to a programmable end of line device which will also ramp up or slow down the control of the hydraulics to make sure they are soft movements that don't throw fireman off of the ladder if it were to stop moving suddenly.

Having the scope hooked up showed that the AIM module was the culprit and powers and grounds checked out on it just fine. The module was replaced and it's been in service for an entire year with no issues."

To see the original post on FB please click here - https://www.facebook.com/groups/PicoSco ... 6266910853

Thanks again Edward and look forward to seeing more.

Kind regards

Ben

Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 1581
Joined: Sun Aug 25, 2013 7:22 am

Re: One waveform One picture

Post by Steve Smith »

Note quite One Waveform One Picture as there is no waveform involved, however we just had to share this.

Many thanks to Randy Janson from over the pond for sending in this gem

Chevrolet Corvette

Long story short, customer complaint is an “irritating noise” from the suspension

The Technician had just watched the following video https://www.youtube.com/watch?v=SCuwReAWIGw&t=4s describing how to generate vibrations throughout the vehicle

Once the resonant frequency of the offending component had been identified, the subsequent continuous rattle could then be located.

The image below speaks volumes
Image 1
Image 1
The vehicle had recently had a new rear leaf spring installed and the incorrect nut had been used.

The nut had tightened on the crossmember instead of the eccentric bushing itself (Therefore no clamping force applied to the eccentric washer)

The witness mark clearly demonstrates this fault condition

The correct nut repaired the vehicle

Thank you again Randy and I hope this helps others, take care…….Steve

Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 1581
Joined: Sun Aug 25, 2013 7:22 am

Re: One waveform One picture

Post by Steve Smith »

Following on from this case study here https://www.picoauto.com/library/case-s ... -non-start this is one of those vehicles that simply does not want to leave the workshop!

Long story short, as described in the case study, the vehicle has a history of Engine ECU replacements ranging from second-hand to finally a refurbished unit.

After what seemed like an “age” of getting this vehicle out to the customer all seemed well until it returned approx. 5 weeks ago with multiple engine codes and limp home activated!

As you know, all kinds of things play on your mind when this occurs and you run through the various scenarios of “blame” in the process.

Remaining objective is key and so a vehicle scan revealed the following DTC’s:

Engine control:
P0220 Throttle valve position sensor/App sensor 2 signal too high
P1526 Throttle vale control unit malfunction

ABS:
U2105 Communication with Engine ECU faulty
U2106 Communication with Transmission ECU faulty

Based on the above and the known vehicle history in the case study above, the focus had to be upon the throttle body where the “Throttle valve position sensor” was clearly at fault (See below)
Image 1
Image 1
Note above how channel D (TPS sensor 2) remains fixed at approx. 5 V during the initial operation of the throttle, followed by a sporadic signal during periods of intermittent “contact” about the throttle position sensor

The included image reveals the fault and it beggars belief that a component as critical as throttle body uses such narrow gauge wire to transfer the signals from the connector housing to the TPS track! The solder joint to which the detached wire should be attached is much smaller than the remaining 4 wires which may also have contributed to the failure, food for thought as ever

I hope this helps, take care……Steve

ben.martins
Pico Staff Member
Pico Staff Member
Posts: 539
Joined: Tue May 16, 2017 1:02 pm

Re: One waveform One picture

Post by ben.martins »

The following post is from a Pico user who kindly shared this with the Pico Automotive FaceBook group but I thought it would be good to have it on display here as well! Thanks here to Kris Williams at KW Diagnostics who has kindly allowed us to share his work.

"Here I had a Golf in for AC not working, which already had a AC pressure sensor fitted to it. A quick scan for fault codes reveals a fault for no signal from the Pressure sensor.

Now you would be forgiven for thinking the wiring is ok from initial tests as checking with a multimeter the 12V supply and ground was good and there was expected voltage on the signal (LIN bus) wire. Unplugging the sensor voltage stayed the same on the LIN wire proving there is a connection at least to another sensor or control unit, this could lead to condemning the sensor.

Using the scope to take a look at the actual LIN messages we can see we have a noisy poor quality signal, from experience we know there must be some corrosion within the circuit somewhere distorting the LIN messages.

Now this is where the wiring diagram saves us time. The diagram shows that between the pressure sensor and the AC control unit there is a fresh air control unit and a Air Quality sensor sharing the LIN signal wire. Looking at live data and the codes shows that the other 2 components have no problems communicating to the AC control unit, safely ruling out a control unit fault, or indeed wiring between those components and the AC control module.

That leaves just the wiring between the pressure sensor and the air quality sensor under the scuttle panel.
Again the scope helps here, carefully wiggling the loom and watching the LIN signal on the scope meant the wiring fault is found within a small section of the loom, saving time from otherwise removing all of the insulation in search of the corroded wire."
Before and wire.png
Hope this helps and shows how your mulimeter can sometimes put you on the wrong path. Thanks again to Kris for sharing.

Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 1581
Joined: Sun Aug 25, 2013 7:22 am

Re: One waveform One picture

Post by Steve Smith »

Following on with this case study viewtopic.php?p=103362#p103362 the turning point during the diagnosis came as a result of using NVH which revealed an E0.5 engine order vibration.

E0.5 vibration orders develop as a result of incomplete combustion (arguably called a misfire).
See here for further information viewtopic.php?p=86961#p86961

With incomplete combustion on a single cylinder (4-stroke multi cylinder engine) we have one disturbance for every 2 revolutions of the crankshaft.

To help with the interpretation of E0.5 let’s start with a known good example:

Think of a 4-cylinder engine with complete and balanced combustion. Here we have 2 disturbances for every one revolution of the crankshaft (E2)

Remember during the 4-stroke cycle we have 2 combustion events for one revolution of the crankshaft

Now think of this same engine with incomplete combustion on one cylinder

During one revolution of the crankshaft there will be complete and balanced combustion from 2 good cylinders.

During the next revolution of the crankshaft one cylinder will perform OK but the other will suffer from incomplete combustion so placing a huge and dominant disturbance into the crankshaft at the rate of 1 disturbance per 2 crankshaft revolutions (1 / 2 = 0.5)

Armed with this knowledge, using NVH during investigations of complaints such as “holding back”, “jerking” and poor ride quality, we are able to quickly direct our diagnostic route (and more importantly time) to the correct onboard system

In the example below, I knew my concern surrounded combustion and therefore could focus attention on fueling etc. and move away from ride quality / transmission errors.
i.e., what I could feel in the cabin was relevant to the symptom/complaint
Image 1
Image 1
Note above how poorly defined the injector nozzle spray holes appear as a result of exposure to "hammering" or detonation within the combustion chamber and hence, incomplete combustion under specific load conditions

Here is an example where using the accelerometer enabled the correct diagnostic route relevant to transmission and not combustion related (Often it is difficult to separate the two from just the symptom alone) topic22833.html

I hope this helps, take care…..Steve

Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 1581
Joined: Sun Aug 25, 2013 7:22 am

Re: One waveform One picture

Post by Steve Smith »

Diagnosis before dismantling pays dividends when it comes to areas of “focus” once permission has been granted to carry out a repair

Here is another example of In-Cylinder analysis using the WPS500 pressure transducer to highlight low compression

Whilst low compression can be diagnosed with a conventional compression gauge, lets take a look at the additional information within the attached image/video (sorry, not quite One Waveform One Picture)

Starting from the left-hand compression tower, I think we can agree we have low compression (approx. 0.431 bar)

Interestingly, we do not have leaning compression towers (as small as they are) suggesting no cylinder leakage.

Our expansion pocket is most certainly “deep” given our cylinder pressure starts at 0.431 bar and drops below atmospheric approx. 18.61 degrees ATDC.

Assuming the cylinder swept volume has not changed we have 145.4 (EVO) – 18.61 = 126.79 degrees of crankshaft rotation in which to generate such a low negative pressure (approx. -980 mbar)

A cylinder which can produce such a deep expansion pocket and no leaning compression towers can be considered to be “sealed”

At 145.4 degrees, our expansion pocket incurs an abrupt change of direction towards 0 bar (atmospheric) which is confirmation of an exhaust valve open event (EVO)

Typically, EVO occurs around 25 to 30 degrees BBDC on the expansion stroke and so this suggest our EVO could be advanced (Confirm with a capture from the Reference Waveform Library)

Whether or not the EVO is advanced, we can confirm that an EVO event is taking place on or near the expected time

The exhaust valve duration (approx. 145 to 365 degrees ATDC) looks to be near the mark to as the exhaust valve close event (EVC) appears to be shortly after 360 ATDC.

Remember this is difficult to prove with all the events that take place at 360 degrees but looking at the “typical” shape and timings of this 4-stroke waveform we are not far away
Note, during the exhaust stroke our cylinder pressure remains at or around atmospheric pressure (0 bar) suggesting the exit of exhaust gases is clear. We must take this with a pinch of salt as there are no combustion gases present and so this would need to be confirmed.

Next, we have the intake pocket which has a deep “dish” appearance with a peak negative pressure of -988.4 mbar!

The “dish” effect is typical of a cylinder “stroke” where there is no change in pressure attributed to a valve open event.

Compare the expansion pocket with a clear EVO event to the intake pocket where the intake valve has failed to open

In such a scenario, the deep dish is formed by the progressive reduction in cylinder pressure attributed to the decent of the piston on the intake stroke (with no IVO event) and the gradual increase in cylinder pressure on the compression stroke. (No IVO event means no IVC event)

Returning now to the right-hand compression tower, it is no surprise we have low compression given we cannot pull air into the cylinder and therefore have nothing to compress!

Based on the above we can therefore confirm our low compression is not due to cylinder leakage, our cylinder is sealed (from a negative pressure perspective) our EVO and EVC events are near typical (but must be confirmed) and the reason for low compression is due to no IVO event. (Refer to video & image below)
Image 1
Image 1
INTAKE.mp4
Video 1
(1.23 MiB) Downloaded 255 times

Armed with this knowledge (that could not be gained from a conventional compression gauge) we know where to focus our attention when dismantling is permitted

I hope this helps, take care……Steve

Post Reply