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 »

Luckily for me I get to head out to some of the more interesting cases where we can use the scope. This one was no exception as it's something I've never worked on before - a crusher!

The purpose of these machines is to do just that, crush large lumps of rock and concrete before feeding into a screener which would separate the different rock sizes and deposit them into their own individual piles. To see what these machines can do please see the following video - https://youtu.be/mEL_aVoL86o?t=81

The reason for visiting is to determine the cause of an intermittent cutting out. It would then start up again with no issue before randomly cutting out again.

For those that have read some of the off highway case studies we've been, you will have read that technical information really isn't that easy when it comes to this type of machinery. The issue being is that the machine may come with one manual but the engine would be made by someone else. This machine was manufactured by Extec but the engine was made by Deutz. Having to look between to different drawings was fun, especially as the wire colours didn't match the machine and made worse by the fact the weather was not on our side!

Fortunately, the technician I was supporting had a much better idea as to how these machines operated than me, with the main focus being towards the EPU - Engine Protection Unit. As with a lot of plant machinery there is often additional protection circuits fitted to the machine in the event that something is going horribly wrong, they can be switched off quickly. This ECU is designed to protect the engine should it overheat, stop charging, suddenly develop low oil pressure and more. These crushers are typically left to their own devices and once running continually operate till their work is done. Having something to make sure the engine doesn't destroy itself is vital, especially as it's the only way you can move these things!

The problem being was we didn't know what the ECU was doing in order to shut the machine down. Looking at the connector we had 8 wires which meant bringing out the 8 channel scope. When it comes to this type of environment I try to steer clear of the 8 channel, as dealing with all those test leads in the rain and mud is not fun. However, it comes into it's own when you need to understand how circuits interact with each other.

With the machine up and running and all 8 channels connected, we started with a wiggle test and fortunately the machine cut out when we lowered the electrical panel lid. Pulling on the wiring we then located the issue being a power supply not correctly installed into the electrical board and the terminal screw was actually loose.
crusher.png
Granted, I know this should be one waveform, one picture but I didn't have the one picture to give the best impression!

Using the 8 channel scope we could see that when all is well with the machine, 7 of the wires should be at battery voltage (or close to) with one of the wires at 0V (possible ground). They should stay at a high voltage till a sensor is grounded or has an open circuit. At this point the protection ECU can trigger the emergency relay which then shuts the machine down. This was proven by ground one of the wires in the connector with the machine running which as expected, shut the machine down.

As with everything we try to diagnose, understanding how it works is paramount to understanding how to diagnose it. In the event that technical information isn't available being able to see the interaction between circuits can be extremely useful which is where the 8 channel scope shines.

I hope this helps

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 »

It has been a while since we posted here which is not for the want of trying. Like us all, there are not enough hours in the day to do everything

I quickly wanted to share this one as it was a fault that raised its head on my own car!

Vehicle:
2013 BMW 320 d Touring (F31)

Symptom:
Engine warning light on accompanied with loss of power (Minimal boost from the turbo charger)
Once the engine had cooled, the vehicle performed as normal until hot, where the symptom returned

Fault code:
25D100 Boost pressure sensor Short to ground
27F100 Charge air hose Hoses faulty

A brief inspection of the air intake confirmed no leakage or connection issues

As always, when it came to diagnosis, the fault was illusive and required considerable testing before I could confirm the output from the boost pressure sensor was at 0 V (hot engine) with serial data reporting a fixed “charge air pressure” of 60 mbar regardless of engine speed/load or minimal boost applied
Note: Sensor “power & ground” were good when “signal” at 0 V

So, there you have it, replacing the boost pressure sensor resolved the issue but it got me thinking, “how could I speed up the diagnostic process?” rather than having to wait for the optimal engine temperature for the fault to occur

This was one of those moments where you answer your own question, “by applying an external heat source to the boost pressure sensor”

I know it is not rocket science but simulating symptoms mentioned by the customer or displayed in freeze frame data “theoretically” should reproduce the fault condition

Using the Universal break out lead set https://www.picoauto.com/products//brea ... t-lead-set below, I have supplied an external 5 V & ground to the sensor whilst applying heat until the fault revealed itself
Image 1
Image 1
Note; when the sensor cools, how the output signal was restored, exactly as per the symptom

I know there is never enough time to conduct such investigations in the real world, but the technique can be applied to components in situ, be that with an external heat source, freezing/moisture sprays, or wiggle testing

“Food for thought” as ever and sorry the One-Waveform-One-Picture has not been strictly adhered to above (i.e., 3 images used, hopefully revealing the relationship between output signal and temperature )

I hope this helps, take care…..Steve

User avatar
Rfmotors1
TwoWaves
TwoWaves
Posts: 86
Joined: Fri Aug 31, 2012 8:14 am

Re: One waveform One picture

Post by Rfmotors1 »

Hi All,
I would like to share one case here on this thread created by Steve as it probably matching the "One Waveform One Picture" criteria, hopefully it's OK.

We had Toyota C-HR interesting complaint in the workshop and the Pico WPS500X transducer showed the potential where all other common methods did not help. Decision based on the result of this test was to replace engine for second hand instead of trying to repair it. (leaking valve would lead to repair for example)

COMPLAINT: Customer complained the engine vibration especially at idle and the diag tester showed misfires on cylinder 3. The shop swapped coils, plugs and injectors with no fix.

PREVIOUS DIAGNOSIS: The cylinder analogue compression test result was 6 Bar on cylinder 3 and other cylinders were 10-11 bars at cranking with TB wide opened. The in-cylinder leak test did not reveal any leak, all cylinders had only about 10 to 15% leaking which was good result.

PICO TEST: Engine at idle, pressure transducer in cylinder 3 and then 2, each at one time. Good cylinder 2 RED overlayed with bad cylinder 3 GREEN reference waveform for comparison.
Pressure difference 3.87 Bar, cyl 3 lower pressure but the valve timing was identical for both compared waveforms.
Cylinder 3 and 2.jpg
RESULT: The reason was bend conrod on cyl 3 resulting the piston did not compress the gas enough as it did not reach TDC.


Original waveform:
Cylinder 3 and 2.psdata
(3.83 MiB) Downloaded 271 times

Thanks for reading,
Roman

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 »

I have to say Roman, that picture of the Con Rod made me smile and I used it as "Picture of the week" some time ago in my weekly report to Pico.

Moving on:

Following on from a recent support request, I thought this case was a nice fit for one waveform one picture. I will add the finer details in a larger post ASAP as there is more to this story.

Here we have a 2016 Mercedes Sprinter 2.1 litre 314 CDI with Engine code 651.955

Long story short, the vehicle had a severe lack of power and a fault code for Exhaust gas differential pressure sensor “Internal fault” P245296

A new Mercedes Catalyst/DPF had been installed due to fracturing at the turbocharger along with a new Mercedes differential pressure sensor which utilizes the SENT protocol

DTC P245296 could be erased but returned immediately upon road test where a distinct lack of power was noted

Given the symptoms above, verifying exhaust back pressure using the WPS 500 pressure transducer removes all the variables of scan data interpretation and serial decoding by returning a physical pressure measurement

With the WPS fitted upstream of the DPF a road test was carried out

Below we have:
• Differential pressure sensor (SENT) Channel A
• Digital MAF Channel B
• Exhaust back pressure pre DPF via WPS500 Channel C
Image1
Image1
Referring to the measurements alone, we can see our exhaust back pressure prior to the DPF at 2.242 bar at WOT with a minimum of 590.9 mbar which in itself is excessive

Based on the vehicle history, a blockage downstream of the DPF had to be the cause and the images bottom left confirm the SCR/ASC located in the tail section of the exhaust had indeed become blocked?

I will create a full case study surrounding this vehicle ASAP but the takeaway for me with the above was the minimal intrusion of the pressure transducer allowed for graphing of exhaust back pressure against MAF; returning objective results without the complexity of serial data interpretation or worst-case scenario, misinterpretation

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 »

When new talk about DC/DC voltage convertors you would be forgiven for thinking EV

In the example below we have one such DC/DC convertor fitted to a 2016 BMW 535 GT F07 with the N57 engine and no sign of high voltage components; so why a DC voltage convertor?

On the above vehicle, the DC/DC convertor provides support to the electrical systems during periods of Stop/Start (i.e., to prevent dramatic and momentary fall in battery voltage during cranking)

Typical circuits that require “support” from the DC/DC convertor include Audio/Video system, Navigation/Telephone, and interior lighting

So, how was this unit at fault?


Two words “Parasitic drain”, in the image below we have captured the parasitic drain on the vehicle using our 60 A BNC+ clamp https://www.picoauto.com/products//curr ... rent-clamp
Image 1
Image 1
Note above how channel A captures parasitic drain returning to the 12 V battery whilst channel B captures current flowing into the DC/DC convertor on the Red/White (RW) wire from the fuse box

Channel C is monitoring the 12 V battery voltage whist channel D is our 2000 A BNC+ clamp also mounted around the battery negative lead. Why 2 clamps around the same battery lead?

Channel A is using the low current clamp on a fixed range of +- 2 A to ensure we have good resolution and accuracy sufficient to capture parasitic drain at a milli-amp level

Channel D (+- 20 A) is there to catch current spikes or power up events such as remote locking etc. as this will most certainly exceed the +- 2 A limit set on channels A

Note the “Mean” value of channels A (4.456 mA) and channel D (293.8 mA) between the time rulers. All things considered, whilst we could not rely upon the 2000 A clamp at such low current flow, I do not think it has done a bad job at all. More on the BNC+ 2000 A current clamp here topic22430.html

Focusing on the sudden drop on channels A and B around 45 seconds, this occurred upon physical disconnection of the large 4 pin (N6*2B) DC/DC Convertor connector

The fact that both channels A and B fell to 4.456 mA and 27.94 mA respectively (at identical points in time) places our parasitic drain with the DC/DC Convertor

The discrepancy between channels A and B can be attributed to the inherent drift within all current clamps. To find the true value from a clamp that may have drifted, note the parasitic drain (mA), remove the clamp and note the new value. Subtract these acquired values from each other to find the actual parasitic drain

For example:

• Channel D displayed a parasitic drain (between the rulers) of 27.94 mA
• Remove Channel D clamp from the cable and note what should be the current clamp “zero”, value acquired + 20 mA (Our clamp has drifted "positive" from zero to +20 mA)
• Subtract 20 mA from 27.94 mA to find our parasitic drain of 7.97 mA (Our target is < 80 mA)

Returning now to the faulty DC/DC Convertor, note how the thermal image camera confirms this device to be “warm” along with further confirmation after dismantling to find water ingress!

For further parasitic drain information we have the following links:

https://www.picoauto.com/library/automo ... tic-drain/
https://www.picoauto.com/library/traini ... itic-drain
https://www.picoauto.com/library/traini ... in-testing
viewtopic.php?p=96541#p96541
https://www.youtube.com/watch?v=20hByAi ... Automotive

I hope the above 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 »

Following on from the previous post on parasitic drain, here comes another one!

To begin with I can take no credit for this post. All the hard work was done by Paul Foster who kindly sent over the article. I’ve just purely put the picture and the waveform together. Enjoy!

I wanted to bring to your attention an interesting case study involving a Polestar 2 vehicle's low voltage battery performance.

The Polestar 2 in question has been experiencing a low voltage battery discharge concern, with the battery going flat after approximately four days, despite the vehicle's Amp hour rating being 70Ah. If we worked on the rough guideline of a vehicle pulling around 40mA once fully shut down, a 70Ah battery should be able to last approximately 1750 hrs. Dividing this by 24, gives us approximately 73 days. These are basic maths and don’t take into account random power ups that a vehicle may encounter during a sleep period, temperature changes and other factors that can affect batteries. Needless to say, it should last longer than 4 days.

With initial measurements made we could see that in a period where the vehicle should be shutting down it was still consuming, on average, 2A. With a constant 2A draw, a 70Ah battery will last just under a day and half.

Using a 0.1 ohm resistor, a parasitic drain test was carried out using PicoLog as per Steve’s video - https://youtu.be/20hByAiW-UQ?si=xXwnHFTOnE_EKcoI. Once set up we can just leave PicoLog running whilst looking for the issue.
PicoLog Polestar.png
What we observed was that the keyless entry system on the driver's rear door handle remained active, causing the vehicle to believe that someone was attempting to enter by pulling the door handle. This continuous activation was preventing the vehicle from entering sleep mode, which, as you know, significantly affects battery drain.

In an attempt to address the issue, the connector for the door handle was disconnected, and the vehicle appeared to go into sleep mode with current consumption being significantly reduced. It is worth noting that the infotainment head unit (IHU) on this vehicle can take around 12 hours to go to sleep. This is likely to be similar to other vehicles where certain ECU’s may take longer than expected to shut down. There are also ECU’s which may power up at set times during a sleep cycle. For example, a telematics ECU may switch on to send out vehicle data which isn’t a fault.

After leaving overnight, there were no further changes in the current draw which when measured between 30mA and 10mA and the battery voltage remained stable throughout. Happy that the fault had been found, the door handle was replaced. With PicoLog still running the vehicle was left again to ensure it went back to sleep.

I hope this helps

Post Reply