Compression testing using the WPS500 Pressure Transducer

Ask any questions about using PicoScope Automotive software here.
Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 669
Joined: Sun Aug 25, 2013 7:22 am

Re: Compression testing using the WPS500 Pressure Transducer

Post by Steve Smith » Thu Jan 10, 2019 4:52 pm

Hi Victor, thank you for the heads up and the additional feedback.

I will look into this ASAP, possibly over the weekend

Take care......Steve

Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 669
Joined: Sun Aug 25, 2013 7:22 am

Re: Compression testing using the WPS500 Pressure Transducer

Post by Steve Smith » Fri Jan 11, 2019 2:08 pm

Hello Victor, just going through your data thank you.
In your first post (Jan 05) you mention calculated relative compression of 71%, 91%. 85%, and 100% followed psdata file “Toyota_Avensis_2004_Diesel_20190108-0001”
Here I can see we have battery voltage against pressure.
Please forgive me here, what pressure was measured and did you calculate relative compression from this data?
How did you calculate these values?
The relative compression test “190104_CompressionTest_001” indicated 1 cylinder at 81 % where I can also see noise present on the voltage signal.
Image 1 PD Rel Com.jpg
If we then compare the results above with the actual values below we can try to determine why we have this anomaly
Below I have taken cylinder 2 measurement from file “Toyota_Avensis_2004_Diesel_20190105-0001” (Nice touch by the way having cylinder pressures for cylinder 1, 2 & 3 on one capture)
Image 2 CYl 2.jpg
Clearly cylinder 2 and 4 are the best performers here (as far as the PD relative compression test is concerned) leaving 2 x poor performing cylinders. I believe with the pressure transducer removed from cylinder 2, the voltage drop recorded across the battery would be the same for cylinders 2 and 4. (As we know the compression test hose and dummy glow plug increases cylinder volume)

Looking again at the PD relative compression results “190104_CompressionTest_001” (no transducer installed) the software has detected 2 x cylinders at 100% and 1 under-performing cylinder at 81 % based on the voltage fluctuations during cranking.

There is mixed news here as certainly the PD Relative compression test has identified a problem (remember this is a first look, non - intrusive rapid test) but it does not reveal the full truth unlike the PicoScope absolute pressure test. I have a gut feeling this is due to the noise/ spikes present on the voltage during cranking.

What the PD test has done is indicate a problem requiring further investigation.

You have highlighted a condition I had not thought of Victor when it comes to relative compression testing.
We know and accept it is a relative compression test and we highlight that if all compressions are low then our results will be 100 % (however our cranking speed will be high)

But what about the effects of noise on the battery voltage and what about the effects of oil contamination to the cylinder.

The noise issue will most certainly effect the results (PD Test) given the software is trying to determine turning points in the voltage whilst calculating average min and max voltage values.

As for oil contamination, this really is a curved ball as we know this has the potential to artificially boost compression so misleading the software to identifying the contaminated cylinder as the best performer and presenting all remaining cylinders as low.

Thinking this through a little further, an exhaust valve that fails to open would be another curved ball as the voltage drop across the battery would be dramatic during one compression event as the companion cylinder would also generate a compression on the exhaust stroke so producing misleading results. The results never the less indicate further testing is required.

Looking now at “140610_CompressionTest_001” which includes the pressure transducer the voltage appears more stable and the results indicate a low compression due to the influence of the pressure transducer.
Image 3 ABS TEST.jpg
Could you confirm the location of the transducer for this test?

This would appear to have had the undesirable side effect of balancing compression according to the cleaner battery voltage and PD test results.

At this stage in your testing had the test conditions started to change?

For example, has the engine temperature changed, battery voltage/support been included, starter motor operation changed (sound or noise symptom) hydraulic valve lifter intervention/activity or oil clearance from the offending cylinder due to repeated cranking?

Moving onto “190110_CompressionTest_001” with 3 x glow plugs removed this is not good as the results in the bar graph view do not represent the blatant anomaly in the battery voltage view. What is intriguing is the cranking speed appears realistic.
Image 4.jpg
At this stage, where we have uneven cranking speeds, or an odd sound to the engine during cranking where the bar graph view indicates uneven cylinders we must select the “Display Raw Data” button as there is clearly a concern surrounding the calculations in this example above
IMAGE 5.jpg
Switching to PicoScope “Toyota_Avensis_2006_Diesel_20190110-0002” we can see the true results which certainly highlight the displayed bar graph error above
Image 6.jpg
I can see in file “Toyota_Avensis_2006_Diesel_20190110-0005” that the compressions have now stabilised, can you confirm a PD Relative compression test to perform correctly now the engine concern is resolved?
Image 7.jpg
We will most certainly review the PD Relative compression test for sure. The Pressure button “not working” has been resolved and under test as we speak. (Thank you Victor and Ben)

I will include this forum post in a bug report to the software team rest assured as we have to investigate the bar graph display where only a single compression was evident (3 glow plugs removed)

The PD relative compression test most certainly remains a “first look” non-intrusive test where engine running complaints are reported by customers.

The following 2 case studies spring to mind, where the Toyota GT86 below produced 4 compression events within 360 degrees of crankshaft revolutions ... -non-start ... fire-p0302

I hope this helps, take care……Steve

Posts: 271
Joined: Thu Jan 28, 2016 7:17 am

Re: Compression testing using the WPS500 Pressure Transducer

Post by victor2k » Fri Jan 11, 2019 3:46 pm

Hello Steve and thank you for your support,
First of all I used 2 different cars ,one in 04 jan 2019(reply 7 & 9),other in 10 jan 2019(reply 10),nothing was resolved(until now) on the first engine.
On the first car(1CD engine) I have the glow plug no.4 broken so I used relative compression,but without the pressure measurement-sensor was installed on the engine but "pressure" button not worked in PD.I changed only laptop(for an older version what let me to use the pressure sensor_at cyl 1 of course),all setup was the same(scope,cables,probes,nothing touched-I didn't know about the PD issue) so I don't know why on the first PD is some noise,on the second PD not,
maybe some LP filters will help us,the entire bandwidth of scope not help here
(in the attached file I used LP filter at 100Hz).
The engine was hot,the battery was charged,don't know about the engine temperature drop,maybe was some degrees.
I used the voltage drop to calculate the relative compression from file "Toyota_Avensis_2004_Diesel_20190108-0001" inspired by this overlap
(324.42 KiB) Downloaded 17 times
also the time for voltage drop give me almost the same fractions,not the same story for the time to increase voltage.Of course the largest voltage drop was noted as 100% :)
On the second car(2AD engine) all tests was performed at 0*C,before the first start of engine(cold start issue on this engine),the battery was charged before.I made some new captures trying to eliminate/confirm my doubts about RelativeCompression from PD.On this car all glow plugs was removed,I measured the voltage drop/starter current-to keep the engine moving,then I put one glow plug back to measure the voltage drop/current increase to compress the air ...I try to find some answers based on the energy conservation and work to compress the gas,until now the math is (not so ) cool (for me) :oops: .
Best regards

Post Reply