G'Day Tom and members,

Tom -  I recently had a problem SR5 and wanted to capture secondary but was unable to get a stable pattern.
The secondary probe pickup, obviously larger than the HT lead type, had to be twisted to one side for me to enable a capture. Added to that, at higher than idle rpm the screen would freeze.

If I positioned the pickup to come into direct contact with part of the HT lead at least I could achieve an idle secondary waveform and this was stable.

The leads are genuine Toyota and there were no problems with secondary for any cylinder other than I needed to capture a specific event at a loaded rpm on one cylinder.

Trouble was, when I tried to do a dyno run, ( with me being the only person in the shop ), I found my arms were not quite long enough to hold the pickup in situ so I could capture a specific event under load!  ;D
Just kidding!

Ever had a problem when attempting secondary captures with these thin HT leads? Can I pack the lead out with aluminium foil as to enable a better fit of the lead to the pickup probe?

Cheers!
Kevin

Hi Kevin,

I have not encountered that issue.  I think using tinfoil to shim the clamp around the lead should be very helpful.  Try that and let us know.  

The other issue I'm not clear on is if you are placing the pickup on the plug wire or the coil wire.  With distributor ignition, you will only see the complete secondary picture on the coil wire.  You can then trigger off of a plug wire using an inductive clamp on the external trigger port, or the other channel, so you can focus on the cylinder of interest.  (Note:  the capacitive secondary pickup will not trigger the external trigger because the output is negative and the external trigger is a fixed positive trigger at around 2V.)

Anyway, if you are on the plug wire and the plug breaks down under load the signal may fall below your trigger point and you may not see much useful info anyway.

Hope this is helpful.

Tom,

CHA for Coil HT lead and CHB used for individual cylinder monitoring.

Experiments on a range of vehicle manufacturers utilising dizzy ignitions yielded varying results but useable individual cylinder captures during a loaded dyno run just didn't happen!

Distributor contacts ignition and electronic distributors are still plentiful here and I had hoped I'd find a way to obtain some captures for training purposes to help out our youngsters.
Our big box scope is analogue and although the time base is expandable for viewing burn time etc. I can't save anything.
Two DSO's do help, but again I prefer the PICO for "getting up close and nasty"

Last but not least ... Is the Mac Vandenbrink COP adaptor useable with the PICO?

"Hope this is helpful."
Yes Tom .. as always, and, I thank you for taking the time.

Cheers!
Kevin

Hi Kevin,

If you email me some psd files of your captures, I can take a look at your settings and see what you are seeing.  Might be able to offer some suggestions.

Mac's COP III works fine with the Pico.  It's only attenuated 10:1 though, so you won't see the full primary spike.  The primary spike has no real diagnostic value anyway, so it's not a big loss.   ;)

http://www.autonerdz.com/images/tar7.gif

Here is a capture of a 3.0 Chrysler firing order 123456.  Channel B (top) is an RPM 80 inductive clamp on #1 wire.  Channel A is coil wire secondary parade.  One of the disrtibutor towers has a grounded test light in it instead of the plug wire.  The game is to identify which one.  

Triggering this way divides the sample rate in half on the 212/3 because we are using both channels.  A good altenative is to use the inductive clamp with the external trigger and then using only one channel.  This gives full sampling to the secondary capture, while still triggering off #1.  You cannot use the capacitive secondary pickup on the external trigger though because the ext trigger is a fixed 1-2V pos trigger.  The secondary is a negative spike so the ext trigger won't see it.   :P

Do you want us to answer which lead is grounded through the test light or E-mail you the answer?   ;)

Also, I wanted to point out the nice detail of those patterns, even though your at a relatively slow timebase.  The coil turn on oscillations and the turn-off oscillations are both there...some scopes won't even display this unless the timebase is sped up, but then you can only look at one or two firings.  This would be cool to zoom in on the burn lines, turn-on and turn-off oscillations.  Later, Matt.

Hi Matt,

Feel free to answer.  I don't want to deal with all the emails   :P

High resolution detail  is what PicoScope is all about.  As you wish:

http://www.autonerdz.com/images/tar8.gif

This is #1 with the inductive sync trigger.  I had the channel B vertical zoom on x5 in the last picture  :o
Here, I have put it back to x1 and zommed x10 horizontally.

If you would like to see any of the other cylinders this close I will do the same with them.  

Because we are using two channels and with the sample step used for this time base, we have 9385 samples in this capture for each channel.  This is 46,875 samples per second for each channel.  Had we used the external trigger and one channel, this would be doubled.   ;)

You get detail by simply collecting more samples.  

Well then, I'm going with cylinder #4.  It has a lower and longer spark line.  I would like to see that one zoomed in.  The zoomed sequence you posted looks good, to me.  It is a tad difficult to see with the 20ms/div timebase, but it still sticks out to me.  As a dumb question, why isn't the firing line lower?  Were there issues inside of the cap also?  Thanks, Matt.  Give the full details weather my answer is right or wrong, please.   :D

I had better quick add that I know that the firing line represents the "greatest gap", so if you ground a spark plug wire and the kV drops, the gap would be greater in the cylinder as apposed to the rotor...so in your example, the greatest gap was still in the cap?  I might be letting all this theory run me around in circles, but this is stuff I need to know.  Thanks again, Matt.

Very good, Matt.  You are correct.  Here is #4 zoomed:

http://www.autonerdz.com/images/tar9.gif

There may have been a little cap and rotor wear but not bad.

If you were using PicoScope, you would have simply zoomed and then scrolled through the capture to see them all up close like this.  Good eye  ::)

Yes, the 'Greatest Gap' is in the cap on cylinder #4.  At this time base using both channels though you can't rely on capturing the full kv spike every time.  You have a sample every 20us.  Firing lines can be only 10us.  So, we are really pushing it here.

Hi Tom,

I have e-mailed 8 PSD files ( 600kb ) to you so you can examine them.

Cheers!
Kevin

Hi Kevin,

That's a nice bunch of captures.  You are a quick study!  Here is one we thought would make for a good discussion.  I'm off to the Vision Conference tomorrow but will post zooms on any areas, as requested, when I get a chance

http://www.autonerdz.com/images/kg1.gif

Note the trigger settings.  He has used the external trigger on #1 to devote the full sample rate to one channel while still syncing the capture for cylinder ID.

He has his samples turned up to 32.000 so at 5ms/div, he is sampling at 375,000 samples per second.  This means that there is a sample every 2.7 millionths of a second.   This is fast enough to display every max kv event and all the detail needed.

For comparison, a Fluke 97/98, at this time base, would be sampling at 6104 samples per second (true sample rate)  Pico is over 61 times faster!

Ok, enough math.  My head hurts. >:(

I'll let you take it from here, Kevin.  The game is, what's wrong with the car?

Also for comparison, my Tektronix THS720a would be sampling at 50kS/s w/o the peak detect activated, 50MS/s w/ peak detect activated....but I only have 2500 record points compared to your 32000.
Lookin at that waveform, we have some lean issues.  The second and fifth patterns are the worst with the first closed behind....thinking some restricted injectors.  Coil seems plenty strong with about 9kV and 1 ms duration.  Am I close?  I'd like to see a zoom on any of the previous mentioned cylinders, but my mind is still made up...I'm looking for why they're going lean.   ;)

Wow, it says misfire (singular) present I suppose #1 could be lean from the pattern but the rest look like cylinder turbulance (engine mecanical or vacuum leak) to me but can't be, not on all cylinders? EGR probelms? But I'd be surprised if the EGR did that to all cylinders on a snap throttle. I have seen some Ford COP coils with carbon tracking inside the coil cause a pattern that looks like turbulance but I assume this nissan (no year or model) is 1 coil? I'm anxious to know the answer.

Matt,

I call 'real sample rate' the number of samples per second that can be captured and shown to you.  Peak detect strategies are designed to compensate for a lack of 'real sample rate'.  Yes, the Tek is sampling at 50ms/sec in peak detect but it is only 'capturing' a very small percentage of that information prioritizing the minimum and maximum samples.  You are still capturing the same amount of information, it's just different information.  'Real sample rate' is the same with or without peak detect.  The strategy works very well for picking up on transient events though, and is a good feature when needed.

Misfire,

Just a thought...combustion turbulence is much higher under load.

Here are the zooms Matt requested:

http://www.autonerdz.com/images/kg2.gif

http://www.autonerdz.com/images/kg3.gif

http://www.autonerdz.com/images/kg4.gif

Hmm, now looking at those, I'm thinking carbon tracking or arching outside the combustion chamber...especially on cylinder #2 with the "shark fins" on the spark line.  That seems like a tell-tale on arching.  Why, the previous image, did the spark lines ramp up so much?  I usually only see that on lean conditions.

There is well less than 1 msec of burn time on all these blowup samples. If I had the car here I would probably current ramp the coil to cover my bases but my gut tells me it's a cylinder turbulance issue. Notice how the burn starts and uses the available fuel then restarts over and over. The avaialable fuel in these cylinders does not look like it is distrubuted or atomized evenly rather it is in uneven pockets causing the burn to start and stop over and over. Could be caused by an EGR open when it shouldn't be causing the intake air to tumble in an abnormal fashion.I guess you could call it lean in a sense but a classic lean pattern (a fairly smooth upward sweep) looks nothing like this. Of course talk is cheap and I've been wrong before but that's my 2 cents worth.  As far as what is wrong with the car we'll have to wait and see what Kevin says.

I'm no secondary combustion analysis guru.  We do have one in our group though.  Mac?

I have a few thoughts.  All the spark lines begin at about the same voltage.  Firing lines do not start high.  A lean mixture would show a high firing line as well as a higher start for the spark line.  Cyl #1 in particular runs out of available HC and extinguishes in only 430us.  That one is probably misfiring.  

I initially thought lean but It's not 'lean' because the spark strikes at a normal kv level.  But we do run out of HC early and have a lot of turbulence.  The engine has to have a restricted breathing and inhaling less air as well as less fuel.  Perhaps a lot of carbon on the intake valves restricting intake and creating turbulence as they do not seal well consistently and disrupt swirl effects.

Ok Kevin.  Tell us the story.  

During a snap throttle test, I would expect to see the firing line higher. It appears as though there is not enough energy left to maintain spark duration (firing time). #1 shows a definate roll off of reserve energy. Look at the burn time on #3 in the paraded pattern. Lack of complete combustion there! Looks like others are lacking as well. #2 might be a suspect for carbon tracking or leaking boot. The misfire during that capture points towards #3 to me to be the worst. And it appears as though there are other cyl's not contributing like they should.

I'll go with low coil output and carbon or boot tracking.

Hi Carl,

I agree that during the initial snap the firing lines should have been higher.  I'm thinking that because he was not triggering on high kv on the channel being used, that he may have missed the initial spike as the engine gulped and the capture is after that event.  If it was a weak coil, I would expect to see the firing line short but not rising in voltage.  The rise in voltage says HC depletion to me.  

Here is a weak coil:

http://www.autonerdz.com/images/tar10.gif

The HC is still present when the spark runs out of energy so the spark line voltage remains at the same level.  This one is at idle though, so there is little turbulence.

Here is Kevin's #3 for you:

http://www.autonerdz.com/images/kg5.gif

Interesting capture huh.  That's only one of them  :o

G'Day Guys,

First up ... Good to see Tom initiate this 'diagnostic game' and the great participation by forum members so far. This is 'what it's all about' and I'm sure each and everyone of us will benefit somewhere along the line.
Not only Pico capture techniques, captures interpretation but also diagnostic methodology could also be discussed during the 'diagnostic game'

For the purpose of keeping things relatively simple, let's use the capture and the vehicle it was obtained from, as a sort of 'general guideline' if you like, to the Pico useage, screen capture interpretation and possible diagnostic strategies in this particular instance.

As with any diagnostics, something should be known about the vehicle symptom or symptoms and the vehicle ignition, fuel and air systems.
This particular vehicle is of our local production/consumption, but I'm sure many of you can relate to it with ease.

THE SYMPTOM : service R/O was written thus - " runs roughish and lacks power "
mmmm ... front desk needs some educating me thinks!  ;D  

Vehicle general info:
300,000km on speedo
In line 6cyl - firing order 153624
external modular coil
optical ignition - cranksensor 120° and 1° signals
multiport injection - current controlled injectors
2 x fuel pumps ( intank pre pump, external pressure pump )
Bosch Hot Wire Air Mass sensor
3 wire heated O² sensor
No EGR system
No data stream - DTC's only available

I approached this firstly by confirming the symptom before attempting diagnostics/analysis.

Confirmation of the symptom proved out.
Bad multiple mis-fires and poor performance.

A check of DTC's showed 1 stored - 14 (speed sensor )

I chose to connect the Pico and sample the tailpipe emissions with the 5 gas analyser with the thought that the two could support my initial 'confirmation of the symptom' as a possible fuel related issue.

I chose Kv parade to give me an overall view of all the cylinders and the 'effect' of the symptom.

As most of you have ascertained, the capture is showing a ' leanness' issue during combustion and it pertains to all cylinders.

The capture shows the effect but not the cause.
What would you do next?

With regards to the cylinder Kv's in the capture .. Tom has quite rightly noted that the initial higher Kv normally seen with 'lean' during a snap throttle was missed because of my scope settings.

What could/should I have done during initial scope setup before the snap Kv throttle test?

Would using the ' Save On Trigger ' scope function have caught the event?

Would you have approached the symptom with a different diagnostic/analysis strategy?

I won't provide the answer to the 'effect' of the fuelling issue just yet, but, the 'cause' was multi - layered  :P

Cheers!
Kevin

    These are two captures I made a few years ago when pico was new to me. I was playing and didn't have the proper settings but was just looking for a pattern. These are from an early 90's BMW Coil On Plug ihnition primary and were taken at idle. All the cylinders were the basically the same. These two are the same capture the second being a x10 blowup. The issue with this car was extreme carbon buildup on the valves causing some of the fuel to be soaked up into that carbon resulting in a lean misfire. A valve job fixed the car.
www.members.aol.com/bgoode007/bmw1
www.members.aol.com/bgoode007/bmw2

Kevins added description of 300,000 miles and no egr would lead me to agree with Tom, Carboned Valves?

Hi All,

Those that identified with a carboning or restriction of cylinder breathing are on track. However,
at this stage I still needed to identify totally the lack of HC during combustion.

A current ramp of the fuel pumps provided yet another clue. The pre-pump waveform exhibited a poor commutators/brush signature while the main pump signature showed an expected 5amp, good speed and reasonable signature. The pre-pump is the fuel volume to the main pump so now I knew I had a fuel volume issue as well.

The Bosch Hot Wire Air Mass meter relys on PCM controlled 'self cleaning' and uses the speed sensor in the controlled hotwire burn off process after ignition shut down.

Looking at the waveform signature of the air mass signal output during snap throttle, showed a woeful signature and initial voltage rise. Not to hard to get to see the condition of the sensing element on this model and I confirmed a dirty sensing element.

Re-connection of the speed sensor connector and confirmation of PCM control of burn off was verified.

Whilst I did not mention O² sensor operation and fuel control earlier, the system was in control albeit rather poorly.

So ... I had a 'domino theory' diagnostic and effected repairs had to be done accordingly.
Intake valve restriction was my last stumbling block. With a high miler vehicle such as this one the client opted for an upper cylinder de-carbon service.

After the repairs, the vehicle operation and engine running during all modes ( cold start, warm up, operating temp. idle etc. ) was great and not a sign of mis-fire. The dramatic engine power output improvement would more than satisfy the vehicle owner.

Whether this scenario will be effective for some time? I don't think so. Carbon build up occurs for a reason. Valve guide and seal wear? You bet!!

Now ... how about some answers outlined in my previous post! :)

Add to the question list ... can engine speed ( rpm ) be calculated from the screen capture?

Cheers!
Kevin

I'll getback to you on your questions, Kevin.  I'm here in Kansas City preparing for a class at Vision tomorrow.  I will answer when I can give you a real response. :P

Hi Guys,

Nice captures, similar to captures I've seen on a Ford Ranger pickup with a 3 liter V6.

would you agree with the following

the captures you have shown at idle do show a nice spark-line indicating normal combustion and spark duration, because the mix is right. ( pressure and volume adequate for the occasion )

the waves as shown during snap throttle start out with a good mix and turn in to a lack off conductive HC, likely due to a fuel delivery problem, that is likely caused by a lack of volume from the fuel-pump ( volume tested at the return line, since a weak pump working against a pressure regulator will be flushed out )
of-course restricted injectors would play a part here as well, but the low volume needs to be corrected first.

the sawtooth pattern has also shown up when there is excessive exhaust back pressure so the cylinders are not fully charged up with a pure combustible mix of fuel and air.

Hi Kevin,

To reply to some of your questions:

With regards to the cylinder Kv's in the capture .. Tom has quite rightly noted that the initial higher Kv normally seen with 'lean' during a snap throttle was missed because of my scope settings.

What could/should I have done during initial scope setup before the snap Kv throttle test?

You could trigger for the high KV spike on the capture.  You could also use Min Max setting to leave shadows of the previous traces on the screen but I think the first option is better.

Would using the ' Save On Trigger ' scope function have caught the event?

Maybe so may be not.  There are gaps between the screen captures.  Odds are fairly good.

Would you have approached the symptom with a different diagnostic/analysis strategy?

There are lots of approaches and many ideas of what 'correct diagnostic strategy' is.  IMHO, correct diagnostic strategy reaches the correct answer  ;)

can engine speed ( rpm ) be calculated from the screen capture?

Sure, why not?  If you know what a rev takes in milliseconds just divide into 60,000.

Hi Tom and Kevin

If you take the coil turn on time between each event (with out cursors) I measure 8 Ms, if you multiply this by number of cylinder which is 6 you get 48.Then divide 60000 by 48 you get 1250rpm That's approx but if Tom gives us some cursor up the exact rpm will be found.

Kevin What portion of the snap test is that image from Ie. tps state.  

Hi! Guys,

Romain Apologies for not replying sooner. I was unaware if your question was being addressed to my captures that Tom posted or to that Mis-fire posted.
I guess both could be representitive of the same occurrence? :-/

I do believe you have some merit in the reference to exhaust backpressure.

Tom I believe you have a capture with no injector operation at all? If you could pop it up, perhaps Romain and others could expound upon this.

This particular vehicle comes standard from manufacturer with tuned extractors btw....

There are lots of approaches and many ideas of what 'correct diagnostic strategy' is.  IMHO, correct diagnostic strategy reaches the correct answer  

Sure Tom .... I agree ... but I also believe that learning different approaches may also allow one to gain an insight into why things are done a certain way to expediate the diagnostic process, and, in doing so, to arrive at a quick and logical conclusion, without cutting corners. ( only my opinion matey!  ;D ).

Mike Thanks for the participation. I have tried both your and Toms calculation methods but in real time with other scopes which record rpm, for some reason with the Pico captures, the calculated Pico capture rpm always reads much lower than the actual rpm as measured with the other scopes ( I've tried digital and analogue ).

I've instances of 4 cylinder measured rpm with one lead disconnected, the recorded other scope visual rpm is low, yet, when I check on the capture and use either calculation, the Pico rpm is always much lower!  Guess I'm going nutso or something. :o

The TPS is a simple 3 pin connector, double ( idle/full throttle ) contacts type. It was correctly set and the idle switch contact was in 'off' idle mode at the time of the capture.
If a TPS type failure occurs on this particular model, idle rpm will  be above 1000rpm and base idle cannot be adjusted.

I did verify TPS operation with and when checking the airmass.

Thanks for reading Guys,
Cheers!
Kevin

Kevin,
       I have been using Pico for a research project where I have found it necessary to use it to calculate rpm from the waveforms I collect that display one complete combustion cycle for all cylinders. I think what you are forgetting is that for every rpm of the crankshaft there are 2 rpms of the camshaft which is what you are looking at when you measure an ignition reference from #1 combustion stroke to #1 combustion stroke again. Therefore it is necessary to divide 120,000 by the time in msec this takes to calculate crankshaft rpm. I have found that the Pico is extremely accurate for this. Here is an example of a COP Chrysler with #3 coil disconnected to induce a misfire. The rpm was 720 RPM. Hope this helps :) www.members.aol.com/bgoode007/rpm

Kevin,

Misfire's comment is great.  I neglected to mention what a rev is.  It would be the time between one companion cylinder firing to the other.  From one cylinder firing to the same cyl firing again would be two revs.

If you are referring to my weak spark capture:

http://www.autonerdz.com/images/tar10.gif

The injector is working fine.  There are no performance symptoms with the car at all.  It was just hard to start sometimes.

There is an article with more on this case Here:

http://www.autonerdz.com/clst.htm

The article is 'Weak Spark Case Study'.  For those of you who do not have the articles password, just follow the directions under the eyeballs on that page.

So we don't clutter up this thread please direct comments on the article to the Articles Feedback section of these forums. :)

Hi Tom and Kevin

Your car has a distributor, so I would use the same calculation 8msx4 =48 then 60000/48
-1250 cam rpm x2 =2500crank rpm

Now if you had dis and a crank sensor and used the duration of one reluctor tooth x
number of teeth, then 60000 divided by that number you should get the rpm of the
crank

Kevin when I asked about the tps I wanted to know at what position the throttle was
open at that capture, as in where you just opening so as to have max air intake or just
snapped close as to have full fuel gulp.

I  was not suggesting the tps as a fault.