Help My Guys Understand Pulse Waveforms - Live Broken Vehicle
We recently received a Nicholson Pulse Sensor and we were fortunate enough to have a broken vehicle that lacks compression and we would like a lesson or three from those experienced in this type of analysis to help us understand how to interpret the signals.
We've performed running compression tests on cylinders #3 (good cylinder) and #4 (bad cylinder) and have included screenshots and the raw files below. We're posting this bounty hoping that we can generate some interest by providing an incentive. One of my technicians, Andrew Hernandez will be participating in this discussion.
Thanks in advance.
Is it possible to get a "cranking" capture? It's much easier to decipher a cranking vacuum waveform. When the engine is running it gets pretty messy.
I can get you a capture of cranking vaccum in the morning and post
Those captures are much much better. That being said, I can't say I see anything that obviously stands out. Although it's a funky looking pattern, they all look relatively the same.
I need more time to study them but at the moment I'm stumped. Time to call in the waveform master, you know, the guy whose name starts with a B and ends with a Steckler.
BTW, have you made sure that those compression pattern results are repeatable? Another words, not an equipment or setup issue? Just trying to cover all the bases.
I should have been more clear. I was hoping to see cranking vacuum with an in cylinder capture. With cylinder sealing concerns, it can sometimes be more obvious (In the in cylinder waveform) as the crank speed decreases.
We wanted to see cranking vacuum because that is easier to analyze. That said, I do not see anything going on in the cranking vacuum trace.
Let's talk about the cranking vacuum trace.
I like to see a descending trace as pressure decreases. First look sensors will do that. The ATS Inches of water true transducer will do the opposite but that's the only one that I know of that's used in the auto diagnostic field that does this intentionally. It looks like Mike's sensor is responding opposite to pressure when compared to a first look. There's another company that makes a pressure pulse sensor and it is hit or miss how it will respond to pressure but I think it's not intentional. I think that it happens when the wiring polarity is reversed during manufacturing. I wonder if that's what's happening here or if all of Mike's sensors respond this way.
That is some awesome scope manipulation, thanks for posting that. See what you think of the following capture. This is one of the new running captures posted today with more appropriate filtering. I added a piston position chart and numbered the intake pulls. The problem definitely starts at cylinder 4 but some disturbance seems to continue through cyls 5 and 6.
It's not a bank issue. The first 3 in the firing order look ok and the second 3 in the firing order look bad. It's almost like the cam has snapped in the middle and the rear half has shifted slightly.
I think if I had this in my bay I would want to go in cylinder on some more cylinders.
But, before doing that, I think a Relative Compression test with a current probe, cylinder sync and a vacuum transducer is in order.
I think that would give us a better picture of the overall engine condition.
Thanks, Bob! The running waveform is too subtle for me. The compression does not start until the intake valve closes (60ish degrees after BDC?), so if there is an issue with the intake valve, the effect may go quite a bit further into the compression stroke, but that would just be explaining away something I don't understand...
Excellent, thank-you for taking the time to reply.
What do you think about the flat section of the previous cylinder wave form? The other cylinders all have a series of higher peaks and valleys across that section.
Hi Eric, regarding the previous cylinder part of the waveform -- this cylinder's intake process looks less "turbulent" in general, but the waveform is being quite flat only for that engine revolution. For others, it is still bell-shaped.
Thank you for your brilliant demonstration. We believe that there's an over fueling event taking place within #4 washing down the rings. We'll know more come Monday. Thanks again!
Google is not helping me find the "Nicholson Pulse Sensor". What is it and where did you get it?
It is like a first look sensor, I believe there website is down, I ordered one right when the website went down.
Guess he wasn't charging enough to keep the web site running. ;-)
Hello I spoke to Micheal and he would like to thank everyone for the ongoing support. Due to unforseen family circumstances and technical difficulties with the website. The website has been taken down. If you have any questions please look him up on Facebook or feel free to contact me. He sincerely apologizes and hopes to bring the website back up as soon as possible. Thank you!
Thank you Tomi. I will guess he is "Michael Nicholson"? Hence the Nicholson Sensor?
I assume that channel A is the change in manifold pressure? Is it possible to get a cranking capture of the known good and known bad?
Yes channel a is the Nicholson sensor. I dont have access to a known good but I will post a cranking capture. Did you want a RC test as well?
Here is what I see at the moment but I'm not confident on the accuracy. We can do much better with a cranking pattern.
The cylinder id chart is synced on cyl4. The numbers indicate the cylinders intake strokes. It looks like cylinder 4 compression is pushing into the intake manifold but usually a compression leak will cause a lower exhaust pocket, so I'm not positive on that.
After looking at the compression captures a little more it appears that the cam timing in cylinder 4 is late compared to cyl 3. Since it's a single cam engine I would have to guess that the valves on 4 are opening later due to a worn lobe or something like that.
I'm just spit-balling at the moment. I think a cranking capture will much more useful.
I know you have way more experience with this than me, and I agree that a cranking might be useful. But the only thing that concerns me is that it seems manifold pressure is affected before the intake valve closes on the compression stroke, and I can't seem to find any appreciable signs of leakage (Inconsistent/lower exhaust pockets, towers leaning, erratic compression peaks). especially a 15 psi difference
Valve timing looks consistent in both captures, and I know this is grasping at straws but the intake pull on number 4 looks not as deep (didn't want to use high or low otherwise I would confuse myself :)
The only time I have seen something like is on a 3.8 gm with a bent connecting rod, ran around in circles with that one until it hit me over the head.
of course I can't explain the effect on vacuum later on
If you have any thoughts I would appreciate it , I am always trying to get a better perspective on this stuff.
I meant valve timing looks consistent to me. Wouldn't be the first time I'm wrong
I will definitely post a cranking capture in the morning first thing. Thank you for your time and input.
See my response and new data here: diag.net/msg/m3zwuqb1ne…
Thanks for the input, we appreciate what you’ve shared with us this far.
Where do you find that application you’re using?
I haven't had a chance to look at all the new info yet. I was in a John Thornton class all day today. I'll try to get a look at all the captures tonight.
The program I'm using is a Windows 10 App. "Pressure Waveform Overlays" by Steve McGrew who I believe is a member here.
At first glance, it does appear that the cylinder #4 intake valve leak is pressuring the manifold during its compression stroke.
I would not only suggest an intake pulse waveform along with a synced relative compression test, but also to do so with the throttle body capped, and at least a 10 second cranking duration(or at least until the waveform stabilizes). This will clean up the waveform a whole lot and will make it that much more obvious.
from what I gather valve timing is the same between both cylinders , EVO - 150DEG , EVC 370DEG , ICV 603DEG . both towers are symmetrical also indicating proper valve timing , both expansion and intake pocket on even plane indicating no valve sealing issues . would like to see a cranking waveform to see the low compression ? as I cannot see it on the running waveform
I'm looking forward to seeing a cranking in cylinder and intake capture but from what I can tell, there's no obvious discrepancies between the exhaust valve opening and intake valve closing (less than a few degrees) events to cause the cylinder 4 no contribution concern that you are experiencing. I'm also not seeing any obvious signs of cylinder sealing concerns (Even towers, intake and exhaust pockets like Jeff Colton mentions) but maybe we will see something in the cranking in cylinder waveforms. Some have said that they think that they are seeing an increase in manifold pressure during the cylinder 4 compression stroke but I'm not seeing that.
One thing that I've noticed when looking at exhaust valve opening and intake valve closing, you will sometimes find that they are very close when physically measured with cursors but when you take a 30 foot view and compare both, you will see a difference. The difference that I'm speaking of is when the exhaust ramp and intake ramp cross the 180 and 360 degree marks. Again, I would need more captures before I can commit to a direction and at the point that I commit, I will do what I can to describe what I see happening.
I posted this earlier....this is cranking only. Also did a leak down test of cylinder # 4. Found no leaks. In my opinion the capture I did this morning looks better than yesterday's. I also did another engine running test. The last file is second running test sync is on cyl # 4 on all these files.
I see the #4 exhaust pocket is of concern, as the cam timing relationship to the 180 degrees moves to the right the cam is advance to the left is retarded, there off
slightly, I would focus on the exhaust pocket lack of definition, my question is would this cause a 13 to14 psi difference in compression?
Once you find the problem, the waveform information will become clear
Hi Scott/Andrew, Nice capture. The thing that grabs me about cyl #4 is the delay opening the exhaust valve. Cam timing is ok, so there is either a loose valve, or a worn cam lobe. This is causing the low running compression due to low V.E. (exhaust not flowing/breathing correctly).
I was thinking along the same lines because the cyl 4 exhaust event is definitely happening later than cyl 3. Since this is a single cam engine then the problem has to be cam lobe or rocker arm related. The only problem is that this usually causes an increase in pressure because the cylinder can't expel all the spent gases.
There seems to be some contradictory findings. Very strange.
I wish I could respond better but I've been so busy can't find anytime to sit on front of the laptop. Anyway just a couple pointers. I also have a Nicholson sensor and it works flawlessly. Your waveform seems extremely noisy. I've found that putting a longer hose an hanging the sensor with a bungee cord from the hood striker helps this. If you notice on your scope literally just moving the sensor in hour hand will create a signal. Also I've found during an intake cranking capture I like to connect directly to the intake and use a large port such as the brake booster. Also make sure the throttle blade is completely closed( in other words I'd refrain from clear flood when doing intake cranking). Also in response to Brins comment about the sensor looking like it possibly spikes for vaccum. Yours may be different but mine clearly goes down with vaccum. A quick verification test for this is just suck on the hose while watching your scope. If it goes down your good. And vice versa blow into the hose and it should go up on the scope. Also i would love too see a cranking crankcase capture with the sensor. Your initial in cylinder running capture on #4 looks like valve timing an closing is good. I personally would want to check for blowby or a bent rod. But also haven't opened your cranking files from today yet either.
Okay after reviewing the files again maybe noisy isn't the right word but"busy". Anyway your initial running in cylinder captures have identical vaccum pockets as far as the psi reading goes on 3 and 4 and between the intake and exhaust. It looks like all the vaccum pockets are roughly -10.5 psi if I'm seeing correctly which is -21 inhg which is literally perfect. Because of this I find it hard to believe it has a valve issue. But I could be waayy off lol.
Would have liked to have seen the TB capped version of the pressure pulse. But nonetheless it looks like there is no real cylinder leakage going on here. This overlay is matched at 0-720 degrees, and also matched at 0 psi and 50 psi. looks like piston height may be the issue. Hydrolocked in the past?
You might be on to something there. The in cylinder waveforms look normal except for the low compression in cyl 3. I'm not sure why the vacuum waveform looks like it does but it sure doesn't seem to be due to leakage.
Here you go: diag.net/msg/m3zwuqb1ne…
Interesting point about being hydrolocked as I've seen that in the past as well. We believe that the rings were washed down due to a lower seal leak on the #4 side feed injector. More details can be found in my linked reply. Andrew will be following up as needed. Thanks for the input, it's much appreciated.
If possible could you get us a relative compression capture using an amp clamp, ignition trigger and intake vacuum waveform? I think those together would give a good overall picture.
Sure I will get that up before the end of the day.
Add an in cylinder capture. Turn the filter off when capturing please.
Brin, I wanted to see a relative cranking test with all the plugs in for a baseline look at all cylinders.
I agree Bob! I always start with a relative compression capture when I suspect a engine mechanical concern.
But I was hoping all along that he was going to duplicate his original captures under a cranking condition. My last comment was made because I'm hoping that he'll get starter current, in cylinder pressure and cranking vacuum on the same capture.
Yeah, I guess having the transducer in it won't affect anything since it will be a cranking capture. Might as well get as much data as possible in one shot.
Thanks for uploading that. I put a piston position chart on it and I see a couple things that are conflicting.
First of all, when cylinder 4 comes up on compression the corresponding intake pull is weaker indication the compression is leaking past the valve into the intake.
What's confusing however is that the current waveform seems to indicate that cylinder 3 is the weak one.
I'm not able to make a concrete conclusion at this point.
If you suspect the rings are getting washed down, a pulse sensor on the dipstick tube should help you nail that down.
Hello, Bob Scott did post a waveform from the dipstick. Scroll down . It is in response to Mario .
It is not uncommon to have a slightly weaker intake pull after a low contributing cylinder. I think that it has to do with crank speed. There's less of an air spring effect after TDC on a lower contributing cylinder than there is on a good contributing cylinder.
I don't look at this capture and think that the intake valve is leaking.
I agree . Last week I did a cylinder leak down test on suspect cylinder and found no leaks. What I did find is my injector on said cylinder was leaking an washing down cylinder. Today I will be pulling intake for further inspection and possible replacement and retest. I will try to keep everyone updated.
I spent a little time with Andrew on this today and I'll provide you with an update and let him chime in and answer others individually.
A little background might help. This vehicle was previously in the shop to correct an intermittent misfire - P0302. Andrew performed testing and identified #2 injector as a problem. Current ramping the group revealed low current on #2 injectors compared to the others. After checking the injector circuit and finding it OK, we recommended all injectors. New injectors were no longer available so we opted for remans. These are side-fed injectors.
During testing we observed that the #4 spark plug was usually wet upon removal. We were thinking that the #4 injector might have flow balance issue so we proceeded to perform an injector pressure drop test. We were unable to perform that test because there was very little rest pressure. After clamping off the supply line after cycling the pump off, the pressure continues to drop. We suspect that the #4 lower injector side feed seal is leaking fuel into the intake port and resulting in ring sealing difficulties.
On Monday, Andrew will be removing the upper intake and fuel rail for inspection.
I took a couple of samples as requested.
All cranking compression tests:
Cyl 1 - Blocked Throttle - [NOTE: This file was previously mislabeled as cylinder #3]
Scott thank you for the response. First time I see the cranking capture, makes all the difference. I'd agree with Caleb about doing a relative compression test vs an crankcase pulse capture. After all, not only would it validate your suspicion, but it would also serve the purpose of the original request for pressure pulse understanding, furthermore justifying manifold disassembly.
I came into the shop this morning to perform the tests you outlined. See my result below:
Cyl 1 - Compression, Pulse sensor on dipstick. [EDIT - previously I misidentified the cylinder under test as #3]
I appreciate your input.
Wow Scott these cranking captures are pure gold!! Like I said earlier if you could finish it up with a cranking crankcase capture that would be awesome. Very cool to see this come together.
Hi Scott, I am looking at your last captures, is the red trace intake vacuum? Do you know the firing order also?
The red trace is a piezo pulse (differential pressure) sensor.
Firing order is 1-2-3-4-5-6
I am following this thread as I am new to scope patterns also.
Am i right in saying that you think that you have cylinder wash in cylinder 4 and that has reduced the vaccum pull for that cylinder?
I have studied the three patterns labelled cylinder 4, cylinder 4 blocked and cylinder 3 blocked.
My question is when looking at the cylinder 4 (your misfiring cylinder) scope patterns, the lowest vacuum pull is on the power stroke of the in cylinder pattern and the peak cylinder pressure is higher than pattern labelled cylinder 3 blocked (your good cylinder)?
Looking at pattern labelled cylinder 3 blocked the lowest vacuum pull is at the start of the intake vacuum pocket of the in cylinder pattern?
Am i reading this wrong or are the patterns labelled incorrectly?
Can someone please help me understand this?
Thanks for chiming in. I double checked my cylinder 3 callout and I'm embarrassed to say that this is actually cylinder #1. My mistake and I apologize, for the others following along I'll go back and correct my markups.
Is cylinder 3 capture actually cylinder 4?
Which file are you referencing? You can right-click and copy paste the URL in your response. #4 should show lower compression than that of #1.
In the first file Cyl 4 at the peak of the first compression tower blue trace the red vacuum trace has its smallest fall, I assume that this is your area of concern?
From what i understand this is air drawn into the cylinder, an even pattern is ideal, a decrease in amplitude is less air entering the cylinder (lower vaccum generation of the cylinder).
I would have thought that as you are in the misfiring cylinder 4 that the red trace vacuum would have its smallest fall starting at the 360 degree marker (start of induction stroke) not at the point of power stroke as it is in this file.
The second file Cyl 1 has the red trace with the smallest fall starting in line with the start of the intake vacuum pocket shown in blue, in cylinder pattern.
I think you have labelled these the wrong way around or can someone correct me with this and explain why.
The pull following cylinder 4's compression stroke is cylinder 5. Cyl 5 also shows up in the compression test on cylinder 1.
Based on the time/date stamps, the cylinders under test are correct.
The other thing that's changing here is that we've had the fuel pump disabled on this vehicle along with the #4 plug out, a number of cranking cycles that could have lead towards better ring sealing. I'm a few thousand miles away from the shop this coming week so we'll lean on Andrew to keep us updated.
Let me know if this makes sense or not.
It is possible that the intake pull following cylinder #4 TDC is caused by a change is crankshaft speed after a lesser contributing cylinder.
I see that Bob Powell had picked up on this earlier with his analysis of the low vacuum amplitude during Cyl 4 compression peak could be down to intake valve sealing issue.
Brin i think you have a good point also,
Will wait to see what the problem is, may explain things.......
We learned quite a bit from this post and I thank everyone for their contributions. Although I'm awarding this bounty to Dmitriy, I'm tipping a few others as well.
The issue with the #4 cylinder misfire was due to a leaking #4 injector resulting in washed out rings. Here's the video of the test that Andrew performed (and forgot to connect the return line).
What did we learn?
- The pulse sensor is best used to look at intake manifold activity during cranking.
- Don't over-filter your signals
- Capping off the throttle body can help at times when attempting the capture cranking vacuum pulses.
- You should try to isolate the sensor housing from any sources of vibration
- The pulse sensor sensor should be used on the dipstick tube and measured against a reference cylinder to help identify ring leakage to a specific cylinder
I'll let Andrew chime in to add anything else he learned on this subject as well.