Polishing pressure analysis skills question
This vehicle came in with two concerns. One that the check engine light was on and the second concern was that the vehicle would overheat at idle.
The vehicle had a bunch of lean fuel system codes and low voltage oxygen sensor which I quickly Identified as a ruptured crankcase ventilation diaphragm in the valve cover. That isnt what this post is about however.
The second concern of it overheating was looked at next. I performed the typical tests you'd perform on a cooling system including a pressure test ( no leaks found). I then started the vehicle and wanted to watch the behavior of the thermostat, cooling fans, engine coolant temp sensors etc.
Once operating temperature was reached, I noticed coolant temp continuing to rise. It got to approximately 235 and I noticed the electric cooling fan would not come on. I performed a "percussion test" by tapping on the electric cooling fan motor. Wahhh lahhh the fan is working I'm a magician! I checked connections just to make sure I didn't disturb a poor connection. All is well, Needs a fan,
This vehicle has been overheated many many times according to the customer in our oh-so important interrogation interview when the vehicle was dropped off. So lets make sure the engine was not damaged. The vehicle smells "charred" to me so I had a sneaking suspicion that the vehicle may have a damaged head/head gasket.
A simple block test was performed with the magical color-changing blue fluid and within seconds, it changed from blue to yellow. I know this test isnt the most reliable when it DOESNT change colors, BUT when it does, You know for sure you have combustion gasses entering the cooling system.
OK so, this vehicle needs a cooling fan and head removed for inspection/gasket replacement.
Now to the purpose of this post. Pressure analysis is a fairly new diagnostic technique that we ALL can benefit from and also ALL need more practice on. So, being the nerd I am. I naturally will whip out my scope any chance I get to perform some tests even if I already have a vehicle diagnosed.
So I get out my pico and WPS and make my connections by pulling the hose off of my pressure tester pump and place the hose onto my WPS pressure transducer. I like to suspend the transducer with a bungee to avoid it rattling against engine components and also decrease any noise on the signal.
I have a sync on #1 ignition coil for a 720 degree of engine rotation reference as well on channel B (Red)
As the vehicle is warmed up, I am noticing the flat line of zero pressure I had when it was cold, begin to rise and the cooling system warms up. As its rising slowly, I also notice pulses slowly beginning to present themselves in the capture. The warmer it gets, the stronger the pulses get. Once operating temp is reached I get a capture that looks like this.
To me, It appears that I get a pressure pulse in the cooling system shortly after # 3 Ignition event.
I spoke with a friend of mine Mario Rojas about this after pulling the vehicle out (going to scrap vehicle) and he suggested unplugging the injector on the suspect cylinder (#3) to prove it is the cause of this pressure pulse. Great idea! Wish I still had the vehicle here to do so :(
My question to y'all would be, am I on the right track as far as identifying a combustion leak into the cooling system via this method? Have you done this yourself? Any advice?
Seems like a pretty sound analysis Brian. I think the fact that it was near flat line when cold and then started to pulse when it got hot confirms it. It would have been great to see what happened when disabling the cylinder. Next time right? ;-)
As you've mentioned, pressure analysis is a hot topic and one that keeps opening our eyes to new things and different ways of testing and I've tried doing something similar in the past when I was working at Toyota. We had a car come in which we had already diagnosed as a blown head gasket using the leak detector fluid. However, like so many of us sometimes I just wanted to know more and could the scope have told us anything else. So, when everyone else has gone home I set up the scope as you've done, connected to injector 1 and added WPS500 onto the header tank.
This didn't take long to pressurise either and that is something to be very mindful of if using WPS in this way, there is no safety relief if the pressure goes up quickly! Also make sure you relieve any pressure but hitting the button on the side of WPS to remove it safely before taking it off the header tank. Sorry if it sounds like I'm teaching you to suck eggs but it can catch you out if you're not expecting it.
I've attached an image of the capture I made and using the overlay tool you can see that there is a definite increase in pressure in the header tank as cylinder 4 came up on compression. This was also confirmed one we had the head off that it had blown across into the water jacket.
It's great to always see new ways of testing and this is a relatively non intrusive way to record a head gasket failure and give you the evidence when informing the customer. Hope this helps and keep up the great work I keep seeing on here! I'm learning every day!
Awesome! Great minds think alike. However, you said the pressure pulse exists during #4 compression. How can we confirm its that cylinders compression stroke vs. Potentially the POWER stroke of the cylinder before that (#3)?
Combustion pressure is obviously higher than compression so I'd imagine a failure of a head gasket would display a failure during a power stroke vs. Any other preasure event. This is indeed what the magic fluid is sampling and proven bad as well.
This is why I thought Marios suggestion of disabling an injector to prevent combustion and verify the pulse goes away woukd be a great to prove your suspect cylinder.
I love how this posts start the brain ticking!
Yes I think by disconnecting injectors would change the way the pulses look if we have a head gasket failure.
You maybe right in saying that combustion pressure is higher than compression. However, (I'm just throwing this out there as to be honest I'm really not sure!) would this actually be correct?
I'm just thinking about the piston position during the compression and power stroke. During compression the piston is travelling up the cylinder and as we know pressure increases, and when combustion has taken place the piston is being forced backed down. As the piston is forced back down does this not change and even reduce the pressure in the cylinder as the amount of space available for the gas is increased? I know the capture I've put up is an extreme version, but if we look at the pressure rise and it's position in the compression stroke. Notice how it appears to be closer to the power stroke of cylinder 4 then power stroke of cylinder 3. If we assume that pressure in the cylinder is decreasing during the power stroke I would have expected to see the peak pressure in the header tank closer to the cylinder 3 compression and power stroke.
Really just my thoughts and we are learning every day and I'd love to know more. Not saying anyone is wrong or right here as like I said, I actually can't prove this. What we can say is there is so much more we can do and look at with pressure and I believe there is still a lot more we can do with pressure analysis.
I've been playing around with this using a homemade pulse sensor connected to the radiator using and old stant pressure tester hose and cap. I installed a small tee because I was only interested in seeing a pulse and not pressure. Like above the test vehicle was already proven to have a bad head gasket but I wanted to try my set up and see if I could pin point the cylinder. Like above my scope did show a pulse but I also was torn between which cylinder was actually causing the pulse, the cylinder on its combustion stroke or the cylinder on its compression stroke. Never thought about unplugging the injector. I'll have to remember that next time. I do have captures somewhere but I have to find a better way of saving and organizing them to make it easier to search and find the files I'm looking for.
What were the operating conditions during this capture. RPM, Load etc. Were the "puffs" noticable from the surge tank? Just casually observing the pressure waves. It seems fairly similar to pressure waves that are created by combustion swelling the cylinder. I am maybe unfamiliar with this sync, but why does it not pull to ground?