Prius 2 with a mystery misfire

Hi everyone,
I have a Prius 2 with a 1NZ-FXE engine. I bought this car with a misfire issue amongst several other irrelevant problems. The car was VERY poorly maintained.
Compression test showed (dry/wet): 1. 135/155psi, 2. 150/160psi, 3. 130/155psi, 4. 55/60psi.
Started off by taking the cylinder head off to assess the health of it...ended up ordering a used head that came from a good engine. 
As the project was supposed to be minimum budget I went ahead and after flattening and cleaning I installed the head. I did measure valve clearances, not great everywhere but nothing crazy either. Keep in mind: DOHC so no easy adjusting. New timing chain, vvt solenoid, plugs and coils. Cleaned the intake manifold and bench cleaned the injectors. New gaskets and seals everywhere. Started it up and straight away misfires. Cyl 4 intermittent.
Swapped injectors, plugs and coils, stayed the same. I think this leaves me with compression, wiring and ecu. 
Did a compression test and saw 50/40psi on cyl4. Yes, lower with wet test. Started to lose my mind here.
Leakdown test came out very good for all cylinders. Absolutely minimal leak at TDC on all cylinders. My diag laptop is not with me right now so I only have a few shots from my phone that I can attach.
Pictures:
Pico measurements with a COP inductive probe. (I do have access to a wps500x pressure transducer if needed.)
Spark plugs: left from misfiring, right from good
Techstream data from test drive

I desperately need some fresh eyes on this. 

Thank you for the captures and the data!   It looks like the fuel trims and the ignition waveforms both agree that there's a compression problem.

I really don't mean to be insulting when I ask this, but you removed the Schrader valve from your compression hose when doing the leak down test, right? Seen too many perfectly good techs go down the wrong path to not ask.

Assuming the leak down really is good on #4, there's only a couple ways I can think of that you'd still have low compression. One would be a bent connecting rod. Head gasket failure and hydrolocking wasn't very common on the 1NZ, but could still happen. 

Another would be a cylinder leak that (weirdly) gets sealed while doing a leak down test. I've seen this a few times now, where the pressure from the leak down tester at 100 PSI actually seals the leak you were looking for. You can sometimes catch this by first testing at 40 PSI, then at 100 PSI. The leak down percentage should be the same at both pressures.

You mean like a cracked valve spring letting the valve hang open, then filling the cylinder with air pushes the valve back against seat?

Like that! I've also seen worn valve seats/valves that leak under light pressure, but seal when the pressure goes up.

Not saying your replacement head has these problems, antone, just throwing ideas out there.

Thank you for chiming in, solved it thanks to you.
My leakdown tester has its own hoses without valves so the measurements were accurate. Made a quick measuring jig to compare piston heights at tdc. As I feared no.4 is 8mm or 5/16" lower than all others. I am feeling super dumb for not checking the pistons for a bent rod when I had the head off. Fortunately this is my own project and there is no real hurry. This was certainly a lesson I won't forget. Thanks again.

antone wrote:

Made a quick measuring jig to compare piston heights at tdc. As I feared no.4 is 8mm or 5/16" lower than all others.
 

Wow! That sucker must be pretty bent to be that much lower in the bore.   Nice job making that measurement.

I am feeling super dumb for not checking the pistons for a bent rod when I had the head off. Fortunately this is my own project and there is no real hurry. This was certainly a lesson I won't forget. Thanks again. 

I'd say don't sweat it! Might feel dumb in retrospect, but really, how often do we see bent rods? That don't result in a hole in the block, that is. I don't think I've ever run into one. Hell, I just had the head off a PT Cruiser and didn't check those pistons, either...

Pico measurements with a COP inductive probe. (I do have access to a wps500x pressure transducer if needed.)

Just curious, you didn't happen to use the WPS at any point, did you? I'd be interested to see the in-cylinder waveform on #4 compared to a known good, just for giggles.

Sure did!