little.red.skittle

Alright, the leak down tester arrived and I tested the engine... took it for a drive until it was completely warmed up to temp then a little more. Pulled it into the garage and started testing
Test Results:

Cylinder 1: 4%
Cylinder 2: 17%
Cylinder 3: 32%
Cylinder 4: 4%

So obviously this is a massive issue with the engine, during the test I heard lots of air coming from the exhaust that I could easily feel during cylinder 2 & 3, and just a little bit from the oil cap but barely
Now the machine shop I used is very trust worthy, and well known. I've had work done there before, and the head was completely rebuilt. However I did the valve lash myself, so I am hoping the exhaust valves on 2 & 3 have negative lash and thus are leaking. I will probably check for that tomorrow and if I'm lucky that's the issue. Please let me know if negative lash would have that effect

curly

iTrader: (25)

Sorry, had to. Side note, if your m54 has similar leak down issues, it's enough for BMW's DME to cut fuel to that 32% cylinder during runoffs at Road America. I didn't build that engine. Second side note, the m54 I did build finished 15th overall. Have fun rebuilding.

little.red.skittle

I mean alright

technicalninja

If you actually have zero lash you're screwed. The valves are burned and require re-cutting.
You don't, lash will not be your problem, you have reasonable compression test results, if you had zero lash your comp test would be ****.
Did you read the bit in my leak down test thread regarding the legends car? Perfect leak down and bad compression is a valve lash issue.
Your engine might be fine... with one exception.
What didn't occur is the valves were not lapped in. Fancy machine shops commonly use SERDI valve cutting equipment.
SERDI advises their customers that hand lapping the valves is un-necessary and can lead to issues.
The machines will commonly cut the face angle and the seat angle a tiny amount differently. Like a 15 minute difference.
They are hoping that the valves will beat themselves into a good seal.
If that doesn't make sense GOOGLE angles below 1 degree to educate yourself...
This is HORSE ****.
I built a MSM for a customer that I had a machine shop do the whole head on. The only machine shop I trust...
Normally I just send them the head and valves, nothing else. In this case I didn't have a nice supply of Mazda shims to complete the final valve adjustment.
I sent them the whole shooting match. I took my own feeler gauges to "verify" the adjustment before I left the shop. They are a 45 minute drive away from my shop.
The valve adjustment was off on both intake and exhaust at delivery. I brought it to their attention and they re-adjusted 16 valves on the spot. Measured, pulled cams, changed all 16 shims and replaced cams.
No harm no foul....
Right...
During assembly of the head onto the totally stock used 155K motor I did a leak down as soon as I had the head studs torqued to 40lbs.
Bad leakage on the exhaust valves, minor leakage on the intake. It helped diagnosis having the exhaust and intake manifolds off of the head.
My worst leakage rate was 25% and the best was 17% **** POOR!
I ripped the head back off and dissembled. No lapped valves, I didn't really expect I would see lapping evidence anyway...
After I hand lapped the valves and re-assembled I had cold leak down readings below 5% on an old bottom end.
My lapping the valves reduced my valve clearance readings by .0005, 5 tenths, or 5 ten thousands which ever term you want to use.
It tightened them up a tiny little bit.
I retested the car after initial break in 1k miles in.
180 across the board- no variation at all and leak downs in the 3-7% range.
Not perfect for a full on race motor but very good for a hot street car with a used short block.
Car has 25k more miles on it now and the owner (guy who has been a customer for 3 decades and handles my website) occasionally still calls me to say "thank you for my Miata, It's still running great!

My failure in this situation was relying on another person to properly do something I could handle because I didn't have a proper set of shims. A "money" decision.
The wrong one...
If you don't do it yourself and you don't check the work that you sub out you deserve what ever results you end up with.
That machine shop you used most likely had a yellow tag on the work that said you and only you are responsible for the finished product and if you find any issues you have to bring it to their attention BEFORE assembly...
You have to be better and more precise than they do...
You have to really know what you are doing. You are young and in-experienced and didn't know this before.
This is why learning this stuff can be expensive.
The single most important thing a beginner needs is experienced oversite...
You need people like Curly to help you and it looks like you might have pissed him off.
You need to pull the head back off and hand lap the valves, this will most likely solve your issue.
This still does not fully explain your original problem.
It may be something else and this just happens to be another facet in your education.
I'm out. Good luck!

little.red.skittle

Thanks for the insight! That is really bad news for me. But I guess I'll try that. I don't entirely understand all of that and how it works. I figured machine shops would test leaks after rebuilding the head

curly

iTrader: (25)

I ain't pissed, just gloating that I was right.

Basically I've been around long enough to read these posts as this: nothings wrong, but it's down on power.

Which tells me one of those things isn't correct. I'm assuming you picked identical roads and VD settings, therefore something is wrong.

little.red.skittle

Yeah, I understand that. I did chose same settings and roads.

Also update: pulled the head last night, measured all the valve lash and then removed the cams and measured the shims again. I am pretty sure I accidentally subtracted the difference that I need from the shim instead of add because all the gaps were at least twice the size they should have been. Oops. So remeasuring and hopefully I have enough shims to get it right. Then I ordered a head gasket and I'm lapping the valves hopefully today or tomorrow, just the exhaust valves in cylinder 2 & 3

technicalninja

And yet another BIG mistake...
I would NEVER go through all the work/parts required to pull the head and ONLY lap the valves on your currently bad reading valves...
Lap ALL of valves, you have 16 lapping procedures to do NOT 3!
And this thread take a depressing turn, it's not fun anymore...

You NOW need to do a bunch of video watching, thread chasing BEFORE you lap the valves to know how to do that crap correctly.
I have lapped 250-300 sets in my life and the first few I did were NOWHERE near the quality level I can achieve now.
This learning cannot be done in an hour...

I also take lash measurements and do the valve adjustment with the head bolted to the engine if I can.
I'll set a new head up on the bench before installation BUT I will recheck the measurements before I fire it.
If I'm lucky I will not have to change a shim; most of the time I'm lucky but sometimes I'm not...
I catch that **** before I send it...

Curly, glad you're not pissed; you seem to be a good guy. He's gonna need you...
I apologize if I mis-spoke regarding "only work on Miatas" stuff. I was trying to make a point with the OP.
You obviously kick as on German **** as well. nice results.

He's your puppy now!
Enjoy...

andyfloyd

I just did my lash on my fresh head and plan to check it when I bolt it in the car, how much have variance have you seen after bolting a head down?

TalkingPie

Giving OP a prop cat for owning up to his mistakes, learning from them, and really, how many guys his age are rebuilding their own engines?

andyfloyd

For sure, I think OP is as legit as a young dude can get and deserves help at this point.

Dr.Sep

I think a lot of people are afraid to come to MT and ask questions.... for better or worse, thats definitely the reputation here. On one hand, kinda cuts down on the repeated n00b questions, but I also know how much it sucks to try to ask a question and just get crapped on because my background wasn't as mechanically extensive as someone else who has been with the platform for decades or the search function turned up hundreds of loosely related topics.

little.red.skittle

Thanks everyone for the help, I am trying to be very good at these things and hopefully own or be a part of a speed shop at some point. I have the head off the car, I already lapped the valves on the exhaust side of cylinder 2&3. I am going to bolt the head up to the engine with a fresh head gasket torque it down to about 40 ft/lbs and redo the leak down test before reassembly. I understand if I was highly experienced in lapping valves and such I would lap all of them but right now I'm more confident leaving what I know is good. And research has told me from sources such as real street performance, and hp academy that lapping valves is more of an old school technique and a fresh cut with very light lapping just to double check is the best strategy. So because of that I'm leave cylinder 1 & 4 alone and who knows maybe I'll have to take the head off again, but more experience I guess. I'll give an update when I'm all done and the new shims are in

little.red.skittle

Update after lapping the cylinder 2 & 3 exhaust valves I mounted the cylinder head and did a leak down test, cylinder 3 was at 8% leak down and somehow mad cylinder 2 worse at 22%. So I took the head off one more time, lapped all four exhaust valves again extra good mounted the head one last time now all leak down numbers are 5% or less within like 2%. So I would say I am good to go and should hopefully be able to crack the 300whp mark if stock ignition will let me

Mr Plow

We have always checked valve lapping with head upside down, an air line sealed into the port, valves held closed and liquid in the bowl. Fire air into the port, leaks show up as bubbles, experience helps.

Have never seen a head come back from a shop that didn't need lapped and where the difference in the test I describe is very clear. Once the test is good; we have never then seen a sealing issue in assembly from lapping.

Given the issues experienced - it would be madness to not go through re-lapping all 16 valves thoroughly to save further pain down the road. It is simple, worthwhile investment.

little.red.skittle

Sort of an old thread however I figured I would update it so it is useful in the future. When I lapped the 2 valves I also moved the shims around and used the stock ones that came on the head. My valve lash was dramatically tighter however still anywhere from 30% to 50% too loose. This made the car pick up about 25whp through virtual dyno and the VE at high rpm was increased. I assumed that the valve lapping was the reason, however after that I removed the head again and took it back to the machine shop, they inspected it and did a leakdown test, they said the head was sealing perfectly. At first I was confused however after thinking about it I realized the valve lash being so far loose would create less lift and duration through the camshaft. This is why the virtual dyno plot looks the same as an engine with a camshaft limiting power but making very good low end torque. So I finally found a source for new shims and will have those installed tomorrow and hopefully the car will finally break the 300whp mark