Diagnostic post. Guess the noise. Will try to make short, at my wits end with this noise. 2.5% leakdown on all cyl, car runs great.



-fresh professionally built BP4W shortblock .040 bore balanced decked (.040 quench)

-fresh professionally built stock head with Supertech springs



Engine presents with a tapping noise on cold start from 2-3k rpm free reving. Audibly remains until the car is fully hot then goes away to the ear however it can still be heard with a scope free reving and when under load while driving in the same rev range, acceling lightly. My knock sensor is picking this up in megalog. Scope has pinpointed the sound to the top half of the block in between cyl 2 and 3. The engine has been pulled and disassembled twice now. Absolutely nothing obvious on teardown. Everything looks good. An entirely new head (minus cams) has been machined and installed and the sound still persists. Valve lash spot on on both heads. WTF is this noise? Link to the sound - time stamps 1:59 and 3:20

https://drive.google.com/file/d/14IX...ew?usp=sharing

 

Couldn't access the file. Might be me...
What pistons?
What rods?
What piston to bore clearance?
What head gasket? Looking for thickness and/or brand here
What deck height?
Were all the pistons measured for deck height and then checked for "wobble" without the rings?
Do you have pictures of the piston tops after being run but before being cleaned?
On my Supertech 9.5-1 the deck height on ALL of the pistons was different. One was off .005" compared to the others.
A "loose" forged engine can have piston slap noises when cold but those go away during warm up and don't usually create noise after reaching operating temps.
If you had piston to head contact there would be witness marks on both pistons and head.

Your leak down test readings are excellent but not normal. In 40 years I've not seen leak downs all perfectly exact and this is the first time I've seen someone include a decimal point.
My gauge (Matco) has markings every 2 psi (or 2%, they are the same) and there is no way to determine something to the first decimal point.
A more common reading would be all below 5% or 10% and so on. If you can see below 5% on all cylinders hot you are golden regarding leak down.
Was the leak down cold or hot? I normally see a difference between the two. What I care about is the hot reading but if I see a significant difference between hot and cold it usually means a head job is in the near future.
On a fresh motor I'm looking for less than 10% cold and then a smaller number hot. I'd do the happy dance if I hit 2 on anything.
The engines that I have seen leak downs below 5% usually have gapless top rings which Flyin Miata had serious enough trouble with that they don't even try to use anymore.

Start a "howdy" thread and let us know your background (mechanically) and the build level of your car. Lots of serious folks on here that will help.

 

Thanks for such an in depth prompt response! I will add the audio file as an attachment once on a computer. hopefully that clears it up.

Answers to questions -

Supertech 9:5:1
Manley Rods
Piston to Wall - .0040
Cometic .040 HG
Deck Height - 0 deck with pistons just proud of block is what I requested of the shop. They are a high class performance shop that have machined over 30 builds for Jesse Prather, a pro Miata builder/driver friend of mine. I do not have the appropriate tools to accurately measure the deck height myself. I know they mocked up the whole rotating assembly for the decking and did so without rings for checking piston rock/wobble

- only marks on the head are circular marks from the cometic sealing diameter. Piston tops look good, no marks.

-If this is a case of piston slap, it is only happening on one piston. For sure. And if it’s something to live with, that sucks because I won’t be able to acceptably and accurately tune knock control, my knock light and knock sensor are picking this noise up. Even when fully hot. Even in rpm ranges that I can’t hear it at with my ear. I can hear it freaking out on a scope at high rpm as well free reving.

-I will include pictures from teardown in a link once on computer

-Leakdown. I have a tester that reads not leakdown percentage but two individual pressures. I used 80 psi (should have used 100 but it’s fine). All 4 cylinders came back with 77-78 psi in the combustion chamber. So I did some math and came up with the 2.5% figure. Despite not using 100 psi, and despite my math skills leakdown is still definitely consistent across all 4 and healthy. Leakdown was done HOT off the press.

I will do an intro post to my background later as well. Thanks in advance!

 

Sounds like piston slap. Sounds exactly like the piston slap on my built engine.
But you say it doesn't go away when hot. Mine does...

 

@Erat It does go away when hot to the ear, free reving. But I can also hear it with a scope free reving hot. I can hear it with my ear faintly under load in the cabin at the same rpm range driving around. So can my knock sensors lol

 

Are you on megasquirt? You can upload your tune and datalog right to the forum as well.
Just add as an attachment. Be sure to add context.

I know on my MS3x there are a lot of settings for knock sensing.

 

@Erat Yes on MS3. See tune and datalog from a drive attached. For context, this tune was ran on a rods-only BP4W for 10+ events/full season of pure hammering. Ran GREAT. You can see the knock sensor on cyl 2/3 picking up the noise. You can see how the noise actually extends into the higher rpm range as well. Just can't hear it without a scope. Disclaimer, Knock control has not been tuned yet. Just baseline and go atm. Diagnosing this issue has taken over, so I haven't been able to actually experiment and manipulate input sensitivity yet. I have the G4 knock light as well picking it up. Do piston slap prone guys see this much knock sensor activity on a fully warmed up motor? I can't imagine they do...but maybe they do!

 

Looks nice!

I purchased this
MSM BEGi Parts Tub [SOLD!] - Miata Turbo Forum - Boost cars, acquire cats.
The engine is very similar to yours, It was completely gummed up from improper tuning and is currently in many, many small pieces now.
I am a machinist, I have the tools to measure stuff myself. I found enough inconsistencies in my Super Tech 9.5-1 pistons that I am not going to re-use them.
I also believe completely in a quench for this motor of .040 and am machining accordingly. My bore to piston clearance (right now) is .0035.
Because the engine had been recently built (also by a decent machine shop) I did a "blue-print" style tear down, check everything...
The only thing wrong was serious variation in the deck height and as built (.065 quench) it didn't matter.
BEGi tuned MSM computer with an Exide piggy back were the root cause of the failure.
Because I had vastly different deck heights I swapped pistons, rods, and holes to determine where the variations were.
The tops of the pistons were the culprit and having +/- .005 at .040 quench is outside my comfort zone.

I may use these as I prefer flat tops.
JE, Mazda, 3.287 in. Bore, Kit (jepistons.com)

Now, I agree, your stuff does not look damaged.
The audio sounds like piston slap- bad piston slap like one might hear on a loose build started in sub -freezing weather...

If you were together I'd say short each cylinder while running to see if you can isolate to a single cylinder.

It looks like you're apart right now so I'd be measuring everything myself.
I've found enough issues with my Super-Techs that I'm planning on replacing them.
I did not compare the barrel shape (not round, it's oval) of mine between pistons but I'd be checking that closely on yours...
I'd take measurements not only in the normal spot but at 120 degree and 90 to it and at the top of the piston as well.
There are no published figures for what you will be measuring, what you're looking for is consistency (or lack thereof) between each piston and the others in the set.

I say pistons are your culprit but I'm pissed at mine right now so I'm not objective...

Something strange I noticed.
The minor bore scratches evident go all the way to the top, Normally this scratch stops where the top ring stops (.250 down the bore).
The one picture of the side of the piston does not show evidence of this...
Did you run it without an airfilter?
The scratches are not uniform either. I'd check the bores for taper and out of round.
Also weird mark in #2?, I assume it is a camera induced flaw but if you really have a hickey that size in the bore I'd be unhappy with my machinist...

 

I have not had time to look at your tune, so just a shot in the dark. Do you have any cold advance working on that car? I find too much cold advance will give me some extra slap on warm up. Also since you are picking this up on your knock sensors, have you considered that you have too much timing in the free revving and low RPM areas of the map?
Just a thought, but my car used to slap pretty bad in the morning until I turned cold advance off. Made absolutely no difference in drivability when warming up.

 

The nick in #2 kinda looks like it could just be a drop of liquid to me, photos...

 

Has anyone else looked at the tune? 37* at 100kpa at red line, doesn't drop below 30* until 6psi, doesn't drop below 20* until 13psi. Doesn't target richer than 12 until 10psi. Is this E85? Am I crazy?

edit: also constantly running on 5% WUE.

edit2: don't get me wrong, knock readings in the log do say you've got a mechanical issue going on, but yikes.

 

Ok, game face on.

Looking at the pictures, I see two different patterns on the head. One looks like it's from the head gasket seal surface, one is much closer to the diameter of the pistons. Are you sure it's not just touching? Techninja and I are slightly different in our approach, if it were me, I'd throw a .060" Cometic gasket in there slap that bitch on the dyno.

 

@technicalninja I do agree with your assumption of the noise being an odd presentation of piston slap. That was my intuitive guess when I first heard the noise for the first time. Because it either has to be bad machine work, bad assembly, or a part defect. And I trust the machine work and I know it’s not assembly error. So part defect seems to be the most logical variable remaining. I know Supertech has had quality control issues with their line of valves before so anything is possible I suppose. I have considered dropping the short block back off with the performance shop and having them inspect and run the numbers for some kind of confirmation on something.

@sonofthehill it is just a bit of amsoil on the cylinder wall. No nicks. Everything clean. I have not touched cold advance. It’s definitely a mechanical issue of some kind and isn’t affected by timing. Even if fixed.

@curly I understand everybody will have their thoughts on tunes. I don’t want to talk too much about the tune on this particular post unless you think something is relevant with it to the presentation of the noise. I worked with who I believe is a great tuner and don’t know enough to speak to or justify why the spark table is the way it is. All I can say is that the car made great power and performed knock free and drove great all last race season on this tune.

 

So I actually lined up the cometic gasket with that diameter you are referring to curly and I believe it’s just the diameter of the cometic from sealing which is slightly larger diameter than the pistons. Also, I would think that if the piston(s) were somehow touching the head, the presentation of the noise would actually get worse as things heat up and materials expand and the quench area reduces.

 

Curly is one of the "serious" folks I was referring too.
Excess timing is a killer of BPs and they seem to be more detonation prone than other platforms.
Curly is far far more experienced than me regarding tuning and his methods (the way he tunes) are spot on IMO.

If he wasn't 2k miles away from me he would be the tuner I would use.
I've already suggested the he move to Texas but he's a PNW guy and likes it there.

If Curly looked at my tune and said "Yikes" I'd change it BEFORE re-start.

Curly also has direct experience with getting quench too close, he trashed an engine once and is gun-shy about running them tight.

If you are running heavy fuel, E/85, real race gas, methanol then quench doesn't matter nearly as much and running larger quench clearances makes sense.
On an engine that is not detonation prone removing/reshaping the quench pad can sometimes reduce valve shrouding and make more power but these mods are at "the razors edge" and are only applicable on an "all out" build.

I've never encountered detonation free revving, I'm not sure it's even possible. Your piston tops and the bottom of the head show no evidence of either contact or detonation so...
Your noise is mechanical IMO. A slightly bent rod, a piston that is slightly misshapen, a tapered/out of round bore, are what I'd be looking for.

On my pistons it was not the machined circle that was the problem, it was the area just inside the circle that was different. You could actually feel a difference in the tiny little step between each piston and one (#4 for me) you could see/feel a difference between each side of the piston.
This piston had the greatest divergence from the others and it had .008 difference between each side. I checked how far the quench pads in the head extended past the outer ring on the piston. 5mm on exhaust and 8mm on intake so these pistons REQUIRE a greater quench distance than something with tighter tolerances.
I actually prefer the material they are made out of for street car use. The alloy requires less clearance as it expands less as it heats up and the tighter bore clearances should REDUCE the tendency for piston slap.

It sounds like you have bad piston slap, on a fresh set of Super Techs installed with normal clearances that really should not make noise even when cold.

I'm interested in what this turns out to be.
My "money" is on pistons...

PS Take Curly's advice, just compare your tune to others on this forum and really look at your spark advance tables, You have had TWO heavies (SOH, Curly) both say the same thing. They're not wrong.

Personally, I 'd just beg Curly to help me...

 

@technicalninja I am a long time listener and first time caller, I know of @curly and a bunch of the other vets here and know how awesome everyone here is for sharing their knowledge and taking time to help folks for years on end. I am definitely interested in networking and learning from more folks, I just want to kind of earn a place before I just start asking people for all sorts of help and to share their knowledge they’ve spent years accumulating, especially without compensation. I’m a big believer in compensation. Thanks again for all the advice everyone. Hopefully on the right track.

edit: I only have HB mics and no dial bore gauge. So if I am gonna search for a defect via measuring I will probably need to take it back to the shop which they are cool with. I also have an entire second set of new 9:5:1 Supertech sitting in a box in the garage so I could certainly rule out a piston defect. I just need the next time this motor goes in to be the final time. I’m getting tired of cleaning and re sealing with RTV. And cometic gaskets are not cheap. LOL

 

Just a stupid question; the marks on the top of the piston in the picture above seems to indicate the that valves are coming in contact with it. Is this correct?

 

@olderguy I think what you are seeing is a reflection of light on a layer of fluid/oil sitting in the valve relief. Haven’t cleaned Pistons yet. Piston tops look good. No piston to valve contact.

 

If those pistons have not been cleaned yet just how long did you run it?
They are very clean for actually seeing fire.

Purchase a bore indicator. OK ones are not super expensive
This is what I use
6460 – Cylinder Bore Gage (centraltools.com)

This used one is nicer than mine
Central Tools 6462 Cylinder Bore Gauge Periscope Style 2.6 to 7 In | eBay

And this will suffice

I prefer the type with a sled, you can really see OOR and taper with it.
You zero the indicator in the bore and then use your mics to get a reading.
Then you use the same mics to measure the piston. Even the cheapest mics ARE accurate between 2 readings on the same mic so if used this way the inexpensive mics are as accurate as a Starrett or Brown&Sharpe mic.
I have the expensive mics but I use my "cheapos" more often.
If you are serious about this (and you seem to be) the tool you buy today you will be using 3 decades from now.
My bore gauge is now 35 years old...

 

@technicalninja a decent dial bore gauge is on the list, I’ve just already spent so much $ in such a short amount of time on this adventure that I can barely afford. I’ll pony up soon. Also I know there is much practice that goes into properly measuring and getting accurate repeatable measurements and I have much to learn. I feel like this is going to take a vets eye’s and some very very thorough accurate measuring to determine a defect in the piston, bore or rod.

The motor was ran for about 30 miles. That second picture is the condition of the piston tops right after removal of the head during teardown.