VVT Bottom-End Rebuild / Refresh
 

Hey guys - first 'serious' post in the engine category on the forum - I typically post over in the 'build-threads' section (link in sig) – so be kind ;)

This is also my first crack at rebuilding / upgrade engine internals. I've done most other things, but never internals.

As such, there has been / will be LOTS of continuing research as I go along – input welcome


The Symptoms:

-LONG cranking starts
-Horrible exhaust smell
-Horrible gas mileage (5-7 mpg average)
-VERY lackluster power (close to a stock Miata)

I’ve determined that I have been washing the cylinder walls with fuel - my dip stick smells more strongly of fuel than my gas cap does! Haven't completed a leak down / compression test yet – (forthcoming) - but this seems to be a fairly straight forward - I'm at peace with it.


Car Parts / Intended use:

Car daily runs 60-70 miles @ 7600ft elevation (primarily commutes, not track) (Installed parts are in my sig)


The Project:

-Leave VVT head as-is (remove / clean)
-Re-use previously-installed forged rods (assuming no damage)
-Re-use stock cast pistons/ wrist pins (ditto)
-Re-use stock crank (ditto)
-Verify / resurface head/block if required (never overheated – but new gaskets are picky?)
-Send block out for bath and measurements
-Minimum re-hone cylinders - possibly over-bore slightly if req’d (0.5mm / 1mm)
-Install new ARP Main studs (I know I have head studs, not sure of others)
-Install new ARP2000 rod bolts
-Install new ACL race bearings (Main / Rod / Thrust)
-Install new rings (NRG, CP, JE, Wiseco, Mahle, Manley) – (matched to larger bore if required)
-Install full OEM seal kit / Timing Belt / WP / Clutch master & slave / motor mounts, etc
-Billet (race) Oil Pump (Boundary Engineering) – (skip the crank damper)
-Other?


THEN:

-Break in the engine using stock hardware only, disconnect Link & extra injectors


THEN pt. 2:

-Replace modified intake manifold (4 additional injectors) with stock NB2 intake manifold
-Install 650cc injectors (ID 1000cc instead? – No ethanol)
-Install Megasquirt (considering MS2)
-Verify Walbro 190 fuel pump, if not - upgrade
-Tune / get it running!


Questions:

-What else have I not considered / included in my 'To-Do' list?

-Any reason to tear into the intake manifold due to damage (not looking to upgrade)

-Compression. I’d prefer to run my OEM cast pistons – but there’s a possibility of required over-boring the cylinders due to scoring / damage… so I’m assuming my stockers won’t work if over-bored, correct? (In that case, planning to go with the 8:6:1 units)



Work begins soon. Parts are already in the cart – ready to be shipped.

i would first do a compression check.

change oil to flush out the gasoline soaked oil.

Then figure out why you are running pig rich.

your engine might be fine, and you are just drowning it in gas.

In some ways, this thread / plan of action may be a bit early - there ARE obviously some basics to get out of the way first. Agreed - compression/leak down tests are planned, and an oil change will be required, regardless - I'm hoping that fresh oil *might* help re-set my situation.


Would 'cylinder washing' actually cause this big of a power loss and such poor MPG's? I fear the rings are toast for the symptoms to have gotten this bad.

This is where I've been puzzled. On one hand I have good AFRs, but horrible mileage and power (suggesting over-fueling and cylinder wall washing) - so I assume that if the oil change helps me 'rest', my AFRs will be INSANE and require a complete reset, tuning table wise.


Fingers crossed. Oil change tonight, comp check when the tool gets in

you can borrow a compression checker at any autoparts store for free.

was this same tune running reliably before? when did this crazy richness start?

Agreed about the tester - got one in the mail, here tonight or tomorrow.

Honestly, I don't have a lot of history with the car beforehand. The PO did have a very good history with the parts/tune (he is a very trustworthy source). Car has been on the road... idk, 10hrs since all this work - I did have to bump the fuel maps up when getting the car running (this old Link stuff is kinda annoying / vague).

After the oil change tonight I'll do a full tune-reset (I have FM's recommended values) and go from there.

No wideband?

Wideband installed and functioning well.

I *thought* things were doing well (while they weren't) - I did bump up fueling at one point (can't remember exact order of things) to help with idle / AFR values.

I was idling between 14.5/15.0, 11.0ish at full boost (12lbs), 16.0ish at cruise (Link doesn't kick in the extra injectors w/o boost). But at the same time I've been getting horrible gas mileage, horrible exhaust smell and massively down on power (can't maintain 75mph in 6th, have to downshift to 5th @ 4200rpm just to hold 75!)

Like I mentioned above - after doing an oil change tonight I'll also do a full Link reset - returning to FM recommended values. Megasquirt is in my future, regardless.

You have to go with a MS3 to work with VVT. I would not fuck around with a MS2 and then have to install a VVT Tuner.

Thank you for the input.

I had considered the MS2 for cost-sake - I do not want to have to install a VVT tuner.

MS3 it is

May not be a bad idea to post your timing table... you mention it's running rich, but at 70mph cruise you should still be near stoich. A bad tune would definitely cause lack of power and terrible gas mileage.
On the other hand, I know nothing about how a link works, and dual injectors seems utterly ridiculous for a GT2560 :)

Edit: I just looked at the Link manual. You seriously tune this thing with a 1x48 character display and a couple buttons? O.O Burn it with fire, get a megasquirt.

Good god, 5mpg? Are you experiencing misfires? Curious if you are fouling plugs? I'd wanna double check that wideband sensor somehow if you're using that much fuel and still getting normal AFR readings. Did this creep in or was there a sudden change at some point?

Yes, exactly... this thing is... mystifying AND a bit frustrating. I have the recommended FM values and will reset the ECU to those with an oil change this evening. Kinda a mess - would like to get away from the Link and into Megasquirt asap - but that also brings a steep learning curve! It'll be worth it.


Agreed, I've been averaging between 5-7mpg. I've got a 30 mile trip (one way) to work - in that distance I clear roughly half a tank (premium) - not great.

I haven't had any miss-fires (that I have noticed / heard) - a quick check of the Engine Light (will be on forever - check my Kart build) will verify. I understand the CEL usually 'blinks' when a miss-fire is detected? The wideband sensor is brand new, and controller is a nice unit (Autometer Wideband Pro, 5778 Phantom Series). When installing I did look through the manual trying to find anything about open air calibration - didn't see anything noted.

Popping the plugs for a quick look while completing my oil change and checking timing tonight will tell if I'm fouling the plugs! Will update.


The issue has been ever-present (since start-up) - this turbo setup was in a different car that ran dependably before I swapped roller-skates. (Here: https://www.miataturbo.net/build-thr...3/)

I ran the setup on stock settings, but was getting AFR readings near 15.0/16.0 (16.0 is the limit of this wideband?) on boost - so I adjusted up the 'master fuel adjustment' setting off '10' (FM recommended) to around 20, which currently gives me 'good' AFRs. This Link has me pulling my hair out.

Update:

Good News: Completed an oil change last night - have a magnetic plug, didn't notice any metal shavings

Bad News: Ran a compression check after the oil change - on the second try (with teaspoon of oil dropped in) these were my numbers:

1. 105
2. 143
3. 113
3. 113

When I pulled the plugs, these are what the looked like (1,2,3,4):

Attachment 227998
https://cimg4.ibsrv.net/gimg/www.mia...73ab891d03.jpg

So - next steps:

Engine removal / tear down - block inspection - then determine whether or not I can simply do a honing / ring job OR see if I have to do a slight over-bore w/ new appropriately sized pistons/rings.

Oil pump, bearings, head studs, seal kit (etc) are all assumed.

Looks like 1 & 4 weren’t even firing. Maybe check that coil. Did your used oil reek of gasoline?

1 & 4 do indeed look VERY clean - kinda odd - these plugs have... maybe 500-750 miles on them.

Yup - as mentioned above, I was concerned that the engine was getting cylinder-washing - the oil and dipstick did smell like gas. Compression check was with new oil and a warmed up engine.

I was considering checking timing, but at this point that doesn't matter - those compression results are pretty clear.

isnt your engine shaking like mad with only 2 cylinders firing?

im surprised you even drove the car in this condition.

you can check spark on cyl 1&4 with a timing light clamped to the ignition wire for cyl 1&4

did you recently do a timing belt change? could timing on the belt be way off?

I think it would be good to know if you're not getting spark, and have that fixed before the engine goes back in.

No timing belt change. Car doesn't have the many miles on it since the 're-assembly' - and honestly it didn't shake much, it wasn't silky smooth, but wasn't bad. I've got a light and will give spark a look.


Agreed.

Yea based on those plugs, the atrocious mpg, and fuel smell(unburned fuel in cyl 1/4 just leaks past the rings) in the oil, I'd say its a safe bet that the coil bad on 1/4. Since it is completely not firing on either it might not make a huge vibration as the firing pattern may not generate any abnormal as the two firing cylinder are 180 apart vs say it running on 3 cyl with one being 90 out but sure you would have noticed a SERIOUS lack of power when driving with only having 2 cylinder

As for the compression numbers I assume you ran the test at WOT otherwise the 3 dont look bad it it was closed throttle but Id want to know the before and after the oil in cyl to make a judgement.

Alrighty boys - an update:

Seems as the timing in the engine is off minimum one tooth (almost looks like 1-1/2 tooth?)

Regardless - I'm going to reset base timing tonight and re-check compression.

Will update soon. Thanks for everyone's input.

https://cimg1.ibsrv.net/gimg/www.mia...52cbecfbd8.jpg
https://cimg7.ibsrv.net/gimg/www.mia...f6daf32b90.jpg


Attachment 227996

Attachment 227997

Exhaust cam looks ok, intake looks off by a tooth.

VVT will alter cam-to-pulley orientation but not pulley-to-belt or pulley-to-pulley.

Agreed.

Dissagree. In the 1st and 3rd pictures the intake cam looks spot on and the exhaust cam looks off

Well, regardless of what exactly was off - I dug into the engine and reset the static timing (crank, cams, tension, etc) - I put everything back together, warmed up the engine and re-checked compression.

First check gave me 50psi - this was because I had forgotten to wire the intake WOT.

After fixing that, I tried again - I got +/-100psi, straight across the board. :mad:

This was pretty frustrating, and I was doubting my timing reset - so I removed valve cover / timing belt cover and re-verified static timing. Everything looks spot-on.

Camshafts/Lifters/Springs had good movement, no damage / defects, everything looked good (hand turning the engine).

So, at this point it looks like a full removal and tear-down is in order.

https://cimg0.ibsrv.net/gimg/www.mia...bad3f80738.jpg


https://cimg2.ibsrv.net/gimg/www.mia...0ab1ab57af.jpg


https://cimg8.ibsrv.net/gimg/www.mia...9346dabae5.jpg

You said a leak down test was planned. Where are the results?


I've used 3 different compression testers on the same engine and gotten readings that varied by 50psi.

This company makes the compression tester, and other gauge sets for snap on. If you can afford it this set is worth it. Their leak down tester is $85 on amazon, worth the premium over harbor freight for peace of mind.
SKU : TU-30A - Compression Tester Set - Lang Tools

Nothing yet on leak-down info - borrowing a tester today from NAPA, will use tomorrow. I'm hoping I ran into a bum tester. If I don't get good results with the NAPA unit tomorrow I'll just pull the whole things and start a tear-down - just to get my eyes on things, then go from there.


Thank you for the tip - looks legit!

Decided to check ignition with a timing light - my light I reading spark from cylinder 2/3 but not 1/4.

The car does start and run, which pretty much rules out ECU issues.

I double checked grounds, as well as 12+ to the 1/4 coil pack - I do have switched power.

Then I ran the car again, noticed that the coil pack for 1/4 was getting EXTREMELY hot - after running 20/30 seconds?

So I swapped coils - same problem - then I swapped in a back up set of coils - same issue - whatever coil is plugged into the 1/4 harness is getting SUPER hot, and I'm still not getting spark at cylinders 1/4.

Thoughts? (Found out no local autoparts store loans compression or leak down testers - so that's delayed - spark is my foremost concern right now)

I would replace plugs and wires, just to rule out.

Plugs / Wires are new, less than 300 miles on them.



Alrighty. Didn't make much more progress over the weekend from my previous update.

Some Info:
The ONLY wiring on the car that isn't 'stock' is the old Flyin Miata Link ECU.

The Link ECU interrupts the signal from the stock ecu to the spark plugs - (2) stock wires and the 12+.

Seen in this install image:

https://cimg6.ibsrv.net/gimg/www.mia...bce97d7279.jpg

Attachment 227995


SO, I've verified that the 12+ wire is installed correctly, supplying voltage only with the key in the 'run' position. I then double-checked the connections between the Link wires and the stock ECU wires (Brown/Yellow, Black/Yellow) - everything looked good.

I don't believe it's a wiring issue - I think it make be an ECU / trigger issue, possibly turning the injector on ALL the time, therefore over-driving it and overheating the coil. Need to check those supply wires.

Sounds like your Link is fried. I say ditch it asap, there is a reason they are so reviled around here.

100%

Actively looking for an MS3 - suppose the smart route would be Squirt the car, get it running and happy, then look at engine internals.

I've posted on other forums, consensus seems to be the Link - need to verify that the coil trigger wire is seeing constant voltage - that would suggest bad signals from the Link.

Hate to say but with those compression numbers (repeatedly)-- its probably not gonna be "happy" until it gets some new rings and a hone.

Welp, some more progress:

Decided to disconnect the Link ECU entirely. Did so, checked spark with my timing light - still no spark in 1/4 cylinders.

So, I'm poking around now trying to determine if anything out of place - hoping it's not the stock Miata ECU.

After getting the Link uninstalled I took some measurements at the coil pigtail connectors. I did the measurements with the ECU connected / disconnected, and with key on / key off, here's what I got:

ECU connected, Key On
Inj 1
BR/W, 1.86 - B, 4.1 - B/W, 12.24
Inj 2
B/Y, 32.8 - B, 4.0 - B/W, 12.16

ECU connected, Key Off
Inj 1
BR/W, 2.2 - B, 0.0 - B/W, 0.0
Inj 2
B/Y, 2.2 - B, 0.0 - B/W, 0.0


ECU disconnected, Key On
Inj 1
BR/W, 0.0 - B, 0.0 - B/W, 12.3
Inj 2
B/Y, 0.0 - B, 0.0 - B/W, 12.3

ECU dissconnected, Key Off
Inj 1
BR/W, 0.0 - B, 0.0 - B/W, 0.0
Inj 2
B/Y, 0.0 - B, 0.0 - B/W, 0.0


One observation:
When checking Inj 1 (runs coils 2/3) when taking a reading on the BR/W wire with ECU Connected, KEY ON, I got the 1.86 reading (was 32.8 on B/Y wire on Inj 2), AND I got the red "X" lit up on the Battery Load Test function of my multimeter (not illuminated for B/Y wire on Inj 2 which runs 1/4, the "problem" coil)

To add, the Green "check mark" lit up on the Battery Load Test function of my multimeter when checking the B/W wires for BOTH injectors, ECU connected AND dissconnected, KEY ON

Thoughts?

Well, an update:

Double checked - Coil 1 (cylinders 1/4) still gets hot to the touch after 10-20 seconds of operation.

Also double-checked the voltage at each coil pig tail - these are the readings I got (ECU connected, Key turned to "run"):

Coil 1 (Cylinders 1/4)(No spark)
Br/W - 1.6v, B - 0.0v, B/W - 12.0v

Coil 2 (Cylinders 2/3) (Has spark)
B/Y - 25mv, B - 0.0v,. B/W - 12.0v

The only difference seems to be the Brown/White wire and the Black/Yellow wire. I believe this difference is what's causing no spark and overheating in Coil 1 - thoughts?

Didn't read word for word: let me summarize my thoughts.

1) Timing lights aren't how you check for spark. Use a test light connected to ground, or an inline spark tester.

2) Confirm good 12volt input at each coil, it should be constant with key on, engine off. Use a test light, or load the circuit in someway. Open voltage with a meter tells you little; although this is doesn't sound like your issue, since they have a shared power feed. Unless you have an issue between the splice and coil connectors.

3) Swap coils between 1/4 and 2/3. If your spark moves, you have a coil failure. Why? Well either random failure, or environmental. I've had defective plugs out of the box; wires too. No output for 10k volts and higher means dead coil. I would be performing all spark/no-spark testing with the COP unit bolted down, testing from the output of the coil body where the wire connects. Of course, you could have a single dead pole on one coil. Again, doubtful. Swapping the coils eliminates this possibility.

4) ECU switches the ground to control primary side of the coil. If your readings are different at the control wires, you have 2 options.




A) You have an open or voltage drop between the connectors at the PCM/Link/Coils. 12 volt input supply can be load tested at the connectors for the coils. However the other two wires need checked between the coil, and the ultimate source of control. Disconnect all players, Input 12 volts through each relevant wire, ground a test light at the output. This confirms ability to move sufficient current to operate the coil. ALL inputs from PCM to coil connector, Coil connector to link, and link back to PCM need checked. If you want to be lazy, remove the Link piggyback (eliminating it), and check from PCM connector to coil connector

B) You have a failed coil driver (my suspect), which is not fully pulling the coil to ground. I assume pins 3I and 3F are the ECU coil driver grounds on your diagram? If this is the case, I would be replacing that ignition coil before replacing the ECU.

Well, some good progress:

Grabbed a complete ECU/Immobilizer Box/Ignition Ring/Keys from a friends ride, swapped them in, started right up. Verified spark at all (4) plugs, everything looked / sounded good! Took her out the next day for some pulls - ran just like she used to, tons of power, great sound.

After about an hour or two (or three?) of gunning it through canyons, the car just stalled and stopped, temp spiked. She wouldn't turn over quickly enough to fire the engine, requiring a tow. After getting her home, I started going through the whole 'air, fuel, spark' tests.

Pulled spark plug #1, smelled faintly of gas and spark checked out.
Pulled spark plug #2, didn't smell much gas. Went to verify spark and got a mini 'Old-Faithful' out of cylinder #2. (Same for cylinder #3 btw).

Attachment 227993

Attachment 227994

The liquid was watery slippery. Smelled slightly of fuel, but not much. Had a green tint to it, obviously coolant.

SO

Pretty sure the head gasket blew. Maybe I should have seen it coming with compression numbers being so low? IDK. I don't think it's a cracked block/head (fingers crossed), and this kind of failure wouldn't be from any of the other coolant / seals. OEM Miata gasket kit is in the mail, I'll update when I get into things farther.

Change the coil that was an intermittent no-spark. If I'm correct on my hunch the ECU driver was at fault, that coil may have been the cause. Coils are relatively cheap, and if there is a coil issue which killed the previous ECU driver, the new ECU (and its driver) will also fail.


My preferred head-gasket test is using a leak down tester, if you have access to compressed air. If there was significant coolant in the combustion chamber, I would be concerned of a bent connecting rod from hydrolock. IF the pistons can be cleaned enough to be totally carbon free on the crown, a machinist straight edge and a feeler gauge should reveal any significant rod length changes.

I agree - only change I made to get the car up and running (really well, actually) was the ECU. I'll swap the coils just in case, didn't know that a bad coil could back-feed and damage the ECU? Good to know!

I'm somewhat worried about this (hydrolock), there was a *decent* (maybe a cup, cup and half?) amount of coolant in there - rods *may* be on my list of purchases, just to be safe. Your tip about cleaning the crowns and measuring length is noted - thanks!

Potential for computer damage, via coil failure, is really just a product of how an ignition coil functions. Long story short, a short or partial short in the primary winding is what kills the ECU drivers. Total current flow of a straight piece of wire with zero effective resistance vs. the windings of the coil primary side is significant. If there is a short within the primary, current levels can be higher than expected, damaging the driver. Manufacturers and their engineers have finally gotten wise to this, I went to a FCA training event a few weeks ago and basically every driver on their new ECUs have current limiters which will disable the drivers if excess current flow is detected. I seriously doubt the miata ECU has current limiting drivers, but you never know. IMO if there is a concern of a damaged driver in the ECU, the driven component must be thoroughly inspected. Unless you have a lap-scope, there really isn't a good way to do this. Really I would only worry about the coil which was not being switched properly. If the other one worked fine, I wouldn't say it's required to be changed. Of course, coils aren't crazy expensive either.

This is the enjoyable aspect of my day-job, so I enjoy nerding out about it. :party:



Good luck with your exploratory surgery on the motor.

Slight Update:

Pulled the engine/trans. Disassembled the engine down to the bare block (notice the nasty, brownish looking oil - MAJOR coolant dilution):

Attachment 227991

Looked over the whole thing in detail - some minor scoring on the cylinders, pistons/crank/bearings looked fine. No rods bent (a concern was hydro-lock from the coolant in the combustion chambers.) No cracks (visible or felt with fingernails - very 'high tech' analysis.) No worn-in ridge/ring at the top of the cylinders either.

In the process discovered this - beside the oil squirter for Cylinder #4:

Attachment 227992

Some of you know what this is - caught me MAJORLY off guard - those ragged edges looked so bad - I spent the night convinced I had grenaded my block. Turns out NOT to be a hole / casting defect, rather the oil return/drain passage for the head. PHEW.



Haven't gotten the any parts to a machinist for measurement - nothing within 3hrs of my location - will do this soon. The overall condition of the block/bore/crank/bearings seem promising - possibly just need a good cleaning, honing, rings/bearings/seal job?

I did take a look at my valve lash (thanks Greg Peters) - nearly HALF of my lifters don't fully close due to poor adjustment. It's possible this is where my compression test issues (look back a half-dozen posts) seem to have originated. Still doesn't explain my primary engine issues.