Alright I re-checked compression with all the spark plugs pulled this time. Much more consistent #'s now.

205
205
205
205

How many times did you crank?

once it stopped, i stopped cranking.

it was only a few seconds

Just checked mine

2007 w/ 88600mi. 190 +/-5 across all four cylinders

Sounds healthy and ready for more modifications :)

Nice numbers. I don't even want to check mine :/

Sent from my DROID4 using Tapatalk 2

Checked mine on Sunday, car has 45k and is fully bolted + tuned, the results were:

(passenger)
189
187
185
180

0-190-190-195


TappaFak

LOL Ouch.

Woops

Probably sent from a toilet

Performed a compression test. 84k miles. Intake, internals, and freek tune in progress.

185, 185, 170, 185.

@Lex;

I have seen many DISI engines with exactly 150psi on cylinder 3 at work. (I compression test every engine I do timing chain/VVT work on, or remove the cylinder head, ect)

The engines have all run fine and make no noises.

As you can see, MY engine is developing low(er) compression on #3 .

I have searched my ass off for 45 minutes reading hundreds of posts on this forum and it would seem ring land failures are possible.

Since creating this thread (which I have not read completely through, out of time) has there been ANY progress on discovering what the hell is up with cylinder 3?
It's gotten to the point that I EXPECT 150 psi when I'm doing these compression tests now.

Thank you sir.

Cylinder three has the highest flow through the highly imbalanced intake mani and since we don't have individual o2 sensors in each exhaust port, fuel delivery is averaged over all 4 cylinders, and 3 runs lean.

he speaks the truth.

I would be willing to bet any that you test out in the 150 range, compared to others in the 180s has a broken ringland...or bent rod but you can check for piston height while doing the compression checks to confirm a severely bent rod with just a long screwdriver...slightly bent rod may require a dial gauge.

that is exactly what my compression numbers looked like and I have torn my motor down to confirm a broken ring land in #3 .

Since I am here I'll report my new numbers.

150 miles on the new motor with it fully warmed up

182 - 182 - 182 - 182

Glad to see you back up and running Dano !

Consistent compression is consistent! lol

Did this a few months back at about 65k miles in a 2007 ms3 mods are in sig.

Dry

1-171
2-155
3-160
4-175






Wet

1-180
2-160
3-165
4-181

I see. Could you link to some hard evidence of this? Yes, I searched but came up empty.
I've never noted any signs of lean running on spark plugs or pistons. Or even exhaust valves on the engines I removed the head from. It would make sense though. I have a spare exhaust manifold, and an Innovate wideband. It would be very interesting to install a bung on each runner and actually get A/F measurements per cylinder. Has that been done yet as far as you know?
I checked mine with dial calipers, no piston height discrepancies.
The ring land issue. It seems so likely.
150psi on 3rd.. So common it's ridiculous.

When the intake manifold was flow bench baselined it's was 24% imbalanced across all 4. Lol. 3 being the highest flowing.
From my phone it's going to be hard to get you a link.
Seems the heat of running lean leads to piston rings butting together at the gap. Something has to give.
http://www.mazdaspeedforums.org/foru...-thread-88365/
First post has the numbers.

Any word of this compression problem on #3 for the genpu's? I've been away since i sold my 07 MS3 a couple of years ago but interested in joining again by the summer.

yes it damn well better be with all the work I did... esp in #3 cylinder to get it cleaned up proper!

time will tell how it holds up.

not only that but if my #3 cylinder wall is any indication of the larger data sample, they get pretty warped due to all the heat.

PWC bore measurements can be found in my thread.

http://www.mazdaspeedforums.org/foru...09/index5.html

Thanks for the link. I might have missed it but did they flow test an unmodified intake?
I see results there for a ported stock manifold, but the results show that cylinder two actually flows the most, followed by 3.

Since the O2 sensor get an average of the four because it's reading all four exhaust streams, I look at it like this.
The average CFM per intake runner per that test, is 222.75 CFM.
Cylinder 3's result of 233 is only 4.4% higher than average. You could *theorize* that results in an actual AFR only 4.4% leaner.
Even still, cylinder 2 would be leaner yet.
All of this means squat since it was one test on one machine of a non-stock manifold since it was ported.

I'm not sure how conclusive that is. The only 100% for sure test would be individual bungs in each exhaust runner with a standalone wideband to check actual AFR.
If my car wasn't my daily I'd love to do this test with my extra exhaust manifold.

Are there any other tests done to verify the claim of lean on cylinder 3?
Not saying it isn't correct, it's just that one test on a ported manifold doesn't hold much water.

go to maperformance.com...lookup their intake for the MS3/6 and you will find all the flow data you seek.

Will do thank you Dano.

Edit: Given that data, cylinder 3 is approximately 10% above the combined average for flow, and you could *theorize* once again that could equal #3 being approximately 10% leaner.

That, is significant. Thanks again for the information and links.
And I've always wondered why Mazda tuned this engine to be a soot-blowing dragon.

I still really would like to see an individual exhaust runner AFR test and may just do it myself. That is a project for another day however. ha

I also believe the ring gap closing is a possibility but what can we do but speculate?

2009 speed3
77k km

1- 200 psi
2- 200 psi
3- 200 psi
4- 200 psi

Broken gauge.

That's what I thought when I kept getting 205 and that was on a brand new snap on gauge.

I went over to my friends shop and used another snap on gauge, same #. I even tried a MAC gauge, same #'s.

I'm at 3 different gauges now showing the same lol

Weird that there would be that many broken gauges.

build a motor with OEM pistons and open up the ring gaps.

there is little doubt in my case the top ring butted causing the land failure.

hahahahahahah


never seen a 200 before...ive hit 195 I believe in the heat of summer.

edit: strike that I have hit 190 before..

#1 195
#2 195
#3 196
#4 195

Built engine with about 13k miles on it i gapped rings to 23 thou

my wiseco pistons and rings are Boss

stock compression ratio 3 layer cometic HG

my wisecos are gaped as follows

1st = .018
2nd = .022

ive just accepted it as actual.

only so many mechanic friends i can bug to let me borrow compression gauges from lol

i figured to be better safe then sorry and allow them more room to expand especially wanting to go over 500 HP in future, and with my high ass compression i dont regret it my gap is perfect.

believe it or not they dont even really leak down they hold the pressure.

Just curious, where did you get that info? Three thousandths would be out of spec.

This is out of the overhaul manual.
"Piston ring end gap
Standard
Top: 0.14—0.24 mm {0.0056—0.0094 in}
Second: 0.34—0.44 mm {0.0134—0.0173 in}
Oil (rail): 0.15—0.40 mm {0.0060—0.0157 in}
Maximum
1.0 mm {0.0393 in}"

Just trying to clear any confusion. Also, I do agree the low end of the spec is pretty tight and any engines that are AT the low end could have issues like you had.

I was incorrect you are correct.

Well if your gaps work for you, that's where I'm setting mine when the time comes, that's for sure.

for clarification...I am not running OEM pistons/rings.

If I were to run OEM pistons/rings I would run larger than spec...not sure exactly where I would land.

I got my specs from the wiseco instructions they show ring gaps for various power setups i.e natural aspirated, turbo, turbo with nos, nos the higher the expected power the bigger the gap

All of @Dano; s fail (my twin '08 red) and the talk of cracking ringlands from track use made me nervous enough to finally whip out the compression gauge.

'08 Speed 3, 48k miles, approx 20hr of track use, fully warmed up motor (oil temps ~180*F)

195-188-188-190 :bigok:

As for track use, I have been careful to shut it down when oil temps got above 240*F and have always used 40 weight oil.

4 time attacks, and 31k kms now

passenger side to drivers side

180-180-180-165

oil bumps it back up to 180

doh

leak down test coming soon



on a side note, during time attack, oil temps would hit 260 and plateau, very quickly. Within a couple laps during test and tune
i did an oil analysis at some point and i was told the oil looked fine, high silica was the only problem, probably left the filter dirty too long

I had a dealership do a compression test before I bought my car and they came back with:
190 - 165 - 150 - 175; I think I'm going to do it again when I change my spark plugs, because I don't really trust dealership techs. However, I'm just waiting for ZZB with that low compression on cylinder 3. That low compression did save me $2,000 though. Lol.

Have your head or block ever been machined? We ran a compression test on a friends miata after it was rebuilt and our #'s were 10PSI higher than the max spec. It was because the head was decked a few thousandths.

No I bought the car new in 09 and have never removed the head.

Does Mazda actually allow engines to be rebuilt? AFAIK, they will replace the engine, but not allow a rebuild. Ford is the same way, and it is due to the "new age" machining required.

If his head or block were milled, it probably wasn't through Mazda :)

Maybe your high compression #'s have something to do with you making more power than most? Maybe you don't have a magical K04 after all :scratchchin:

Why I have such high compr. #'s I am not sure. Tight engine is tight? I don't really know.

I checked compr. again 2 nights ago and it was still 200~205 in all 4. On a warm day I'll see up to 210. TBH, I don't even care that they are strong #'s, I am just glad they are all the same @ 63,000 miles !

The Skyactiv motors are not rebuildable period.

It just means all of your rings and valves seal well, Your engine is pretty much as healthy as it gets. :biggthumpup:


Why is that?

Machining operations require purified unicorn semen.

Sent from my PC36100 using Tapatalk 2

Sprayed on steel cylinder walls IIRC


Tappin

mind blown... lol steel in a can? So basically it's not possible to hone/bore them?

You got it :)

disposable blocks. but then again Mazda will not rebuild a motor, ever. they put new ones in so to them there is no reason to build a block that can be resurfaced. I suspect that saves them money and with uber thin cylinder walls, heat transfer to the coolant has got to be much improved.

A good 5-6 years ago one of my friends was working in a machine shop where they rebuilt motors. He was going through his apprenticeship. We were building one of my motors at the time and now he is an HVAC service tech for Honeywell. He said there is no future in the engine rebuilding business and that shops are losing more clientele each year. No one fixes TVs anymore either.

It is part of evolution. The processes employed by the factories have such tight tolerances and are increasingly less expensive that it is very difficult to achieve the same for a small shop/business. The shops that remain in business will only be dealing with older engines and cars ... but the good thing for us, the consumer, is that new factory engines are better and better off the shelf.

so making monster power on future cars is gonna be insanely expensive because you would need to start with a new block? sigh... or drop in a different engine completely

Mazda does rebuild the rotaries, but as far as I know the piston engines are all new.

Figured I would post my results here as well; 182, 180, 120, 195. After pouring a little oil in the results were 185, 182, 121, 200. This is a built motor with about 1500 miles at the time of the test. I'll be doing a leakdown test as soon as possible.

Ben, I hope it is related to a valve!

I'm thinking it is. Unfortunately that would mean tearing into the engine again possibly. I kinda hope he just didn't quite get the measurements right when he checked the clearances; I think that would be an easier fix. I don't think it's the injector seals; for one thing, we were very careful about putting the injectors back in. Also I'm not seeing much KR at all on any of my logs.

I don't think its the injector seal.

I had a completely leaking toyota seal once on cyl #2 and my compression was no where near that low.

It never hurts to check these things, however.

I thought the funniest part of the conversation was when he suggested that the rings were still seating, that the motor isn't fully broken in. Probably not, but damn, why would 1 cylinder be that much lower? I hated to hear that, because I liked this guy. He seemed like someone I could trust.

Good luck. :happysad:

A leakdown test will show you where the leak is coming from with such a large compression difference.

Did a compression test today. Rented the gauge from a newly opened Autozone.

From left to right: 190/165/190/195. Wet test of the low one brought compression up to 190.

Main reason I tested the compression is because I'm getting some oil out of my valve cover vent.

have you been watching your oil temps at the track? what oil have you been using?

I have not been watching my oil temps at the track as I have no way of doing so yet. I've been running Rotella since about 40k.

Did this start happening after a track event? It may be the ringlands. If you watch the hot idle vacuum readings you should be getting less vacuum at idle with lower compression. Any other drive ability issues aside from the oil in the intake?

I'm not sure when it started. I track/autox through out the entire summer, the car is and has been running fine so I never had a real reason to remove the intake. Only noticed it when I took out my maf to clean it. Right now I have a small catch can in between the vent and my intake.

I'll make note of my idle vacuum when I leave work and when I get home.

Exactly the symptoms I went through when I broke my ringland in #3 .

First started to "oil" my maf and intake so I put an OCC vta on the VC vent. Fast forward a few months and the failure was confirmed

Seems this has now become a mainstream failure mode. The uber tight OEM ring gaps coupled with high oil/coolant and cylinder temps overall is taking its toll and the rings butt.

Sorry to hear it.


Tappin

I thought I noted my mileage but apparently I didn't. Engine has 117k on it.

I think we have to start thinking about removing more heat from the motor on high power or tracked cars in hot weather or at least monitoring temperatures. This is something I am looking into.

Thanks for the insight. Please do post the warm vacuum readings.

I posted this in another thread but:

is it time to go to cooler thermostats or get better fan control, oil coolers? What would be the least troublesome approach that would yield the best results?

I personally don't like seeing my car run at 217-220ish all summer long. The fan will not even come on until it is 217 or the motor is shut off which ever comes first.

I have no idea what the thermostats are rated at and if cooler models are available, and if that would make a difference in a DD situation. on an open course with a ton of airflow maybe that is all that is needed but in city traffic it may make little difference.
@Ziggo; has replaced his I think and wonder what it is rated at and if he knows of the availability of others and what difference it made in his ECT...is does live in Texas for god sake.

this may not be necessary for a forged motor, yet, but better heat management for the OEM ring gaps is def necessary.

Agreed.

I only did one trackday last year (on one of the hottest days of the year in the Midwest) after you tuned my car, and had to end my day early (before the end of the 5th of 6 sessions) due to heat-related issues (slipping clutch primarily).

I'm not a "high-horsepower guy" by any means, but I do tend to drive the proverbial piss out of my car and am still at the very least going to be running one of the map iterations you sent me that has less timing advance in it the next time I go to the track because of this (at least until I get a "fresh" clutch and perhaps upgrade to a Koyo radiator).

I'd also like to get an oil temp gauge (and maybe even transmission fluid temp guage, a la @Tomas;'s setup) as well...

It depends really. A tstat that opens earlier may lower your cruising temps but it may do very little once you exceed the cooling capacity of the system at WOT. Perhaps delay things a bit at most.

The car was designed to reject 270hp worth of heat and 400whp cars are at 450+crank. That's 50% more heat the must be rejected. Perhaps starting another thread about this is best.

Agreed, a dedicated thread to get some ideas out there would be good.

good idea..I'll put one together with liniks to some of the pertinent information...if somebody gets to it first feel free and link it here.

Just got home from work (50 mile commute 80% highway)

Vacuum at idle was 21.56 and when I got home it was the same. The pic with the red screen is the hot idle. Both were with ac off.



Sent from my EVO using Tapatalk 2

So @Lex; what would it mean if I'm not getting any lower vacuum at idle?

Vacuum looks good to me. Usually with a cold motor I see about 19" vacuum and hot motor 21-22.

Sent from your couch

So if vacuum usually drops with lower compression what does it mean that its staying the same?

Sent from my EVO using Tapatalk 2

Does your vacuum fluctuate at all while the car is running, or does it stay pretty steady (at idle or under steady load)?

The vacuum test is just a roundabout way of checking compression. If you are directly measuring compression then I don't really see its usefulness.

Indeed. Best not to "get lost in the weeds" talking about vacuum...

I'd say a leakdown test is in order. ;)

I was just curious how it was related. As for your previous question, no it doesn't fluctuate. seems to stay pretty steady. I know while engine braking it pulls about 23-24"

Now to find time to do a leak down test...

Vacuum readings are a "constant" gauge on compression and that's why I like them. You can keep an eye on it without breaking out the tester but be careful as engine temperature and ambient conditions (temperature, pressure, altitude) all affect the vacuum reading.

If you have a good gauge (or monitor the MAP sensor) you can see a change in your average vacuum if compression drops and this is something you keep an eye on.

It was really interesting to be able to literally watch the compression come up in my new engine. When it was brand spanking new, it would only pull ~18inHg when fully warmed at idle. Over the first 400 miles, it slowly crept up to holding ~22inHg, and it has been steady for the past 1500 miles. Definitely a good indicator for if/when things start to go south.

My compression tested at 228/228/230/232 once it leveled out.

So i just broke 50,000 on my built motor and i did a compression test last night here are the results..

180
http://i206.photobucket.com/albums/b...ps2c7f3081.jpg
170
http://i206.photobucket.com/albums/b...psd33491fd.jpg
180
http://i206.photobucket.com/albums/b...ps06dd390b.jpg
180
http://i206.photobucket.com/albums/b...psa4254edc.jpg

did mine last night. sobering results
160
155
155
170

MS6
72k miles
1 full year BT
spraying meth for the past... 3-4 months?


I get ~21-22 in/mg

i've had "coolant issues" in the past where tube falls off, gets disconnected, shop that did clutch forgot to put coolant back in....
so my car has "overheated" about 3 times. I usually catch it way before the needle goes up to the top, thank god to Dashhawk and alarm set to 230*. But overheating is still no bueno. I'll attribute my shitty numbers to those instances.
Yeah leakdown test is coming up next.

MS3 - 68k miles. I'm pretty happy with my results, was fearing the worst.

170 - 170 - 160 - 160

It's unlikely you lost compression from hitting 230F coolant temps. If anything you might have a leaking head gasket from this but not compression problems.

I don't even think you have compression problems. Do you have excessive blowby? Consume oil? Did you crank until the needle stopped rising in all cylinders?

As these motors wear compression will fall off a bit but I think some of these lower numbers can also be attributed to the testing procedure and tester itself.

230F, not C. Silly Canadian.

My compression test numbers before buying the car:
190 - 165 - 150 - 175

I have noticed that since I started running 18PSI, it seems like I've been losing/burning more oil. Every fill up (about 1 week/200-250 miles), I lose about 1/4 of a quart of oil. I assume that is PCV related, since it only started happening after turning up the boost, but an OCC should put an end to that, if that's the case.

Clearly there is something up with cylinder #3 , but I have not done a leak down test. The car seems to run fine overall, maybe a slightly rough idle, but it also has not had the valves cleaned yet and needs a new VVT actuator/timing chain. It makes pretty good power, but @cld12pk2go; pointed out that I seem to be making lower numbers than he would expect. However, I'm also right on par with my old speed3 with the same mods in terms of crank horsepower, so I'm not too sure what to think.

Honestly, if anything, I expect it to be a cracked ringland. The oil loss and low compression point right at that. I will do a leakdown test sometime soon and try to figure it out. The best part? The dealership tech who did the test stated "the engine is in very good health" on the work order. It's pretty much outside of Mazda's specs even if you don't count the highest cylinder. It has received a healthy amount of beating over the last 3 months, so I may do a compression test next time I pull the TMIC, just to see if it has changed.

I figured as much about the headgasket, no sign of poopoo coolant in my oil.

I can't tell if i have excessive blowby because i only drive .8miles to work, I get a TON of water in my OCC so I have to drain it nearly weekly in the "winter" due to the milkshake machine I have going on in there.
The intake tract was never connected until I installed the Cobb prototype inlet hose, but I did have a crankcase filter, never absorbed any oil on it though. Doesn't look like the hose has anything in it really either, just a very thin film (light vapors?) inside.

After reading through the entire thread, i saw that some of you guys cranked up to 20 times... i think I only did 3 to 6 at the most. Even though it was levelling off pretty decently. Maybe I'll get a second tester to try this out again.

thanks for the input.

Check for head lifting.... does coolant come out of the over flow or past the cap after hard driving? when the car is cold start it up and see if the water shoots out of over flows really fast.....

Nope, none of that. I think something may have been a little off during the test. car is probably fine.

thanks though.

Guys and gals, cylinder 3 takes more cranking than the other ones to bring up the compression on the gauge. Don't just crank 3 times. Crank until the needle stops rising with foot to the floor (14-18 cranks) and all spark plugs out.

If you want to do a more thorough test do a leakdown.

the "gals" part was for me huh?

I stopped because friend said the needle stopped after about 6 cranks.

like i said, i'll try it out again and i'll be a little more thorough. we did this when going out to eat and decided to do it on a whim since he had the tester in his car. We were pretty hungry so maybe we were rushing a bit.

Still, they're within tolerance spec and arent horrible numbers, so i feel somewhat OK about that :)

180 180 150 180

cranked 3rd alot with hopes and dreams.

Oil in the intake?

Yessir.

:frown: