My car's compression tested low. Help.

Status
Not open for further replies.
Joined
Dec 21, 2009
Messages
1,577
Location
Tulsa, Oklahoma, USA
I bought one of the compression testers from Amazon ( this one).

Cyl 1: 130
Cyl 2: 130
Cyl 3: 130
Cyl 4: 135

Car was fully warmed up. Then I parked it and let it sit for 10 minutes as per the instructions. Then I took the Intercooler off, pulled all the plugs, pulled the fuse for Ign 2 in the engine bay. My wife held her foot to the floor (to open throttle plate and turn off injectors). We let it crank for about 5 to 7 seconds. It was enough that the compression wasn't going up at all anymore.

No blue smoke. The first crank yielded I think 80 maybe a bit more.

It's a 2009 Mazdaspeed 3 GT. Car has 13,700 miles on it. Don't know the elevation. Northeast Oklahoma though. All cylinders were 130 except cylinder 4 was 135. I tested 10 minutes after the car was fully warmed up and parked. The directions said to wait 10 minutes something about aluminum heads.

Also, when I added a teaspoon of oil the compression tested 150 across the board.

They are all close to each other but other people's compression for this car is 180 to 185. I think that means I'm below the 28.5% tolerance.

I have changed the oil myself. Currently has Rotella T6 in it w/ about 1k miles on it. Before that I had German Castrol in it for a run of only 1k miles and before that I had Pennzoil Platinum for a run of 2,700 miles before that was Castrol GTX for 1k miles and before that was the free oil change and it went 4k miles and before that was factory fill for just barely over 3k miles. I just bought 12 quarts of Redline 10w40 and was just about to put it in before I tested the compression, now I don't want to waste it on this engine.

The thing I worry about is since there's no CEL them saying well nothing is wrong the car is fine. But the major power loss I had led me to do this compression test as I KNEW something was wrong that I had lost lots of power. So do I just take it in and say this to them? The car wasn't stock, it had..

SRI
TIP
BPV
Racepipe
RMM
HP FP internals (worked great)
and some misc stuff like shifter bushings, new transmission oil etc.

I could probably put it back to stock in a couple of hours.

Also I've been checking and changing the plugs quite a bit. Took the stockers out for the Deno 1 step colder then I pulled them noticed a lot of oil around the edges of them. Tried NGK 1 step colder and it ran terrible so I pulled them and the same thing, oil on them. I just put in some stock NGK's. Would "blow-by" be what's causing the buildup on the outer part of the plugs?

Thanks for any help/advice.
 
Last edited:
The only thing I worry about (numberwise) when using a compression tester is cyl-cyl variance. The absolute number is effected by other variables (cam duration, cam timing, gauge accuracy). If all of your munbers are within 5PSI, look somewhere else besides cylinder leakage for a problem.
 
Where else should I look punisher?

Also, this low is below the 28.5% variance. I think the manual states that if your compression is 129 or lower you have a problem. Mine is 130.
 
Last edited:
I would buy another compression gauge and recheck before you take it to the dealer.

It seems unusual that every single cylinder would be low, especially if the car still runs and there's no CEL. Makes me wonder if the gauge is accurate.
 
Does engin employ VVT? I would think under crank though it would set cams for zero overlap. Does this not use ECT, you confirm throttle was open? 185 PSIG on a low compression TURBO engine? When did you notice loss of power? does engine run smooth or buzzy? Sorry, I am asking too many questions ...
 
Last edited:
I don't know what drive by wire is, but I'm doing the test according to how the mazdaspeed guys say to do it. Holding the gas pedal to the floor opens the throttle plate and shuts off the injectors, according to them. I don't know what VVT is ARCO.

The car come stock with 9.5:1 pistons and is Direct Injected with Turbo. I noticed the loss of power back in late December of 09. The car runs perfectly fine. Very smooth and no other issues other than the plugs seem to foul relatively quickly. But when I lost power it was substantial and very noticeable. One of the things I used to do was go along in 2nd gear and push the pedal to the floor, the tires would light up. No dropping the clutch or anything, just go along in 2nd and then put it to the floor. Now when I do that it can barely get a squeak out.

Also, when I did the wet test all cylinders tested at 150.
 
Last edited:
Originally Posted By: shpankey
I'm doing the test according to how the mazdaspeed guys say to do it.


I've gotten lots of bad info from people that said they knew what they were doing.

Have you checked any shop manuals? I know before they were available online. Maybe reading through one of those could help you out.

If thats not available I'd find a good Mazda or import shop and talk to them about whats going on. Stay away from the dealer by all means.
 
I don't think that's the case here though. These guys are really good and work for COBB, PTP, CP-e, PG and the like. Also, they did their tests like mine and all have 180 to 185 on all cylinders.
 
As long as your computer firmware is the same as those you're comparing against, the only variables are your engine and gauge... unless you got the forum-recommended gauge too.

How about you put it back to stock, reset the ECM, and drive it for a couple days to see if it improves? You can always go to the dealer and complain about low power, pay for a compression test out of warranty (even though you're in warranty), then have those numbers on file if you do have problems later.
 
Yeah, it's supposedly a really good tester. A friend of mine is coming over tonight and trying to borrow his dad's compression tester so we can verify it. Though since the teaspoon of oil took the cylinders from 130 to 150 I fear the worst.

I don't have anything done to the ECU or any kind of chip or what have you. That part of the car is all stock.
 
Last edited:
I wouldn't say that compression testers are useless. They do have their place but the accuracy of them can be variable.

Too many things can effect the readings. Having a slightly dragging starter or discharged battery can effect cranking speed and the readings.

Adding oil will raise the compression if the valves are ok. And it did in your case. No rings seal 100%. If it runs ok and doesn't foul plugs I wouldn't worry about it.

A leakdown test is much more accurate than a compression test.
 
Is this low power problem you have intermittent? What type of fouling is happening to the plugs? Can you post a picture of them? What are these performance parts you have installed? I know SRI is short ram intake but what do the other acronyms mean? You've never had a check engine light? The absolute value of the compression test doesn't matter, it's the variation between cylinders that does. I am concerned about the results of the wet test though.
 
Late valve timing will lower the cranking compression. I notice all cyls are pretty much the same.
Any chance you slipped a tooth?
This would also make the car run way different than normal.
Like mentioned, any variable valve timing on the intake cam could cause low readings, also.
 
My car only has minor mods really. As for the acronyms (you know SRI)...

TIP = Turbo Inlet Pipe (in between the SRI and Turbo)

HPFP = High Pressure Fuel Pump (internals in my case)

BPV = Bypass Valve (recirculate, Turbosmart brand)

OCC - Oil Catch Can (put this on recently)

Racepipe (just a short mid-pipe that has a catalytic converter in it, replacing it with a straight-through pipe, the car still has a cat though, it has 2 stock).

I don't know about the timing stuff. How do I find out if I slipped a tooth? The car runs ok but not as good as it used to. It sometimes has a rough idle and now occasionally pops or backfires when accelerating hard. This is common on Speed 3's though. But mine used to not do it. Also recently started smoking a little bit at idle. Very little though. I could be smoking while driving too but since it's such a small amount I don't see it. The smoke didn't start until I did the OCC though, which was just recently.

Yeah all cyliners are very close. 130, 130, 130, 135 but like you said, when doing the wet test they all went to 150 which scares me.

What's weird is, like you were asking me just now. There are times that I feel that the power is back. Say 8 times out of 10 when accelerating hard the car feels limp. 1 time I'm not sure, maybe in-between. The other 1 time the car pulls HARD, like it used to. What can I make of that? Or is this just in my head?

I will post pics of the plugs. I just took some step colder ones out. They were only in the car for 3 weeks. The inner white part is VERY white. But around the circumference of the plugs they are oily.
 
Last edited:
I would get rid of the OCC who knos what the PCV dynamics are with a turbo setup. HPFP? Why? The stock should be adequate - is this a returnless system? Sparkers; Did you put the plugs in with new gaskets? You cant reuse gaskets. If they are loose or too tight you will damage the head or overheat that plug. Plugs have to be pristene-clean on the insulators, I prewash mine iwth ethanol and clean with a lint free cloth. You dont put a pile of silicon grease on the boots I hope. Playing with the ignition too much can kill it, if just one cyl is mistuned for some reason, the whole car will run bad. BTW, VVT is variable valve timing. Another thing - gasoline is terrible these days. I rarely get a good tank.
 
Last edited:
It could be something in the VVT,my concern would be the 20psi improvement doing a wet test and the plug oil fouling.
Even if the gauge was inaccurate the same gauge being used for both test would be a valid result even though the actual PSI may not be.

IMO being an 09 with only 13k i would do as has been mentioned,return it to stock and get it to the dealer.
 
I'm not sure about MS vehicles, but late model turbo Subies are very sensitive to mods, especially when it comes to changing the intake geometry and removing cats. IMO, you're wasting your money on these mods if you don't tune for them.

Drive by wire means the throttle electronicall controlled. Drive by cable means you have a cable physically attached at one end too the gas pedal and the other end to the throttle. I'm not sure if WOT in your car would actually open the throttle plate all the way, since the car is off, and the ECU sees no need to. I'm asking if this is the case, not saying it is.
 
Status
Not open for further replies.
Back
Top Bottom