Compression test on 3.6r?

[Ifixtheinternet]

My 13' 3.6r is about to hit 60k and expire the powertrain warranty, so thought I'd do a compression test to make sure it's in spec.

 

I couldn't find any write-ups on this, but going through spark plugs threads indicates the motor mount on one side needs to be disconnected and motor jacked up a few inches to access the rear-most plugs.

 

I want to do a warm test if possible.

Is it okay to run and/or crank the engine in park while one mount is disconnected? I can't figure out how the dealer would do it otherwise.

 

Anything else common I should check for specifically?

 

Thanks!

 

Sent from my Pixel XL using Tapatalk

[capttris]

Personally, I don't know that a compression test at 60K miles will show much. And even if you are a little lower, you should be well within range, especially if you have done all basic maintenance, oil changes, and don't abuse the car and engine.

 

If you are really determined to do this, a leak-down test may prove more valuable.

 

What you are looking to do is definitely interesting though.

Please report back on the readings you get and if you think you need to visit the dealer before the warranty is up.

 

I had my CVT oil changed around 55,000 miles, before the warranty was up. I took an oil sample of clean oil and the waste oil and the waste oil, while dark, was basically healthy. But I got piece of mind, just you are looking for.

[Ifixtheinternet]

I've done all the regular maintenance, except I bought the Mobil 1 "15k" synthetic and was doing oil changes every 8-10k. I've since decided that was not a great idea and moved to supertech from Walmart and plan on shorter intervals. Always use the dealer supplied oil filter.

 

I figured it couldn't hurt and Ive already cleared my schedule so I decided im gonna do the test. It will also be a good chance to get a good look at the plugs and see if there are any early issues.

 

I've only heard of a leakdown test being performed if a problem is found with the compression test, but that's an interesting thought.

 

Not sure if I'll be able to do that, don't have a leak down tester and I haven't found a place that rents them near me yet.

 

Any thoughts on cranking the motor over only mounted on one side, sitting on a jack? To me it seems like it would be fine, thought I'd ask the more experienced people though.

 

Sent from my Pixel XL using Tapatalk

[Ifixtheinternet]

After a long couple of days, I have my numbers.

Wow, if anyone thought changing the plugs is difficult, try to thread a compression hose on these cylinders!

 

Add a clogged/dirty compression gauge, my wife locking her keys in her car and needing a rescue the minute I tore the whole car apart, and I'm really glad this is over.

 

I ended up with cold numbers; it is basically impossible to do a warm test on this engine.

You have so many things to take out, including the battery to get the gauge in there, the engine was cold by cylinder 3, so I redid all cylinders cold. Surprisingly, the numbers were not dropping noticeably on the cylinders I did twice as the engine cooled.

 

Here are the numbers:

 

Cylinder

 

6. 163 5. 157

 

4. 165 3. 165

 

2. 170 1. 160

 

The spec for a new 6 cylinder is 185-213 PSI.

 

I called SOA to see if this is in spec or merits a visit to the dealer and they said they could not look up that info over the phone, but the dealer could help. I called the dealer and they said they could not give out that info, but SOA could help.

But they'd be happy to do a compression test for $150 [emoji849]

 

With a 7.5% spread, I believe that would be in spec if comparing to most other vehicles.

 

Does anyone know what Subaru considers good numbers for the 3.6R?

 

Sent from my Pixel XL using Tapatalk

[FLlegacy]

With a hot engine you would be in range. The extra heat expands the rings and pistons and seals the combustion chamber better.

 

Sent from my SAMSUNG-SM-G920A using Tapatalk

[Ifixtheinternet]

Further along now . . .

 

A supervisor got back to me from SOA and confirmed the acceptable range I found, 185-213 PSI for the 3.6R, but also that the maximum variance should not exceed 7% between cylinders.

 

Since I did get a 7.6% split, that is a Little out of spec. Of coarse that is with a cold test.

But with one cylinder being 28 PSI below spec when cold, I think it's likely not to reach 185 when hot.

 

The dealer wants $400-$450 to do the test, but would be waved if there's a problem in warranty.

 

I'm thinking I'll try to test the outlier cylinders HOT, and make sure I still get out of spec numbers before dropping that kind of cash.

 

Sent from my Pixel XL using Tapatalk

[dgoodhue]

If they do warranty it, Subaru may ask for documentation of oil changes.

[capttris]

I agree with dgoodhue

I think you might be chasing something that isn't a problem.

And even if you can accurately proove a little bit lower compression numbers on all cylinders, that would probably also be considered normal wear. Especially since all of your values are about the same.

 

 

Remember, warranty technically covers defects in manufacturing and workmanship.

 

The PSI for a "NEW" engine is what you listed.

Engines lose compression over time. No matter how you baby the car and what oil you use. (good oil is still critical, of course)

 

If you are intent on chasing this, a better quesiton to ask your dealer, is "what compression PSI results would be considered low enough (while still under warranty) to qualify for a repair" (I say "repair", as it's hard to say what specifically what SOA would offer)

 

My guess is the answer to this quesion may not be published. The dealer would certainly have to do their own compression test, and they may also want to do an oil consumption test.

 

As you are right at the cusp of hitting 60K miles, I have to admit that from a dealer's perspective, this might look suspicious, that you "all of a sudden" have "perceived" compression issues.

 

Again, I'm all for trying to catch and fix everything possible BEFORE the warranty expires, but you might be chasing a problem that doesn't exisit.

[DrD123]

The measurement needs to be warm/hot - otherwise, you will read low and have inconsistencies. Compression info for the 13 is below. Is the car exhibiting a problem that drove you to worry about compression (misfires, burning oil, etc.?)

[Dispatch20]

How well is your measurement gauge calibrated? And how long are you cranking the engine for? 1 second? 5 second? Until the gauge stops rising? These are huge variables.

 

For example, on the same exact 3.0R engine with 176,000 miles, under the same conditions (engine temp), I used two different gauges.

 

Snap-on Gauge:

Bank 1: (#1,3,5): 195, 200, 200

Bank 2: (#2,4,6): 220, 220, N/A (didn't bother)

 

Autozone loaner Gauge:

Bank 1: (#1,3,5): 160, 160, 160

Bank 2: (#2,4,6): 190, 190, 190

 

That's right. 30-40psi difference in the gauge readings with all else equal. In my case, the Bank 1 timing chain was off by 1 tooth which caused lowered compression and misfires. Doing a compression test was a worthwhile use of time to debug a major issue. 160psi compression was enough to misfire 5% of the time (although offset ignition timing also played a factor). The snap-on was reading far too high (220psi on a good cylinder), and the Autozone junker was working well (190psi on a good cylinder).

 

In my opinion, it's a poor use of time and resources to measure compression on a relatively low mileage H6 with no symptoms. You aren't seeing excessive blowby, misfires, issues with the plugs, or any other symptom. But your #'s look fine, and if you pay Subaru to do a compression test with their equipment, I'm sure you'll see that the numbers will be actually in spec. You're wallet will be lighter, though.

[Ifixtheinternet]

How well is your measurement gauge calibrated? And how long are you cranking the engine for? 1 second? 5 second? Until the gauge stops rising? These are huge variables.

 

For example, on the same exact 3.0R engine with 176,000 miles, under the same conditions (engine temp), I used two different gauges.

 

Snap-on Gauge:

Bank 1: (#1,3,5): 195, 200, 200

Bank 2: (#2,4,6): 220, 220, N/A (didn't bother)

 

Autozone loaner Gauge:

Bank 1: (#1,3,5): 160, 160, 160

Bank 2: (#2,4,6): 190, 190, 190

 

That's right. 30-40psi difference in the gauge readings with all else equal. In my case, the Bank 1 timing chain was off by 1 tooth which caused lowered compression and misfires. Doing a compression test was a worthwhile use of time to debug a major issue. 160psi compression was enough to misfire 5% of the time (although offset ignition timing also played a factor). The snap-on was reading far too high (220psi on a good cylinder), and the Autozone junker was working well (190psi on a good cylinder).

 

In my opinion, it's a poor use of time and resources to measure compression on a relatively low mileage H6 with no symptoms. You aren't seeing excessive blowby, misfires, issues with the plugs, or any other symptom. But your #'s look fine, and if you pay Subaru to do a compression test with their equipment, I'm sure you'll see that the numbers will be actually in spec. You're wallet will be lighter, though.

 

I haven't had any obvious issues with the car, aside from possibly losing a little oil between changes but that seemed to stop itself. Also the engine had more vibration than I'm used to when warming up, thought there might be a power Imbalance. But it's my first boxer engine so that's probably why.

 

I just wanted peace of mind the engine is still healthy.

 

I wasn't using a calibrated gauge, just a basic loaner from O'Reilly's.

 

I was cranking for about 10 seconds with WOT, and recording a video of the gauge on my phone to see what the actual peak PSI was. If it didn't top out, I'd crank some more.

 

Today I received a brand new gauge from Amazon that is extremely well rated.

 

Redid the test with a hot engine (ouch) and was able to get a test about 10 minutes after shutting the engine off. I tested the cylinder that was reading the lowest before (5) first.

 

I only got 137 PSI?!

 

Thought I did something wrong so unhooked it and redid connections, same. Tried a different cylinder, same. So this very well rated gauge is even less accurate than the cheap loaner.

 

I didn't realize it was such a crapshoot with compression gauges!

 

I did find that cylinders 1 and 5 were basically identical, maybe 1 PSI off with the engine hot.

So to me that says it's gotta be less than 7% split when hot which would be in spec.

 

I wanted to have peace of mind that the engine is healthy before I'm outta warranty, but maybe I was barking up the wrong tree with this.

 

Guess I'll be sending this gauge back.

 

The dealer said they'd do a multi point inspection free of charge so I'm just gonna do that.

 

Thanks to everyone who responded with advice.

 

[emoji2]

[DrD123]

if all the cylinders read low like that, and within a psi or two of eachother, the gauge might not be sealing in the plug hole.

 

I'm all for inexpensive tools if they are decent quality, but I have found for measurement equipment, you sort of get what you pay for. I have a mityvac digital compression gauge kit I bought years ago - autozeros, logs data for each cylinder, tons of adapters and extensions, replacement o-rings and valve cores, etc. - really nice gauge.

 

The spec is not 7% between cylinders - it's 7 psi. Also, for compression tests, elevation matters - not sure what your elevation is, but for me I am over a mile above sea level, so I need to correct the data.

[Ifixtheinternet]

if all the cylinders read low like that, and within a psi or two of eachother, the gauge might not be sealing in the plug hole.

 

I'm all for inexpensive tools if they are decent quality, but I have found for measurement equipment, you sort of get what you pay for. I have a mityvac digital compression gauge kit I bought years ago - autozeros, logs data for each cylinder, tons of adapters and extensions, replacement o-rings and valve cores, etc. - really nice gauge.

 

The spec is not 7% between cylinders - it's 7 psi. Also, for compression tests, elevation matters - not sure what your elevation is, but for me I am over a mile above sea level, so I need to correct the data.

Ya I saw that gauge, it's really nice.

Maybe I'll save up and get that one.

SOA told me 7% was the rating, but maybe they mispoke, I see 7psi on the chart. Either way, I think my motor's in good shape.

 

Sent from my Pixel XL using Tapatalk

[Perscitus]

Another way to skin the cat here is to pull an LV (Learned View) using RRs logger, FreeSSM or BtSsm and see how your Learned Timing and Fueling corrections look, where IAM is hanging out, etc.

 

This, together with some targeted logging would give you or whoever finds the patience to review the log/s a good sense of the engines health.

 

Best part - no tune required and either no or limited upfront cost (say VAG com Kline cable and free software or borrowed dongle and free software).