Compression check .... Uh Oh!

[priapismic]

A few weeks ago I posted about my '04 twin-spark R and it's irregular idle, bouncing around between about 1000 to 1400 rpm. I had set the idle speed up slightly so it wouldn't stall, as it had been doing when coming to a stop. Running at open throttle was, and remains, just fine, no issues there at all. Several folks suggested I look into replacing the stick coils. A fresh valve adjustment and TB sync didn't cure the idle problem, so last Saturday I rode the 100 miles to the nearest dealer, and had their ace tech look at it.

He ran a computer check and found no problems or faults, so he was kind of stumped. When I mentioned the coils, he said his diagnostics didn't indicate any problems with voltage. So then he ran a compression check, and that's when things got scary. One cylinder is running at 230 psi, the other at 200! That's a big difference, and could explain the lumpy idle.

So now, a couple of questions for you all: what would cause such a big variance? I know that short of a leak-down test, we can only speculate, but could I have a valve problem in the making ('04 model, twin spark, 58k miles)? Sunday I decided to re-check the valves on the weak side, and sure enough, both exhaust valves were tight. Really, really tight. Would that cause a big compression difference? Of course I re-adjusted them back to spec, but now I think I'd like to re-check the compression on that side.

I've never done a compression test before, but from what I read, it's seems pretty straight forward. But do I pull ALL the spark plugs to do the test? Primary and secondaries? And what do I do with the plug wires? Do they need to be grounded to something, or can I just let 'em hang?

If the compression difference is still there, would it be best to just park the bike until I can get it back into the shop for further evaluation, i.e. leak-down test? Or am I taking a risk of blowing a valve, and hence doing MAJOR engine damage if I continue to ride it?

TIA for any suggestions.

[ruddy]

The tight clearances for the exhaust valves could have caused the difference, because the valves may not have been seating correctly. I would definitely try another test. The way I have always done it is with a warm engine. Insert or screw the gauge into the spark plug hole and crank the engine until the pressure quits increasing. If the pressure seems to be low, the next step is to squirt some oil into the cylinder and repeat the test. If the pressure increases, the rings are worn.

I would leave the secondary plugs in and remove the primaries. I never worried about the plug wires when testing car or outboard engines.

[NoRRmad]

Hm. any danger of the secondary plugs actually igniting the mixture and subjecting the gauge to full combustion pressure? That'd be an order of magnitude higher than 200 PSI...

[CycleRob]

[priapismic,[/b] You have burnt exhaust valves. As the sealing/mating seat area burns and/or wears away the valve clearance will get smaller. The significant change you measured after relatively few miles signs the deal. They are currently $80.59 each by my Parts search .

About the compression test . . . you MUST:
--Disable the ignition -or- externally ground out sparkplugs plugged into the powered wires.
--Disconnect both injector's power wires.
--Use full throttle while cranking.
--Count the same number of compression pulses (usually 6 to so the readings are comparably accurate.

A leakdown test is preferred because:
--You're not cranking the engine at all, sparing the battery & starter any trauma.
--No need to disable ignition and EFI.
--It tells you exactly which parts are damaged, and with experience, how urgent the repair is.
--It gives you an easier to do "engine health score" that can occur at every valve adjustment.

On the bad side the LeakDown test equipment costs more, you need 100psi shop air and a wrench turned, flywheel marks accurate, TDC of the compression stroke.

[boxermania]

priapismic

First and foremost I'm going to disagree with the BMW Tech's determination and I qoute

he said his diagnostics didn't indicate any problems with voltage.

as the only voltage he can measure is the primary voltage and that tells you nothing about the secondary winding voltage, or high voltage. So that could still be the problem.

That being said, I'm not going to outright disagree with my friend CycleRob regarding a burnt vale, but based on the measured compression pressures, I'd say that your problem warrants further investigation before tearing into the engine.

I would like to know the following:

How long has the engine been running rough?
How long ago (time and mileage) did you set the valves and adjust the TB's.
At the time you set the valves, did you make any adjustments and in which direction, tight or loose and how much?
What was the "tight" clearance that you found on the valves this past Sunday?
Were both cylinders checked? If so what were the values?
Were the valves adjusted to specs afterwards?
Which cylinder has the low reading? (viewed seating on the bike)
Does your bike spend a lot of it's time on the side stand?
Where do you keep the oil level on the bullseye?

The reasons for my asking are:

I've seen you ride and your style is not conducive to burnt valves.
Historically R259's are not valve burners.
You do have ~ 60K miles on the bike.
CycleRob gave me one of his valves when he decarbonizes his engine at ~50K miles and I couldn't measure any wear on the stem or the face, albeit, it had some spots on the seat face. I don't recall if he had run a compression test before removing the heads.

Over time, as the vale contact face wears the valve 'sinks" into the seat, which closes (tightens) the clearance between the valve and the rocker arm. If unchecked, as the clearance tightens it doesn't allow the valve to fully seat and transfer the heat of combustion onto the engine, which is the primary reason that valves burn up, another possibility are carbon deposits that get extremely hot and precipitate the event as well.

I do have a leakdown tester that you can borrow, if you wish, as CycleRob indicated, a source of 100 psig air is required; a small compressor is quite satisfactory as the air consumption is minimal. One can perform the test with less pressure but it does require some calculation to determine the % loss. Your other choice would be if any of your friends have one or even if the dealer can do the leakdown test. Best results are obtained with the engine a normal operating temperature.

BTW, I'm leaving at 2 pm today to visit the family in South Florida and won't be back until December 14th.

Best of luck and keep us posted.

[priapismic]

Rob & Al and others,
Thanks for your thoughts.

I've seen you ride and your style is not conducive to burnt valves.

Thanks (I guess!), Al, but as usual, there is more to this than that. In the interest of brevity I left out some pertinent facts. Here is, as they say, the rest of the story:

Back in late March or early April, I changed spark plugs to Autolites (3923's?) No problem with the primary, but as others have found, our spark plug wrench doesn't fit on the smaller secondaries. I tried a socket, but it wouldn't fit in the plug well, so I did the best I could by twisting the ends with pliers. Both sides seemed secure enough, so I left on a 1,000 mile trip, telling myself that eventually I'll have to re-check those secondary plugs to make sure they're snug. Well, one thing led to another, I never quite got around to it, and some weeks later, while on a day trip of several hundred miles, I noticed the bike was "chuffing". I remembered then that I had neglected to check the secondary plugs, so I told myself that as soon as I got home I'd check 'em. Only I didn't quite get home that day .... well, I DID eventually get home, but only after being stranded on the road, miles from Bum**** Texas on a little back road, where the bike just quit running.

First thing I did was to check on the secondary spark plug, and sure enough, on the right side, the plug had come loose and just .... blew out. I had never seen anything like that before. Here are some picts I took later to document what happened:

Here's the burned out plug:




Here's what's left of the electrode:



Here's what happened to the engine guard:




Not pretty, eh? Well, long story made short, I just happened to have the Bosch plugs that I had removed earlier with me (something told me not to chuck 'em and carry them with me). So with a little plug wrench action to replace the burned out plug, and five minutes later I was winging my way homeward. Subsequent repair of the burned electrode proved reliable enough (first by me, then by a BMW tech), and eventually I replaced all the plugs, and the bike was runnning okey-dokey ever since. Until a few weeks ago.

That's when I noticed the bike was stalling when coming to a stop (I live in rural area with only a few stops signs here & there; traffic lights and actually having to stop are a rarity for me). So I adjusted the idle speed up slightly, which helped with the stalling, but then I noticed how irregular the idle speed was. Hence my original post. Now remember - and this, I think, is important - the spark plug event occurred more than 7 months ago, AND the bike has always run under open throttle just fine. Only hint of a symptom was that lately, my gas mileage was going down slightly. I've done at least two valve adj's since the "event" and never had a problem with clearances until now.

Anyway, I appreciate the comment about my "riding style", but I fear I did myself some damage when I blew the plug out the hole (somehow that doesn't sound quite right ...) and now I expect I'll be facing, at minimum, a valve job on the right side. Oh, and the BMW tech attempted to put fresh plugs in, and discovered that secondary hole was stripped. He didn't have a Helicoil insert for it, so he tapped it out to one size larger and got a plug to go in, but he really wants to do it right and put a Helicoil in there. Weird thing is, I've changed plugs (and removed and cleaned the plugs) since the incident, and *I* never had a problem with a stripped hole. Hmmmmm..

Thanks again for everyone's input, and Al, thanks for the offer of your equipment to do a leakdown test, but since you're in Florida - and you're normally about 5 to 6 hours away from me anyway - I expect I'll have the BMW shop do the leakdown test asap. I'll keep y'all posted.

[CycleRob]

Do you see how combustion gases can melt the steel body of the sparkplug? That happened because it was loose for a long period and was unable to transfer that heat of combustion to the cylinderhead. The bad part is that the temperature of the combustion chamber parts run very hot, just as if you installed a sparkplug hotter than any one ever made! That can/will melt piston crowns and damage exhaust valve seats. Don't be surprised if, after the tech looks into the cylinder thru the sparkplug hole, that he'll tell you you need a new piston. My guess is the vibration you feel is caused by the burn away loss of significant piston crown Aluminum, adversly affecting the dynamic engine balance and lowering the compression ratio on that cylinder.

Look in both the sparkplug holes with a bright light in a dark place. I say it looks like the right side head & jug is coming off soon.

.

[mad1150]

...and if your going to use the Autolites 4164's on bottom, buy a cheap plug socket for them and grind the outer surface down to fit into the plug well.

I admit, it's not pretty but it works! I may have the "Tool & Die" guys at work make me one that's more BMW worthy

[paddle007]

As far as worrying about an ignition coil firing without a plug, it's safer to create a path for the spark than letting it "fly". Without a sparkplug the energy buildup in a coil is much higher. I have seen it done a hundred times without ill effects but you don't want to be the one who gets burned. Now if you can get a buddy to hold the coil its all good .
As far as the secondary plug it needs to be installed and dissabled as the engine will run on one plug. Could be usefull for a running compression check but it is brutal on compression guages. Believe it or not the running compression is lower than cranking compression. I do this on automotive engines which will run with one cylinder dissabled. Don't think our beloved oilhead will run on one. It would run with the secondary plug in but the hot gases destroy the schrader valve in the hose adapter.
You seemed a little concerned about running a compression test. It's not rocket science,don't be intimitated. Check both sides for comparison purposes. Always spin the engine over the same number of times. You can count the number of times the compression guage needle moves. Typically I go for five compression strokes.
If the compression is still way off I would not suggest running it more as the problem will not get better and could manifest itself in ugly ways.