When you test your compression, do the readings differ depending on whether you test through the Injector openings or the glow plug openings?
IE: will the reading be less if tested through injector opening?

__________________
Jimmy L.
'05 Acura TL 6MT
2001 ML430 My Spare

Gone:
'95 E300 188K "Batmobile" Texas Unfriendly Black
'85 300TD 235K "The Wagon" Texas Friendly White
'80 240D 154K "China" Scar engine installed
'81 300TD 240K "Smash"
'80 240D 230K "The Squash"
'81 240D 293K"Scar" Rear ended harder than Elton John

I would be somewhat concerned with a tester that used the injector openings.
The tester would need to exactly displace the volume of the injector in order to get accurate readings. If the tester allows any air space, where there would normally be the injector, the results will be artificially low.

This also applies to the glow plugs. The glow plug takes up space. If the tester does not take up the same space, the results will not be accurate.

But, the glow plug is much smaller than the injector.

Surely the compression test is done over a few compression cycles of that cylinder. If the compression tester has a non-return valve, the discrepancy will be negligible.

I have never yet seen a compression tester, the volume of which was matched to the size of a spark plug on a SI engine. The gauge itself has a volume much greater than the spark plug displacement, then you need ot connect the valve to the combustion chamber.
If the internal volume of the compression tester increases the combustion chamber volume by less than 10% of the swept volume, and a non-return valve is fitted where it connects, the error in compression pressure will be less than 0.1% after 3 compressions and less than 0.01% after 4 compressions..

__________________
Tony from West Oz.
Fatmobile 3 84 300D 295kkm Silver grey/Blue int. 2 tank WVO - Recipient of TurboDesel engine.
Josephine '82 300D 390kkm White/Palamino int.
Elizabeth '81 280E, sporting a '79 300D engine.
Lucille '87 W124 300D non-turbo 6 cylinder OM603, Pearl Grey with light grey interior


Various parts cars including 280E, 230C & 300D in various states of disassembly.

I think I understand the nature of Brian's concern. If the adapter displaces less volume than the component it replaces (glow plug or injector), then the volume of the prechamber is effectively increased, and the compression ratio drops appropriately. I guess the question is one of how large that particular difference in volume is with respect to the total volume present when the piston is at the end of the compression stroke.

Here are two adapters from a small Mac set I have. To me, the injector port adapter looks like a pretty good approximation of the business end of an injector. I don't think I'd lose any sleep over using either one.

Attached Thumbnails

 

Let's look at the volume that is prior to the non-return valve. Everything that is past the valve is not relevant, because multiple compression strokes will reduce any error in the gauge itself.

So, say the volume of the cylinder is 22 cc. with the glow plug in place. And, let's say the glow plug is taking up 3 cc. of displacement below the threads.

If you now install a tester that installs into the threads and does not extend down toward the top of the piston, the volume of the cylinder is now increased from 22 cc. to 25 cc. This would equate to a drop in pressure of 12%.

On an SI engine, say the volume of the cylinder is 66 cc. A spark plug, below the threads, has very little volume. There might be 1 cc. below the threads. So, a loss of 1 cc. of displacement due to the spark plug would only be 1.5%.

If the test is performed with the injector holes, the error has the potential of being significantly greater.

I concur.

1. I think yall are using the wrong base to compare from... thus the results are higher than they really should be...even though small anyway...
It should be a comparison between the cylinder volume with the piston at bottom compared to the two situations with the tester in place compared to the original equipment...

EXCEPT that...

2. The figures which are given for making decisions concerning the health of your engines are derived from tests made by Mercedes Engineers using compression testers themselves... and probably exactly like the ones you are using... thus negating the difference completely.... sending this ' problem' to the theoretical file... instead of being something to worry about...

I thank you for pointing out a flaw in the way I approached this mental exercise. If we consider Brian's concern, though (an adapter that displaces LESS volume than what is usually in that part of the prechamber), I still think that it would make for a LOWER reading. It has effectively increased the volume of the prechamber by the same absolute amount on both ends of the compression stroke, yes?

Agreed, I consider the point to be entirely academic! As noted before, I wouldn't lose any sleep over using either of the adapters I have. I'd just use whichever was more convenient in a given situation.

Still, not a bad exercise in trying to understand a measurement and how possible sources of error work (both direction and magnitude). For example, when performing the "2mm of valve lift" measurement to find valve timing, how much is a given error in measuring the valve lift reflected in the timing that is determined?