Testing Compression

[asavage]

[I originally wrote this in Sep-2004 in another venue; now I am reposting it here.]

Problem: the Nissan manuals detail testing the compression via removing the injectors and using an adapter which fits into the injector holes.

Snap-On (and others) use a glow plug adapter instead. However, while it works OK for cylinder Nos. 3 & 4 (using a straight adapter between the gauge and the GP adapter), there are clearance problems with regard to the injection pump when trying to use the adapter for the forward two cylinders. Especially cylinder No. 2.

My solution: a custom short-90° adapter. This adapter trims about 1" off the std 90° Snap-On adapter, and easily clears the injection pump for cylinders 1 & 2.

[for the search engines:
Nissan LD28 = GP adapter M3576 (12mm)
Nissan SDxx = GP adapter M3566 (10mm) ]

(click on any image for larger)


Cylinders 2 & 3 are the most difficult, I found.

It was later pointed out to me that the Nissan injector adapter has no hose, so the volume of the Snap-On gauge's hose may affect the numbers. IOW, this setup may yield lower test results than if I was using the pull-the-injectors method using a hoseless gauge.

Using this setup, I received the numbers below. Cold engine, throttle at idle, ten "puffs" per cylinder, two passes per cylinder. DPC control unit unplugged per manual (round connector to the rear of battery) (so no diesel gets injected and spewed out all the open glow plug holes while cranking). Remote starter switch makes this easy.

Note that, done properly, the engine should be at or near operating temperature and the throttle should be propped open, so these numbers aren't as representative as they should be. So sue me.

(all numbers in PSI)
No. 1 = 470
No. 2 = 445
No. 3 = 420
No. 4 = 440

My generic SD-series manual lists:

Standard = 427
Minimum = 356
Max difference between high/low = 43

I'm OK with those numbers. The 420 could be better, but it is what it is. With all glow plugs working now, if I let it glow for a further 10 seconds after the glow indicator light is extinguished, it fires up on all four and idles perfectly. If I instead begin cranking when the glow indicator light goes out, it comes up a bit fussily and yields more smoke.

I've replaced the valve stem seals, now I need to get it warmed up and adjust the valves. Might see a small improvement there, but I don't expect much.

If anybody wants to perform their own compression test, I will loan out this setup (with a suitable deposit: I want it all back!) for short periods (out of my hands for one week at a time, max).

[glenlloyd]

Nice setup, I like it. I have a kit I haven't looked at in years, supposed to have adapters for just about anything, but it doesn't have a 90 degree adapter I don't think. Where did you get your custom 90 degree adapter from?

Also, does Nissan specifically say that their PSI numbers are taken with a hoseless comp. tester? Aren't these tester hoses specifically designed to withstand high pressure? If anything I can't imagine that it's off by much, if at all, afterall, the gauge will move before that hose expands IMO.

thx

steve a

[asavage]

Nice setup, I like it. I have a kit I haven't looked at in years, supposed to have adapters for just about anything, but it doesn't have a 90 degree adapter I don't think. Where did you get your custom 90 degree adapter from?

Oh, it's custom-made. For the adapter end, I bought a lot of five surplus quick-disconnects that are identical to the ones that Snap-On uses (found a liquidator on the web). For the hose end, bought a std air fitting for compressed air from the hardware store. Took the Snap-On brass 90° fitting and those pieces down the street to a machine shop (job shop) and described the problem and what I wanted to do: low-profile 90° adapter setup. I gave max height dimension. Prototype No. 1 was still too tall. It worked, but was a bit hard to use. I took it back and they cut it down some more. I'm really happy with it. I haven't tried to use it with the LD28 yet, but it works slick on the SD22.

It was later pointed out to me that the Nissan injector adapter has no hose, so the volume of the Snap-On gauge's hose may affect the numbers. IOW, this setup may yield lower test results than if I was using the pull-the-injectors method using a hoseless gauge.

Also, does Nissan specifically say that their PSI numbers are taken with a hoseless comp. tester?

Well, the factory gauge is a Kent-Moore built unit designed for use in the injector hole, and doesn't have a lot of volume. I assume that adding the hose volume does introduce a not-insignificant error to the reading. It should likely be compensated, but I haven't thought too much about it.

When I do a compression check, I'm looking for variation across cylinders, and compression above some threshold. The actual numbers aren't too much a concern, unless they get down below about 370. Below 350, and it won't start reliably in cold weather. BTDT with other diesels I've owned.

I have a Snap-On cylinder leakdown test setup, but I don't know (and haven't tried to see) if it will interface with the glow plug adapters. Might be interesting.

Aren't these tester hoses specifically designed to withstand high pressure?

The compression testers designed for use with diesels have completely different hoses than those for gassers.


That's hydraulic hose, not air compressor hose. The gauge reads to something like 500 PSI. Don't even think about trying to use a gasser tester on a diesel, it'll blow apart.

If anything I can't imagine that it's off by much, if at all, afterall, the gauge will move before that hose expands IMO.

It's not that the hose will change volume during testing, it's that the hose volume itself becomes part of the combustion chamber volume. The compression ratio is altered by the hose's volume, just as if you'd used two head gaskets, or took a chunk out of the combustion chamber. That lowers the compression ratio, and that in turn lowers the peak PSI that's read on the gauge, making comparison with the published test numbers difficult.

I do not know the correction factor. I could take half an hour and calculate the hose volume vs OEM combustion chamber volume, and come up with a correction factor, but it's not of pressing importance to me.

[asavage]

Nice setup, I like it. I have a kit I haven't looked at in years, supposed to have adapters for just about anything, but it doesn't have a 90 degree adapter I don't think.

The Peugeot that I tested (you saw the pics) didn't need a 90° adapter to test. I think I *might* have used a 45° adapter though. I just can't recall.

[glenlloyd]

When I do a compression check, I'm looking for variation across cylinders, and compression above some threshold. The actual numbers aren't too much a concern, unless they get down below about 370. Below 350, and it won't start reliably in cold weather. BTDT with other diesels I've owned.

For past cars that I have felt the need to check that's what I've looked for as well.

The compression testers designed for use with diesels have completely different hoses than those for gassers. That's hydraulic hose, not air compressor hose. The gauge reads to something like 500 PSI. Dont' even think about trying to use a gasser tester on a diesel, it'll blow apart.

Someone suggested that to me once when I said I needed to buy a compression tester for diesel and my jaw dropped. I explained why their idea wasn't good.

It's not that the hose will change volume during testing, it's that the hose volume itself becomes part of the combustion chamber volume. The compression ratio is altered by the hose's volume, just as if you'd used two head gaskets, or took a chunk out of the combustion chamber. That lowers the compression ratio, and that in turn lowers the peak PSI that's read on the gauge, making comparison with the published test numbers difficult.

Bingo, that's what I managed to forget, but I recall reading that somewhere in the past.

Steve A

[asavage]

I've been thinking about the tester's hose volume. Here's an add'l wrinkle: the GP adapters all include a Schraeder valve in them. That means that the hose becomes pressurized mostly/mainly on the first stroke, and its volume has a severely reduced effect for subsequent strokes.

[philip]

SNIP- That means that the hose becomes pressurized mostly/mainly on the first stroke, and its volume has a severely reduced effect for subsequent strokes.

Virtually all professional motorcycle mechanics know this for the same reason.

[asavage]

While I appreciate the splash of color, I don't follow you.

You brought up the subject: differences in compression test results when using the Nissan (direct, injector) adapter or using a GP adapter with a hose. Are you saying that you (or virtually all professional motorcycle mechanics) have known that there isn't a significant difference between results obtained via direct and hosed connections?

Or did you mean something else?

[philip]

Are you saying that you (or virtually all professional motorcycle mechanics) have known that there isn't ...

"We" all have learned how a small combustion size relatived to the compression gauge hose will delay and lower the reading. So "we" continue to crank until the compression pressure reaches a high level. You'd have to use a "direct" compresson gauge to read a useful first push pressure.

When the air throttle is closed instead of opened, the pressure reading will be lowered. And "we" know to compare a dry vs. wet test thanks to rings.

All said ... when choosing between compression vs. leak-down, I will take the latter.

[asavage]

I was posed a question yesterday:

What do you mean by having poor oil control?

Only the topmost ring is designed to be a compression ring: gases under compression above the piston try to escape down alongside the piston, are vented to the piston's topmost ring groove and push the topmost ring outward to make the ring/piston/cylinder wall seal. The "springyness" of the ring has almost no part in the sealing. The ring must be free to move in the groove in order for it to seal to the cylinder wall.

OTOH, the lowest ring performs no compression duty at all: it is an oil scraper, designed to scrape oil off the cylinder wall and return it to the sump. Gases escaping from the combustion chamber toward the sump do not aid in the lowest ring's function.

The ring(s) in between the top ring and the bottom ring generally perform a combination of those functions, with the emphasis on sealing rather than scraping.

Piston_Rings_01b.png (14.38 KiB) Viewed 8228 times

pistonringsplacement.png (73.28 KiB) Viewed 8228 times

Piston_Rings_02b.png (20.14 KiB) Viewed 8228 times

piston-ring-sealing-face.png (44.19 KiB) Viewed 8228 times

It is possible to have good compression and poor oil control when the oil scraping function of the lowest ring is poor. Rings wear and stick from various causes, but the compression ring must move in its groove every cycle in order to get good compression, so it is "exercised".

The lowest, oil control ring does not have any compression acting on it. It is entirely reliant upon "springyness" and if it is stuck in the groove, it is ineffective at controlling oil from migrating up the cylinder wall.

I have found that the SD engines will often have decent compression and poor oil control. One cause might be long oil change intervals that allow crud to build up around the oil scraper ring, but that is only a guess.

All the above applies to four-cycle engines. Two-cycle engines of the non-diesel variety don't need to control oil from the sump. Diesel two-cycle engines do still need oil control rings, but that's another topic entirely.

[scrapman1077]

Thanks for this post !

[asavage]

I have two of these Snap-On compression gauge setups, and one of them has developed a gauge problem. Notice the needle:

Diesel_Compression_Gauge_06b.png (5.52 MiB) Viewed 8219 times

Diesel_Compression_Gauge_06-1b.png (296.54 KiB) Viewed 8219 times

I don't trust this, and I don't have a simple 400 PSI pressure source so that I can compare one gauge to the other. I just bought this 6" replacement, generic gauge from eBay ($28 shipped):

Diesel_Compression_Gauge_07b.jpg (203.84 KiB) Viewed 8219 times

This 6" gauge head will be nice, compared to the Snap-On's 2.5" head, though it doesn't have the kilopascal ranges (which I've never needed).

Meanwhile, I still have the other Snap-On gauge set, though I seem to have misplaced the second low-profile custom adapter I'd had made

[asavage]

Hmmm. I didn't really think about what a 6" dial would look like on this setup. It's, um, huge.

Diesel_Compression_Gauge_08b.jpg (1.29 MiB) Viewed 8188 times

Diesel_Compression_Gauge_09b.jpg (1.13 MiB) Viewed 8188 times

[plenzen]

For those of us that are,,,,,,,,,, well,,,,,,,,,,
getting a few more miles on,

is there ever printing or numbers that are too big ?

[1981kingcabsd22]

So I'm doing the same thing on my 81 King Cab with the SD22. We are just fixing to make our own version of the adapter and I'm like, wait, south Seattle? Can I borrow yours for a day or two? The kiit I bought at Harbor Freight worked fine for 3&4 but won't even come close for 1&2.