Compression test low. So what?

Open for questions of visitors of DAN. Posts of our guests are on moderation queue.

Moderators: Rick, Lance Murray

Hetch

Compression test low. So what?

Post by Hetch »

Hi,

I am in the market for a used DA40 and have found one that meets 90% of my criteria. Sadly the compression ratios aren't close to 80. I am no expert in aircraft engines but wanted 74 70 72 66.

The plane is a 2006 and is in the low 1000 hours. Thoughts?
User avatar
carym
5 Diamonds Member
5 Diamonds Member
Posts: 1021
Joined: Sat Aug 28, 2010 5:00 pm
First Name: cary
Aircraft Type: DA42
Aircraft Registration: N336TS
Airports: KTYQ
Has thanked: 37 times
Been thanked: 83 times

Re: Compression test low. So what?

Post by carym »

It may mean that it needs a top end, but it may also be OK. I would take the plane out for 15 min or so to make sure everything is warmed up and running well, then re-test the compressions immediately. If still low (especially the 66) it will need some cylinder work.
Cary
DA42.AC036 (returned)
S35 (1964 V-tail Bonanza)
Alaska adventure: http://mariashflying.tumblr.com
Hetch

Re: Compression test low. So what?

Post by Hetch »

Thanks Cary! Your answer begets another question :D. About how much would that work cost?

Thanks!
User avatar
carym
5 Diamonds Member
5 Diamonds Member
Posts: 1021
Joined: Sat Aug 28, 2010 5:00 pm
First Name: cary
Aircraft Type: DA42
Aircraft Registration: N336TS
Airports: KTYQ
Has thanked: 37 times
Been thanked: 83 times

Re: Compression test low. So what?

Post by carym »

When I owned a C310 with O470 engines I was changing 1-2 cylinders at each annual :( It cost about $1000 per cylinder, and that was 10-15 years ago. I wouldn't be surprised if it were to cost $1500 now. There are always surprises at annual with ownership :(
Cary
DA42.AC036 (returned)
S35 (1964 V-tail Bonanza)
Alaska adventure: http://mariashflying.tumblr.com
User avatar
rwtucker
5 Diamonds Member
5 Diamonds Member
Posts: 1283
Joined: Thu Feb 21, 2013 11:24 pm
First Name: Rob
Aircraft Type: DA40
Aircraft Registration: N831BA
Airports: KFFZ KEUL
Has thanked: 100 times
Been thanked: 110 times

Re: Compression test low. So what?

Post by rwtucker »

The discussions have centered more on Continental but in general the compression issue is over-hyped. A Continental entity put out a paper saying that a lot of engines with compressions under 60/80 were being needlessly rebuilt. They also debunked the "lower compression = lower HP" myth. They dyno tested an engine with all cylinders in the 40/80 range and demonstrated full book HP. According to Continental, what matters most is relative uniformity across cylinders and no signs of valves that are burned or warped, etc. and no signs of cracked rings, etc.

There is debate as to how much this applies to Lycomings because of the differences in their design. Maybe another DAN member has more information.

As someone who has taken compressions on the same engine multiple times in a 24 hour period, warm and cold, before and after a flight, etc. I can tell you that each of your four numbers represent a range that reflects many variables, including the position of the rings when you shut down the engine prior to the compression tests. If you happen to shut down with all the rings aligned, as happened to me once when my DA40 engine was very low time, you will get a ~40/80 reading which will return to normal in the next flight. It happened to be associated to an annual so I live with the readings in my logs with a notation of the retest value.

Given that they represent a single reading derived by someone of unknown skill, your compressions look absolutely fine. In the presence of an otherwise good bill of health, I would not hesitate to fly an IO-360 with compressions that were all ~60/80. The first thing you might want to do is re-test, separated by at least one full power trip around the pattern. If you want to be objective about it, you need to have a different person with a different set of gauges run the tests and do not tell them the results of the previous tests. Combined with an oil and filter analysis and a quick look inside with a borescope, you should get a clear picture of the engine's condition. I would be wary of an existing oil test, there is no auditing process for submitting oil samples.
hetch

Re: Compression test low. So what?

Post by hetch »

Thank you all very much! Great context. I am no longer feel disuaded. I will give it a thorough inspection from a qualified mechanic and go from there.
User avatar
Chris B
5 Diamonds Member
5 Diamonds Member
Posts: 843
Joined: Wed Sep 26, 2012 1:52 am
First Name: Chris
Aircraft Type: DA40
Aircraft Registration: N171CB
Airports: KRHV
Has thanked: 210 times
Been thanked: 215 times

Re: Compression test low. So what?

Post by Chris B »

Hetch wrote:The plane is a 2006 and is in the low 1000 hours. Thoughts?
Hi Hetch -

Mike Busch has written extensively about compression (e.g.: link, but a quick search will turn up many others). Personally I don't think that a low compression test - by itself - is a major problem. Cary & Robert's suggestions for a retest on a hot engine make a lot of sense. IMO, a cylinder/valve borescope inspection and examination of the plugs by a knowledgeable mechanic is more important. Thankfully a very good borescope is now available for only about $150.

Corrosion is what dooms most aircraft engines. The weak spot of the Lycoming IO-360 is the cam, which is unfortunately very difficult to inspect. The best proxy is probably a combination of tracking engine use (note oil change dates, and look for gaps) and an oil analysis history, if available. Where the aircraft has been based is also important. Cold & dry is good, hot & humid is bad.

Good luck!

Chris
User avatar
Spinner
3 Diamonds Member
3 Diamonds Member
Posts: 89
Joined: Mon Aug 17, 2015 6:55 pm
First Name: Paul
Aircraft Type: DA20-C1
Aircraft Registration: lots
Airports: CYQM CYFC CYYR CYTH
Has thanked: 27 times
Been thanked: 16 times

Re: Compression test low. So what?

Post by Spinner »

i knew that Lycoming had a SB or SI about compression checks.

http://www.lycoming.com/Portals/0/techp ... ession.pdf

It has some pretty good info on checks. I agree that a warm engine is one of the requirements for a proper check. I would not be concerned unless your compression check is around 60.

As stated previous, a borescope of the cylinder interiors may reveal some corrosion issues if the engine is that old and not utilized.

Paul
User avatar
rwtucker
5 Diamonds Member
5 Diamonds Member
Posts: 1283
Joined: Thu Feb 21, 2013 11:24 pm
First Name: Rob
Aircraft Type: DA40
Aircraft Registration: N831BA
Airports: KFFZ KEUL
Has thanked: 100 times
Been thanked: 110 times

Re: Compression test low. So what?

Post by rwtucker »

Thanks for the link to the Lycoming sheet Spinner. I couldn't locate it when I posted above. The sheet seems to suggest an absolute lower limit of 60/80 but raises concern at 65/80 or an acceleration of decline over time. It also points out the importance of repeated measures, noting valve rotation and seating as an independent variable that can explain an artificially low reading. This is incomplete. Ring alignment is also an important independent variable that can create artificially low readings of greater magnitude.

As a measurement guy, one of the omissions that I find disturbing in all of these publications -- Lycoming's and Continental's -- is a quantitative statement of the SEM (standard error of measurement) in these prescriptive numbers. No one needs to understand this at the statistical level to understand that several factors contribute to measurement error and that, together, these factors define a range within which the actual number you see on the pressure gauge represents the true compression. The main factors that contribute to the total error term are the accuracy and reliability of the gauges, the mechanic's behavior, valve and ring position, and CHT. Given the awkward way some of these tests are conducted, visual angle of incidence can sometimes introduce error amounting to a pound or two.

The gauges themselves deserve special comment. Good practice is to re-calibrate them annually. If you do this, you might get to native accuracy of 2% -- per gauge -- under theoretically perfect conditions (e.g., the gauges are not temperature compensated). Digital gauges probably beat this number. On the other hand, a set of gauges that have never been calibrated after they left the factory could easily be 5% or more off, again per gauge. So here we are with many pounds of uncertainty before we add the other forms of error.

To specify an all-in SEM, someone would have to conduct a lot of tests under specific conditions. My gut feeling is that the expected range for the same person conducting compression tests using the same gauges under similar conditions can usually expect to get within +/- 5 pounds unless all three rings are aligned or a valve rotates and closes on a lump of carbon. I like my IA but I always wonder when he agonizes over whether a reading is 78/80 or 79/80 when I could show him a bigger difference by asking him to stand two feet to the right. When comparing tests conducted by different people on different gauges, I would allow for a much larger error term. In summary, the only way to approximate a true compression reading is by repeated measures and many, not two. These gut feelings are partially supported by my own experiences where my IA has conducted annual compression tests on my DA40 since 1989. The readings he gets for a given cylinder typically vary by 3-4 pounds and not in any particular pattern. It might seem that this small range is less than I am describing in general. It is not. My engine is ~ 400 hours TTSN and my cylinders test near the top of the limit, thus introducing a restriction on the possible range of the distribution. I can expect to see a much larger variation when my cylinders are in the 65/80 range.

One independent and interesting form of error is that of expectation bias. Over time, you will obtain different readings based on whether or not the person performing the tests is aware of previous readings. The best way is for the person performing the tests to have no knowledge of the engine.

Having gone into all this detail, I would pay much more attention to oil analysis, including trends (minimum of three reports) and an inspection of the overall health of the walls, head, and valves.

Our reliance on compression tests as indicators of engine problems is a good example of looking where the light is bright rather than where the problem is most likely to be.
User avatar
Steve
5 Diamonds Member
5 Diamonds Member
Posts: 1953
Joined: Thu Aug 12, 2010 1:23 am
First Name: Steve
Aircraft Type: DA40
Aircraft Registration: N432SC
Airports: 1T7
Has thanked: 81 times
Been thanked: 493 times

Re: Compression test low. So what?

Post by Steve »

Wow the discussion has has had a beneficial effect on my compressions! 13th annual started today. So ran the engine, and compression test was: 76/75/76/76. About 750 hours on the engine. My compressions were always good previously, but there was more scatter in the individual numbers. So far, everything is looking OK, I did plan on replacing the brake disks/linings this year based on what we saw last year. Parts are on hand.

We'll see how this one goes...
Post Reply