John..

I am going to run a bleed down on my engine to check for leaks before I pull the heads. I am going to build a bleed down tester out of the following:

Adjuatable air pressure regulator with gauge
Outlet hose which has a second gauge
Hose to engine combustion chamber

The thought is to regulate the pressure to the device and then measure the actual pressure being held in the cylinder for comparison purposes. I can't find a bleed down tester anywhere here and I want to get on with this job soon, so I will make one out of parts I can get together out of the leftovers here at work.

How much pressure have you guys had to use to find leaks in the system. I was thinking around 60-90 psig should do the trick.

This should give me a good indication of where any leaking air might be going. Air out the intake valve should be felt right there, air through the exhaust valve out the tailpipe and air through the rings out the dipstick...I am hoping for the air out the head gasket option...

Thanks

dr bob

I asked a question about the "leakdown" testers once and got a bit of a runaround. It's never been clear to me how they work, since they somehow come up with a result in percent based on pressure differences between two gauges connected together by a pipe. I'll probably never understand... But folks do sell the gauges online, and nobody could give me the scoop on what's inside the pipe that connects the gauges, so maybe purchase is a better method than trying to make something.

Is there a valve in the middle and you look at the pressure decay over time? Is there an orifice in there and you see the pressure drop from flow through that orifice? Anybody got experience here?

My two sense--

90 PSI is a lot of air if the crank is not at perfect TDC on the compression stroke of the cylinder you are testing.

Use the threaded hose from the compression gauge to go to the cylinder. Mine has a quick-disconnect fitting already installed on the non-engine end, and by gosh there's a pressure gauge there too. I still don't know what to do with that thing during a leakdown test though.

Use a stethoscope and just a piece of vacuum hose to hear the air noises. Listen in the intake hole, and since you are going to pull the heads anyway you can remove the whole intake to get a good listen there right at the valve. Otherwise the popular listening spots will be the oil filler cap (for ring/piston leaks) and the exhaust as the last (and maybe toughest) listen.

Seems to me that a leakdown test is a way to identify the cause of low compression in a cylinder. Have you identified a low cylinder with the compression gauge?

I'll be watching this thread carefully, so any guidance on the gauge system itself and the measuring technique is much appreciated.

Incendier

Very intersting topic.

Poached from a 530i site...

"A leakdown test places a gauge in the spark plug hole with the piston at top dead center without moving. The combustion chamber is then pressurized with air from an outside source.

With the regulator air pressure as the standard, the test measures the % of the pressure _drop_ through a particular orifice that results from whatever flowrate leaks out the cylinder being tested. The air flow passes from the regulator through the orifice on its way to the cylinder, and as long as the leakage flowrate out of the cylinder stabilizes, the pressure drop through the orifice will stabilize, regardless of how long you run the test. If air pressure is supplied to the leakdown tester at 100 psi, and the leakage out the cylinders is a stable flowrate through the tester that causes 16 psi of pressure drop through the orifice, the tester reads either 84 psi or 16% leakdown, depending upon how the dial on the second pressure gauge is marked. "

Looks like "an orifice" is the answer to "what's in the tube". (No jokes, please.) I have no personal experience with anything more scientific than the "pressurize and listen" method, but this seems about right.

As a somewhat OT sidenote, a number of wrenches (as well as the technical editor of Excellence) have expressed doubt as to the usefulness of a leakdown test in comparing the actual holding ability of a cylinder. Considering that shop air will never approach the pressures of a running engine, the theory goes, there may well be leaks that appear under combustion pressure that won't be obvious at 100 psi or so.

WallyP

Some comments:

Yep, the tube between the gauges has a calibrated orifice.

If I remember correctly, the Snap-On tester uses 60 psig for testing, so 60 psig on the cylinder gauge is 0% leakage, zero psig is 100% leakage. The assumption is that the leakage rate is linear - that is, 30 psig is 50% leakage.

There are some mechanics that feel that higher pressures will seal better, not worse. The rings are the biggest leakage path on a good engine, and higher pressure will push them against the piston lands and make them seal better. Supposedly, aircraft engines are tested at 30 psig instead of 60 psig for that reason.

Here <a href="http://www.morrowmarsh.ca/concours/techpages/leakdowntester.htm" target="_blank">http://www.morrowmarsh.ca/concours/techpages/leakdowntester.htm</a>
is a site that tell you how to build a simple leak-down tester. Of course, a good one isn't really that expensive.

Earl Gillstrom

Yep, the tube between the gauges has a calibrated orifice.

The orifice is a .040 hole 1" long.

This came from an avation magazine on how to make your own.

John..

I made my own last night with a pressure regulator and a flow control valve (used as a choke or orifice).

I put 100 psig into each cylinder and left the flow control at the same setting across the board. I had to put the car into 5th gear to keep it from rolling. The flow control or orifice is VERY important because it allows the pressure to build up slowly by cutting the flowrate through the unit. Without the choke in the system the unit might be able to supply enough volume of air to show high pressure readings, even though the engine is leaking.

I got the following results:

Cyl 1: 90 psig, leak through rings
Cyl 2,4,5,6,7,8: 95 psig, leak through rings
Cyl 3: 95 psig, leak through intake valve....hmmm?

The ring leak makes perfect sense, since the air simply passes through the three gaps in the piston rings and into the crankcase. This reminds me of the story of a car that was sold becasue of poor running after a total rebuilt on the engine...the new owner tore it down to find the three gaps on the rings perfectly lined up....duh!

The bleed down is a super-value tool, and at $20.00 it was a no-brainer (I sourced some of the parts from surplus stuff here at work). Without it, I would not have known where the problems might be residing.

E-mail me if you would like a picture of the tester.

dr bob

</font><blockquote><font size="1" face="Verdana,Tahoma,Helvetica">quote:</font><hr /><font size="2" face="Verdana,Tahoma,Helvetica">Originally posted by Earl Gillstrom:
<strong>Yep, the tube between the gauges has a calibrated orifice.

The orifice is a .040 hole 1" long.

This came from an avation magazine on how to make your own.</strong></font><hr /></blockquote><font size="2" face="Verdana,Tahoma,Helvetica">Thanks, Earl -- That's the info I was looking for. In my collection of things I'm sure I have everything except a .040" hole an inch long. Shouldn't be too tough to make one of these gismos from a short piece of brass bar.

Cheers!

ViribusUnits

Something occures to me. There are alought of asumptions made in these units.

I'd expect bigger cylingers with bigger valves, and the like, to leak a larger volume of air, and smaller ones to leak less. Also, the leak rate isn't going to be linear. As the leak guage approches 0 psi, the volume of air it had to leak is going to get huge, while even a small volume of air leaked would result in a sigificant pressure drop at 100 psi.

It would seem to me that a wiser methoid would be to use a regulated air source, a valve, and a compression guage. Open the valve, let it pressure up, and then close the valve. Wait till it leaks down to a specified pressure, and record the time.

Then you can compare the times. The longer the time the better. If you get a big diffrence in the times, you know something is wrong. With experence, you'd know what a reasonable time is.

I've never preformed eighter test, but I'm curious why the "leak down" tester is used, rather than the above methoid.

Steve Cattaneo

Author Topic: Leak down testing
Tony
User

Rennlist Member

User # 414

posted 02-06-2003 13:21
--------------------------------------------------------------------------------
Taking this topic from another listing regarding piston ring gap clearance.

What is the general procedure for performing a leakdown test... considerations,steps..etc.

Is it outlined in the workshop manuals??

--------------------
Tony 87 $4
<a href="http://members.rennlist.com/v1uhoh" target="_blank">http://members.rennlist.com/v1uhoh</a>
<a href="http://www.p-caronline.com/directory/v1uhoh/" target="_blank">http://www.p-caronline.com/directory/v1uhoh/</a>

--------------------------------------------------------------------------------
Posts: 333 | From: Las Vegas BAAAAAABY | IP: Logged |

steve cattaneo
User

Rennlist Member

User # 478

posted 02-06-2003 21:32
--------------------------------------------------------------------------------
Tony,
A leak down test will pin point a host of problems. You are going to need a leak down gauge, and air compressor. Bring the piston being tested to TDC on the compression stroke this will ensures the valves are closed Install the gauge , pump air into the cylinder, the gauge instructions will tell how much air to pump. A 0 (zero) reading is a perfect cylinder ;( zero leak down) but we know nothing is perfect. Less than 10% is good, 20% is fair and 30% is poor. If the test indicates a leak of more than 20%, then listen for air leaking at different locations. crankcase leakage , PVC, oil dip stick = worn or broken rings worn or broken pistons. Air from the throttle body or air cleaner = leaking intake valves, air from the exhaust = leaking exhaust valves, bubbles from the radiator holding tank = head gasket, cracked head or block

--------------------
Steve C
The Great White928
85 S2 EURO
02 MB AMG 350HP SC

--------------------------------------------------------------------------------
Posts: 441 | From: new york | IP: Logged |

Earl Gillstrom
User

Rennlist Member

User # 8469

posted 02-07-2003 09:17
--------------------------------------------------------------------------------
Tony,
A little theory on the gauge: Actually there are two gauges, one on each side of a fixed orifice, (~.040", ~1" long. Compressed air is fed into the first gauge and then through the oriface to the second gauge and then into the cylinder. The leakage of the cylinder makes the second gauge read less pressure. The difference is the leakdown rate. FYI, My '88 (now Jim Morehouse's) had a leakdown of ~2%, at 90K miles which is considered excellent for a racing engine.

--------------------
Capt. Earl '91 GT Zermattsilver (factory special)/Classic Gray
'88 S4 Auto Parts car

--------------------------------------------------------------------------------
Posts: 40 | From: Glen Mills, PA | IP: Logged |

marc@DEVEK
User
User # 5789

posted 02-07-2003 21:42
--------------------------------------------------------------------------------
1-3% you will retain 99% power
4-8 you are down around 5% power
over 10%, your engine is lucky to make 80% rated power

Just my experience with leakdowns and dyno pulls!

Marc
devek
--------------------------------------------------------------------------------
Posts: 154 | From: California | IP: Logged |



All times are Eastern



<img border="0" alt="[cheers]" title="" src="graemlins/beerchug.gif" />

Vlocity

There is probably a very simple and obvious answer to this, but other than using a dowel or something in the cylinder and watching it come to TDC how are you guys finding TDC to do the test of each cylinder?

Thanks,

Ken

Randy V

Nice save from the grave.

SteveG

http://www.ebay.com/itm/300607774027...84.m1423.l2661

No affiliation, never used one, but this looks like just what you need. In fact, I was wondering if anyone could report on this?

WallyP

Nope, that won't do the job by itself. The whistle is used only to find the compression stroke, so that you can then find TDC-Firing.

In other words, finding TDC on the crank pulley might put you at TDC-Firing (at the end of the compression stroke) or at TDC-Exhaust, between the exhaust stroke and the intake stroke.

The wooden dowel thru the plug hole is the best way. You need to find TDC-Firing on #1 by using the whistle or by having the ignition rotor pointing at the plug wire to #1, or by watching the valve action.

Once you know that you are on TDC-Firing for #1, insert the dowel and mark it at the edge of the plug well. It might be a good idea to check by removing the dowel, turning the engine in the running direction (as if you were tightening the crank nut) almost one full turn then inserting the dowel and very carefully watching the dowel move out. The instant the dowel stops moving out is TDC. If your mark is perfect, all is well. If TDC found by the dowel doesn't match the TDC mark on the crank dampener, the dampener rubber ring has slipped.

Since you are now on TDC-Exhaust, you need to remove the dowel, turn the crank almost a full turn, reinsert the dowel, and locate TDC-Firing again. Remove the dowel and do you leakdown test on #1, writing everything down.

Remove the dowel, and insert it in #3, which is the next cylinder in the firing order, and turn the crank slowly ninety degrees. When the dowel indicates TDC, do the test on #3. Go thru the firing order, turning the crank ninety degrees each time.

blown 87

Everything you ever wanted to know about leak down testing is in AC41-13 1b chapter 8 section 1 on engines, starts at page 8-7

http://rgl.faa.gov/Regulatory_and_Guidance_Library/rgAdvisoryCircular.nsf/list/AC%2043.13-1B/$FILE/Chapter%2008.pdf

I prefer the single gauge percentage types myself, they are dead nuts simple once you understand that the engine should be hot and each cylinder has to be at TDC with both valves closed, and that it has to be held there. (I use the flywheel lock for this)

And remember, each gauge is different and are really only to be used as a general tool on engine health.
On a two gauge set, anything under 80/60 is a problem and on a single % type anything over ten % is not good.

Vlocity

Gentlemen,

Thanks for the response and link. That will add some confidence to my process. I am chasing a loss in blower pressure that has been gradual. It could be that the blower needs to be rebuilt as there iare no obvious leaks, but I want to rule the rings out. Compression test was good, leakdown should tell me a little more and then off comes the blower and intake.

Regards,

Ken