DIY leak down/compression tests

[gulkster]

I'm a total DIYer (as many here are of course). I want to check the condition of my 3.0l, and would like to do these (leak down if necessary after compression). I have a home made leak down tester (haven't used yet) made from welding a male hose end to an old spark plug (saw it on speed channel). I guess I could use it a compression test too, as long as I have a gauge that plugs in? Any suggestions on what I should buy/make so I can do this myself, AND do a good job would be greatly appreciated. TIA.
Jamie
edit: did a search and didn't come up with anything about DIY stuff I am looking for

[docrodg]

It should already have a gauge on it (about 150psi) if it is for doing a leakdown test, just have to get one that it is high enough reading to add for doing compression. Grainger can be a good source for pressure gauges - get one that reads about 300-400 or so to make sure you do not overpressure it when running compression. Also add a one-way valve just behind the fitting so the pressure stays in the gauge. The leakdown tester you made should have a shutoff on it to stop flow of air, just turn off the valve and hook up the tester with the higher reading gauge, viola! compression tester. When done with a cylinder open the valve to bleed off the air.

Leakdown testing is far more informative of problems than compression testing.

[morrisjm1]

Check this out, I don't think you could build one for a whole lot less than $60.00 and this is a real one.

Mike


http://www.aircraft-tool.com/shop/detail.aspx?PRODUCT_ID=2E-14&ReturnPage=/shop/search_result.aspx?PageNo=5

[dad911]

Leakdown tester is 2 pressure gauges with an orifice in between, and a regulator. Extensive thread here:leak down test

[euro911sc]

I bought my longacre leak down unit off ebay for like $50 shipped... not worth the time to make your own... IMHO and I'm a serious DIYer like your self.

-Michael

[T77911S]

Originally Posted by morrisjm1

Check this out, I don't think you could build one for a whole lot less than $60.00 and this is a real one.

Mike


http://www.aircraft-tool.com/shop/detail.aspx?PRODUCT_ID=2E-14&ReturnPage=/shop/search_result.aspx?PageNo=5

i dont think that one will work. the part that screws into the Splug hole is too short. most have a longer hose attached that allows you to screw it in. looks like a good deal otherwise.

[gulkster]

Thanks for all the input. It looks like ti is worth it to just get one of those "pre-made" ones I will definitely need one with a long hose. I started to do a compression test last night to see if a leak down was needed, and only managed to get two cylinders done before it dark It took forever to try and thread the gauge into the head, it has to be 3.5-4 inches down there...and at an angle. Cylinder 4 was 125psi after 4 bumps, 140psi max (after 6 bumps). Cylinder 5 was also 125psi after 4, and was 145psi max after 6. So far so good. I will do #6 tonight as I only have a bit of time. I love this car b/c working on it is a bit of a conundrum: manuals say to do the compression test when the engine is at normal operating temp, but it also says don't remove the spark plugs until it has cooled down!

[gulkster]

dbl post

[BK911]

Personally I don't even bother with compression tests anymore. A leakdown is easier and better. I haven't noticed much difference in cold / hot engine either, and would much rather work on a cold engine. Plus a leakdown can be done with the engine out of the car. Ok, rant off.

Here is a leakdown tester I made for about $15. All you need is a regulator, pressure gauge, old spark plug and some tubing. Wham bam.

[dad911]

BK, how do you do it with 1 gauge? and no orifice? Pls see my other post.

[BK911]

I didn't have time to read the entire other post, but basically...

The compressor regulator is set at about 75psi.
The leakdown regulator is set at 50psi.
Screw the hose fitting into the plug hole.
Attach the other end of the hose to the regulator.
Subtract the new pressure reading at the leakdown regulator from 50psi then double.

For example, when the fitting is plugged into the 50psi regulator, the pressure immediately drops to 45psi. That's a 10% leakdown.

That's the way I do it. Please tell me if this is wrong!

[T77911S]

i usually do it cold. trying to be fast enough to do it hot is tough. do you make corrections to the first one done and the last one done to allow for temp change?

i dont put much value in a leak down as far as a % number. it is a great tooll to locate the source of low compression, but i would always do a comp test first, if the conditions allowed. i bought my LD because i bought a used 3.0 and i wanted to try to test the motor before i tore it down.
maybe i just dont understand the procedure, but when i was trying a LD yesterday on my truck, i could set the regulator so the % was always 0 or 80 or 70%. BTW, my truck was 145 psi for all 4.
i would think for the LD to be acurate, the incoming air pressure would have to be set for max reading on the piston side gage while connected to something that has no leaks like a plug or a plug with a .040 hole in it like the one between the 2 gages. in other words, some way to calibrate it for a known % of leakage. i may get it back out this afternoon and mess with it some more.
my harbour freight gages reads 15psi when the other gage shows 0% LD. the left gage goes to 100psi, i geuss the right one only goes to 15psi?

[gulkster]

Originally Posted by BK911

I didn't have time to read the entire other post, but basically...

The compressor regulator is set at about 75psi.
The leakdown regulator is set at 50psi.
Screw the hose fitting into the plug hole.
Attach the other end of the hose to the regulator.
Subtract the new pressure reading at the leakdown regulator from 50psi then double.

For example, when the fitting is plugged into the 50psi regulator, the pressure immediately drops to 45psi. That's a 10% leakdown.

That's the way I do it. Please tell me if this is wrong!

Unless I completely misunderstand how the damper (0.040 hole) works, which is entirely possible, the above method makes sense. I don't understand why you would need to double the (edit) difference number though. Can someone straighten me out please.

[dad911]

BK - I get it.

gulkster - BK doubles the scale, because he is using 50psi, not 100 psi, and you want 100% full scale. Multiplying x2 is short-cut, same as taking the percentage, 45/50 x 100.

[gulkster]

Hahaha...I was so focused on the physics, I TOTALLY missed that That is the funniest thing...
And thank you.

[911pcars]

Originally Posted by morrisjm1

Check this out, I don't think you could build one for a whole lot less than $60.00 and this is a real one.

Mike

http://www.aircraft-tool.com/shop/detail.aspx?PRODUCT_ID=2E-14&ReturnPage=/shop/search_result.aspx?PageNo=5

"i dont think that one will work. the part that screws into the Splug hole is too short. most have a longer hose attached that allows you to screw it in. looks like a good deal otherwise."

Seems like a simple matter to insert a short length of NPT-nipple to lengthen the spark plug adapter. Of course, there may be other brands with a longer hose attached to the adapter. Snap-On has one (see below), but not for $60.



Sorry about the giant size.

Sherwood

[Jim Williams]

FWIW, here is the leakdown tester I built several years back from parts lying around in the garage. I did have to buy a second 100 PSI gauge. I installed a smallish restriction between gauge 1 and gauge 2 (don't recall the size). I borrowed a commercial tester, I think a Tavia, to compare readings with. To make a long story short, there was a significant difference in the readings, which I attributed to the orifice size, so I measured the one in the Tavia, and made mine like it. It's been too long back, and I don't remember the actual orifice size I ended up with.

I had searched every source I could find and like others have found, there does not appear to be a standard orifice size.

I have never found the hose supplied with most leakdown testers to be satisfactory for use on the 911 engine. They don't result in a consistent seal, and sometimes when tightened, the connector on the the sparkplug end will twist on the hose, leaving a problem to get the hose out of the head. I made up a a solid connector with an air hose fitting on one end and a spark plug fitting on the other. A deep well socket is used to tighten this connector in the spark plug hole, and the right angle connector on the leakdown assembly makes an easy hookup to connect the gauges.

[dshepp806]

i've always been told (?) to do both compression and leakdowns, as they lend different data.

True?

Doyle

[T77911S]

Originally Posted by BK911

I didn't have time to read the entire other post, but basically...

The compressor regulator is set at about 75psi.
The leakdown regulator is set at 50psi.
Screw the hose fitting into the plug hole.
Attach the other end of the hose to the regulator.
Subtract the new pressure reading at the leakdown regulator from 50psi then double.

For example, when the fitting is plugged into the 50psi regulator, the pressure immediately drops to 45psi. That's a 10% leakdown.

That's the way I do it. Please tell me if this is wrong!

now i see how that works. i need to look at mine again. yours has the female chuck on the regulator, i cant remember if the female chuck on mine is on the hose that screws into the head. if the male part is on the block with the 2 gages and the reg., i cant set the air pressure before i connect it to the hose that goes to the head. duh,
if you connect the regulator to something with a .040 orifice in it, then set the reg to 50 psi, removed the .040 orifice and then connected it to the hose to the head, would that give you a more acurate reading?

[gulkster]

I realize this is a bit off topic from my original post, but its still within sight Can anyone give me some tips/explanation on how to get the socket on to the #1 spark plug? For some reason I cannot get the socket on. The other five are fine, so I'm not really sure what is going on down that hole. It feels like the socket just can't grab the plug. Do I need a really thin walled one due to the odd angle maybe?
TIA.
Jamie