Boost leak testing methadology
 

I have a test fitting from turboboostleaktesters.com which I attach to the end of the intake elbow, after the IC. Since it's being attached after the Y-pipe the two pressure lines coming off the "tee" on the motor-side of the y-pipe must be blocked to prevent air from escaping. Are there any other places I have missed that should be blocked, or any other considerations?

With a compressor tank at about 60psi, I am only able to get the manifold to hold about 6-8psi. I can hear an air/leak noise that seems to be somewhere in the area of ACV down to LIM, but it's difficult to tell.

What is the "proper" way to test for leaks? It is normal for air to leak at the rate I describe above? It seems that while running this test air would travel through the intake runners and into the combustion chamber, and then be allowed to leak past and apex and side seals. Is that not the case?

Are you trying to test for a leak on the engine side or the turbo side? Maybe add some pics of how you have it attached?

The compressor side of the turbo will definitely lose pressure, I imagine that air will escape from the engine itself as well. Since the rotors are offset there's a damned good chance one of your rotors is positioned so that bothe the intake and exhaust ports have overlap.

If it were me I'd just get the car running and test each connection with starting fluid, old school style. That, or hit everything with soapy water and pulse the throttle for boost, looking for the bubbles.

Engine side.

The boost tester looks like this and is connected at the IC end of the OEM intake elbow, so is effectively pressurizing the TB forward.

https://cimg7.ibsrv.net/gimg/www.rx7...268c86610e.jpg


That makes sense about overlap allowing air to pass straight through. Maybe this is a silly question, but is there even any overlap on a stock ports?

However it seems that if the motor is rotated in a way that there is overlap, the thing probably wouldn't even hold the 6-8psi I'm seeing.

I am curious, though, if the motor is not rotated such to create overlap, would you expect the seals to be able to hold that pressurized air? I mean, those seals aren't exactly o-rings but metal-on-metal...

Gotcha! I read "y-pipe" above and was thinking the twins, with your plug being attached to the charge piping post-intercooler to find leaks in the IC/charge piping couplers.

I can't speak with 100% certainty, however I believe there would be due to the rotors being offset 180°. When one rotor encloses the intake port into a compression phase, then the other rotor would likely be positioned in a manner that air could pass from intake to exhaust ports by way of the cutout on the rotor face.

I mean, that's only 10% pressure retention of your 60psi total, right?

:dunno: Your guess is as good as mine on whether they'll hold static pressure or not. Do you have your TB butterflies opened as well?

I'm curious why you're checking the TB/UIM/LIM for leaks? If you're concerned why not just replace the gaskets and torque to spec?

The best way I've found is to look, listen, and feel for leaks. It can take some patience. Also you will NEVER get an engine to where you can pump it up with pressure and it holds - some will always leak through the engine itself.

Those boost testers are nice, but I typically find some way to block the intake - take a thick plastic bag, fold it over a few times, then clamp over the intake. Or get a PVC plug. Then, I use one of the air blower things on my air compressor you use to blow dirt off stuff and hook that into a vacuum line. Crank the PSI down on the compressor, pull the trigger to pressurize, and see what you find.

There are also ways to build smoke testers to find small leaks. They can be tricky as well, though - I've not used a professional tool, however.

A lot of boost/vacuum leaks are avoided with discipline. Make sure gasket surfaces are clean and true, gaskets are good, all vacuum lines are hooked up, hose clamps are properly tightened, etc.

You can also find leaks between the throttle body and engine with starter fluid. With the engine running, carefully spray the fluid in spots where. there could be a leak. The engine will rev up/smooth out when it sucks in the fluid.

Dale

I'm not saying this is the right way, but its the way I did it and was able to verify/find a few leaking gaskets/couplers, as well as a large leak from the exhaust manifold.

I disconnected the air intake, and connected the pressure tester cap like you have to the secondary turbo intake. I then blocked off the primary turbo intake elbow to prevent air from escaping that way.

As someone else mentioned, air would still escape through the exhaust, so I plugged both of the exits on the muffler.

At that point even with my my inadequate compressor (5 gallon + 5 gallon external tank) I was able to keep it pressurized enough to find leaks. Normally for a minute or so before I needed to refill the air tanks and try again. I imagine a real compressor wouldn't have any trouble maintaining a good pressure for much longer.

You'll never get it air tight so its just a matter of making sure you can hold a good pressure to listen for leaks without the compressor engine kicking on and creating too much background noise.

I used these little rubber ball things that you inflate with air to block things off. I think they are meant for plumbing. Use common sense and don't put yourself in front of anything blocking the air pressure incase it decides to shoot off. If you use a similar tool to block any ports, I'd also make sure you have a good air pressure regulator so you don't overfill the plugs.

Don't focus on the difference between the supply air pressure and the amount of boost the engine is holding according to the gauge. Just focus on finding leaks. Take an empty spray bottle and fill it with slightly soapy water. Spray around the engine bay and use it to locate leaks.

So I definitely have a leak somewhere. Was doing some logging yesterday and slowly turning the boost up on my new BNRs, and got to the point where pre-transition would put out about 14.5psi, but then lose a ton of pressure at transition and only rebuild to about 10psi by redline. I also noticed that as testing continued, overall pressure levels decreased between pulls, as if a hose somewhere was slipping farther off or something -- in the beginning I was able to get about 18psi on primary and 14psi after transition, and only 14.5/10 by the end of the drive (pictured below). I had target boost in the PFC set to 1.5kg/cm2 (~21psi) by the end of the session.

https://cimg1.ibsrv.net/gimg/www.rx7...2c96e514d3.png

I LOVE the idea of blocking the exhaust and pressurizing the whole damn thing. Looks like those smoke tester tools are starting to become affordable, too -- I wonder if that's a good candidate for "should have just bought one of those five years ago".

Kinda bummed, I just replaced every damn solenoid, new OEM primary injectors, all the vacuum lines with new 3.5mm silicon, and meticulously zip-tied EVERY DAMN CONNECTION with PTFE zip ties, but then was under a time crunch putting the rats nest and UIM back on so must have left something disconnected during the rush. Balls. Oh well, not the end of the world. Will post back with my impressions of blocking the exhaust as @SpinningDorito suggested.

Yep just be mindful of the pressure you're blocking off and don't potato cannon yourself by accident.

Any leak that large should become super obvious if you can get pressure in there. I doubt its one of the small rats nest lines. Could just be a gasket not sealing great, or a coupler with a leak, or maybe the flapper in your y pipe sticking, CRV issue, etc. Hopefully something easy to get at.

Remember, most boost leaks/vacuum leaks aren't from a vacuum line, it's typically an intercooler coupler, a block off plate, or a gasket between 2 mating surfaces. That's where you need to focus.

Dale

A little update. I hooked up a boost tester with my measly 2gal compressor.

I blocked off the primary turbo inlet accordion, blocked off the tailpipe, and applied pressure to the secondary turbo inlet.

The compressor is set to regulate output at 100psi. The regulator on the tester kit is set to bleed at about 30psi. You can see in the video below that when I open the valve, the system will hold 30psi before the regulator starts bleeding air off (seems like it could probably hold more, but seeing those accordion pipe expand I didn't want to risk blasting my nuts off with one of those caps flying off). When the value is closed the pressure quickly drops to below 5psi, and hangs around 1 or 2psi for a while while slowly bleeding off.

I feel like for the pressure to drop that quickly, there is probably a leak big enough to be worth hunting for and fixing, but also I can't imagine that if a tiny 4-6mm-ish air hose from the compressor can keep 30psi in there, that two spinning turbos could not keep 14 in there.

I tested the CRV externally with a mitivac and found that it starts to leak around 5psi. Some searching turned up that apparently this is normal. In the video I have the CRV hose plugged, and the plug actually pops off at the end of the video. Doing the same test with the CRV unplugged does not yield any different results in total pressure, although there is a good amount of air coming out of the end of the hose.


I get that I cannot expect the system to hold pressure indefinitely, and that hunting down leaks is key, but what I'm hoping to get out of sharing this is an opinion about weather my system appears to be holding pressure "good enough" or if there is clearly something wrong.

Why not smoke test it? Quick and easy way to identify where leaks are actually coming from.

Smoke test kit is at my folk's place! It came with a 1psi regulator, and the one and only time I've used it so far it seemed to hold the 1psi... So I'm going back soon with an adjustable regulator to turn up the pressure and see where it's leaking. I've exhausted the "hunt and listen" + soapy water methods already.

You didn't actually feel around for the leak though did you? Or spray slightly soapy water in the engine bay?

Never had good luck with soapy water for some reason. This weekend I’ll hook up and smoke machine and use a stethoscope.

I take it you saw the video and tend to agree that the motor should hold pressure for longer than it does after the valve is closed?

I do notice that some smoke likes to creep out of the throttle body linkage area. Again, soapy water didn’t help that case and a mechanic friend who witnessed it said it was common and to not worry about it.

I’m hoping using the smoke machine with > 1psi will be helpful.

if you never had good luck with soapy water, then you probably don't have a boost leak (or at least, not many). If you can't feel air shooting around the engine bay from leaks, then you may be barking up the wrong tree.