Boost leak test - Leaking throttle shaft seal
 

Hey guys, got the refreshed VVT motor built and running in the car, and I decided it was time to finally do a boost leak test on my system and found 2 leaks. My BOV to flange connection was leaking (there was supposed to be an o ring in there :eek4:) and the throttle body is leaking at the shaft seal at 15-18PSI.

I can fix the BOV leak simply, but I really don't know what to do about the TB leaking. It's at the bearing on the throttle cable side. Has anyone found a solution to sealing up that part of the TB? Yes... It's an ebay skunk ripoff, just hoping I don't have to replace it. I also haven't seen anything that indicates that the real skunk 2 would hold pressure any better, as there are many posts on Evo forums about them leaking and fighting for returns.

I found a couple old threads on MT of people with similar complaints but no responses after the leak had been isolated to this area.

https://cimg5.ibsrv.net/gimg/www.mia...7a6f13386c.png

Personally, I do not use eBay items for quality control reasons. Sometimes you just get lucky and it works, sometimes you don’t get lucky though... in my opinion I would’ve kept the stock throttle body, but if you’re dead set on using an aftermarket one it’s pretty much deal with the boost leak or get a genuine skunk TB. If you’re near NC I believe Daniel Semon has a TB and manifold from skunk for sale. If you cover shipping he may be interested in selling.

I'm not sure if you read my post or not...

"I also haven't seen anything that indicates that the real skunk 2 would hold pressure any better, as there are many posts on Evo forums about them leaking and fighting for returns."

I have also found posts of the stock TB leaking. I am not running a stock TB as I have had 2 stock TB shaft failures in the last 2 years, and this TB has a much more robust shaft. Recommending that people keep using a stock TB when they are well documented to fail on a frequent basis is not sound advice.
I get that this is an ebay part, but I'm not going to go by a junk2 just to have the throttle plate stick and have it (potentially) leak at the same pressure. I'm mainly hoping to see if someone has a solution, or that they have pressure tested their skunk 2 and confirmed that it does not leak.

I get that skunk have their own problems, but I respectfully disagree that an eBay part has better quality control than a well known and reputable company. As for the stock TB, that’s news to me but I’ve not had any issues since I bought the car 2-3 years ago, nor the miata I owned before that. Not sure if it’s leaking boost though.

I’d take it apart, look for an O ring that you might be able to add, replace, upsize, or grease up on the shaft. Not sure how else it would seal and still be able to rotate. By the sounds of it, it’s not a severe boost leak so I might just forget about it if I were you.

To get this thread back on track - Does anyone have any actual experience or data on how much pressure their stock or skunk TB is holding during a boost leak test?

Why do you really care? I mean the engine is consuming massive amounts of air from a compressor that has dynamic control over the pressure ratio. Can a leak there cause any measurable change in the compressor inlet mass airflow?

Please let me know if there's something I'm missing.

If your target is 20PSI and your system leaks at 16, that is a problem. Even if your target is 16 and your intake tract can barely hold 16PSI, then everything is having to work harder to reach that same pressure because you're leaking air the whole time you're at target boost.

I have my supercharger spinning to its redline, and my old 1.6 setup capped out at 16PSI, and it started seeping air out at 12-14lbs - I could have been putting more air into the engine instead of wasting it. I really don't understand why wanting a system to hold its working pressure would be weird. I would personally want my system to hold 5PSI over my target boost to have some room for connections degrading due to vibration, heatcycles, etc

edit: I was legitimately asking to be corrected here. I understand why I was wrong now

its not weird to want it, it is just that by the sounds of it it’s a small boost leak that might not even make a difference. The effort it takes to fix it might not be worth it to most. With it being an eBay part, I don’t think you’ll find many other people here running it to be able to help from first hand experience. It really just leaves disassembling it to try to find a way to fix it.

It is not a small leak, it is literally reducing the amount of boost I can run. The system loses pressure very quickly after I disconnect the compressor instead of slowly bleeding off like it's supposed to. Please stop 'offering advice' on topics you do not know about.

anyways...

I'll have some time tonight and I'll rig a test on a stock TB to see what kind of results I get vs the ebay TB, was just hoping someone here had already run some tests to compare with.

I believe all throttle shafts leak a tiny amount under boost.
One of the major failures of a blow through carb set up is that they leak both air and fuel above 15 psi.
Carbs in question were PHH Mikunis. These were the best side draft carbs you could buy.
Serious attempts to get these to work failed miserably...
Get it to seal and you had throttle stick problems, eliminate sticky throttles and you got leaks.
Machining o-rings into throttle shaft housings did not work, adding umbrella seals did not work.
2 sets of 3 side draft carbs got trashed in the attempt.
Carbs in question No Reserve: Triple Mikuni 44PHH Carburetors for Datsun for sale on BaT Auctions - sold for $6,666 on February 14, 2022 (Lot #65,759) | Bring a Trailer
These puppies have gotten stupid expensive now

Limit blow throughs to 14 psi OR put entire carb in the plenum (which has its own set of weird ass issues).

It's a bunch easier when there isn't fuel in the airflow.
You just live with the tiny leaks.
It is interesting that the PSI where problems seem to occur has stayed the same over time...

Didn't see the OPs last response before my post.
Testing another TB in an excellent idea.
If the factory unit does better maybe try a real Skunk2 and see if it's better as well.
Please let us know what you find.

THANK YOU

I'm still gonna test and fiddle with some things, but that's good to know. I've been seeing stuff about evo guys literally putting 30PSI into an intake and having it only drop a few PSI a second, but that was a car engineered around boost. My concern is those same people having bad leaks after they upgraded to real skunk 2s since from what I can tell skunk does not use a seal, only a bearing, and the OE evo TB has a specialty set of replaceable shaft seals.

I don't have a real skunk to test, but I can at least do an OE NB throttle body. Also gonna see if I can find any DSM/evo/etc throttle seals that might have a compatible diameter.

I'm curious if any of the crazy high psi guys here have done a boost leak test at their target pressure, or if it's a 'turn the turbo up until you shove more air in than can leak out' situation. I wish I could do the same, but I'll have to switch superchargers to get past where I'm at now.

The piece that you are missing is that during operation you have 8 massive boost leaks venting hundreds of cfm from the intake track. Not holding static pressure doesn't tell you much about how much compressor inlet flow is being lost through the leak. Now you are correct that the SC doesn't typically have the advantage of compensating for losses via dynamically increasing controlling the PR, but it is still moving massive amounts of air. Consider the compressor you probably used to run this test. If it is your run of the mill 110V unit it might flow what 3-5cfm at 50psi?

Google can provide all sorts of engineering references for calculating flow through an orafice with a pressure differential. 5 seconds of search found this example:

Calculator

Edit: did a quick test calculation... ignoring volumetric efficiency, a 1.6l engine will consume ~300cfm at 5krpm, 16psi boost at sea level. A 1.5mm diameter hole with a pressure differential of 16psi will vent ~1.1cfm (assuming the calculator is right). That means approximately 0.3% of the air entering the SC is exiting through that hole. Think about the gap between the shaft and the housing, the area of that gap is going to be small, probably smaller than a hole with diameter 1.5mm.

Hell yeah! This is the kind of stuff I love to learn. I understand that it's a 5 second search for you, but I wouldn't have even known how to ask the right question to find that out, so thanks for posting. As much as I love building stuff I have no engineering background.

So I did get to do some A-B testing last night. The stock TB held 1-2 PSI more pressure in my test setup (small compressor hooked up at the SC inlet), but then you start getting small leaks at injector o rings, etc and it seems like a bit of a crapshoot. That being said, the stock TB shaft did leak, albeit less than the ebay Skunk.

Based on Ted75zcar's info I'll be running the fake skunk (tack welded new throttle plate screws in) for the shaft reliability.

I did take some measurements in case this is useful for anyone else. This edition of the ebay skunk has a fully round shaft and has a pressed in bushing on the throttle cable side. I was not able to easily remove the bushing so I left it installed. When the throttle cam is clamped down there is 1-2mm between the face of this bushing and the cam, so it may be possible to add a seal in there and anchor it to the bushing as it appears there is no leak between the bushing and the throttle housing, just between the bushing and rotating shaft. That being said I'm leaving this all as is and running it.

https://cimg9.ibsrv.net/gimg/www.mia...c150c9fa41.png

Throttle shaft is 8mm and the OD of the bushing nose should be 18mm as I could not get the calipers on quite square with everything installed. Also, yes.... that's an Ace worm gear clamp. Yes it's being replaced.
https://cimg1.ibsrv.net/gimg/www.mia...32211e0c8d.png
https://cimg4.ibsrv.net/gimg/www.mia...dbf693175d.png