Intake Manifold Leakage

[billpedalino]

I performed the manifold seal pressure test, as per Tom Cotton (I love this test!) on my '64 XLCH and I noted that there is minor leakage at about 15 psi. Now I know that when performing the test on big twin manifold seals (Peek or brass) there should be no leakage at all and I did achieve this on my '47 FL. But the later XL's and FL's the O-Ring sealing mechanism works a little differently.

Unlike on the big twins, as the negitive pressure in the manifold increases and approaches -14.7 psi, the O-Ring is drawn tighter into it's groove, presumably affording a tighter seal. However when performing Cotton's pressure test the O-Ring is increasingly pushed away from the groove as internal manifold pressure increases, thus indicating possible leakage when that might not be the case under operating conditions.

The above is not intended to discount the importance and validity of Cotton's test - I believe its an essential step which I include every time I build a motor or install a top end. In this case it tells me that the assembly is fairly close to leak-free and therefore a constructive use of time. But I do wonder if a minor amount of leakage in O-Ring setups can, or should be tolerated.

Has anyone had similar results?

[T. Cotten]

Bill!

Your intake valves would start to open before the O-rings would move, if properly retained in their grooves by the clamps.

Why would Sports be different from Shovels?

The most common problem with Shovel and Sportster manifolds is aftermarket stainless clamps that are too wide, probably metric.
They should be ground down to .750" wide.
Please remember that the function of the clamps is only to compress the O-rings to fill the gap, not to hold the carb on to the machine. Bubbles can warn of over-tightening.

No leak is an acceptible leak.

....Cotten

[cdndewey]

Unlike on the big twins, as the negitive pressure in the manifold increases and approaches -14.7 psi, the O-Ring is drawn tighter into it's groove, presumably affording a tighter seal. However when performing Cotton's pressure test the O-Ring is increasingly pushed away from the groove as internal manifold pressure increases, thus indicating possible leakage when that might not be the case under operating conditions.

I'm a little confused here...what exactly are you doing to achieve near a perfect vacuum and what are you using to measure it? I would think at that much vacuum the o-rings would have been sucked right in.

[T. Cotten]

"Perfect" vacuum isn't achieveable, Dewey!

But we can produce the same differential with pressure, as one atmosphere is just shy of 15psi.
So 15psi is overkill, but it will display, with bubbles, any and all possible places for a vacuum leak.

Even though the motor will never draw extreme vacuum, a tiny leak can still cause grief, or at very least a loss of performance.

....Cotten

[billpedalino]

Cotton,

You're correct - Sportsters would be no different that shovelheads.

I am curious about the intake valves opening though, and I was going to post the question but you beat me to it. I took the spark plugs out during the test and noted leakage into the heads (obviously with both pushrods loose). I'm surprised that a loading of only 14.7 psi would cause a load that would leak past the valves. If you do the calculation using a 1-3/4" (XLR) intake valve and a full 14.7 psi pressure loading, you only develop a total load (force) of about 35 lbs on the back side of the valve. Using a 1-5/8" intake valve, its only about 30.5 lbs force. I wouldn't have thought that this force would crack open the valve with both springs installed, but apparently it does. I also noticed this leakage past the intake valves when pressure testing my 47 FL.

Concerning the Sportster manifold, I'm not surprised about this leak and am shocked that it's relatively minimal. This particular combination of cylinder and head casting geometry is absolutely terrible. I've assembled many Sportster motors over the past 47 years and this one is the worst match-up I've ever seen! No matter how much you try to turn the cylinders and heads before tightening you cannot achieve anything close to a square match-up at the manifold. Also, my experience seems to indicate that this problem probably derives from inconsistent machining of these castings, as the manifolds all seem to be very close. In fact, in this case changing manifolds yielded no difference in this poor fit.

I had a similar problem with my stroker Shovelhead (although not as bad a fit) and finally used high temperature silicon on the O-rings which worked out very well and has held for many years. It looks like I may be forced into this solution with the Sportster as well.

Finally, to address your clamp observation, I used stock manifold clamps as this is a restoration which will be judged, so that isn't the culprit. I never use the stainless steel aircraft clamps because of the poor fit and over-kill tightening that you cited.


As always, thanks to all for your input.

Bill!

Your intake valves would start to open before the O-rings would move, if properly retained in their grooves by the clamps.

Why would Sports be different from Shovels?

The most common problem with Shovel and Sportster manifolds is aftermarket stainless clamps that are too wide, probably metric.
They should be ground down to .750" wide.
Please remember that the function of the clamps is only to compress the O-rings to fill the gap, not to hold the carb on to the machine. Bubbles can warn of over-tightening.

No leak is an acceptible leak.

....Cotten

[T. Cotten]

Bill!

Naturally the valves crack open slightly, even at lower pressures.
This is why it is so critical to have a constant, regulated air supply. With a regulator, the differential is shown on the gauge, and when you turn the motor through, it becomes apparent when the valves are most closed. Then the regulator can be adjusted to make up for any leakage, and loosening the pushrods is unnecessary.

(Most folks who get "false negatives" think hand or air mattress pumps will do. Some have even thought a shop vac blower would work, when it won't even come close.)

My experience with Sportsters was quite limited, as their owners usually presented far more annoying problems than the hardware.
But just like Shovels, I found that the manifold itself could never be used to line up the heads, without guestimating the necessary gap between the manifold and head spigots. So I had a sheet of stainless sheared to 60º for a gauge. (A Sportster owner borrowed it, and I haven't seen it since....)

I have yet to find any "silicones" that survive my local fuel for long, unless you consider Permatex's "The Right Stuff" to be a silicone.
It may be that your stroker's success was due to the nature of common O-rings swelling from modern fuels (attached), and your adhesive gave them time to seal.
I wouldn't trust any common O-rings for more than a few seasons, but you may not get viton to seal in the first place, and when the goober gives it up, they won't have swollen to fill the gap...

....Cotten

[billpedalino]

The 60 degree gauge is a good idea - at least it can be used to make the gaps equal on both sides and equalize the clamping force and seal areas on both o-rings. But in this case, the fit-up and misalignment is too weird. S&S used to make oversized manifolds for their longer cylinders, which might help, but I cant use one because its a restoration. Looks like 'The Right Stuff' is an option.

Bill!

Naturally the valves crack open slightly, even at lower pressures.
This is why it is so critical to have a constant, regulated air supply. With a regulator, the differential is shown on the gauge, and when you turn the motor through, it becomes apparent when the valves are most closed. Then the regulator can be adjusted to make up for any leakage, and loosening the pushrods is unnecessary.

(Most folks who get "false negatives" think hand or air mattress pumps will do. Some have even thought a shop vac blower would work, when it won't even come close.)

My experience with Sportsters was quite limited, as their owners usually presented far more annoying problems than the hardware.
But just like Shovels, I found that the manifold itself could never be used to line up the heads, without guestimating the necessary gap between the manifold and head spigots. So I had a sheet of stainless sheared to 60º for a gauge. (A Sportster owner borrowed it, and I haven't seen it since....)

I have yet to find any "silicones" that survive my local fuel for long, unless you consider Permatex's "The Right Stuff" to be a silicone.
It may be that your stroker's success was due to the nature of common O-rings swelling from modern fuels (attached), and your adhesive gave them time to seal.
I wouldn't trust any common O-rings for more than a few seasons, but you may not get viton to seal in the first place, and when the goober gives it up, they won't have swollen to fill the gap...

....Cotten

[billpedalino]

Issue apparently solved - at least for now. As per Cotton's suggestion and per my past experience, AND... after several additional installation cycles, the manifold/head joints are now leak-free. I used a very light coating of Permatex Right Stuff on the O-rings and let the installation sit for 12 hours after each attempt. The assembly now withstands 15 psi of pressure for an extended time with no leak, using snoop leak detection fluid - even with the lousy manifold match-up angle.

This is bittersweet. On one hand (and most important) I very pleased that it is now leak free using the OEM components. On the other hand, I just spent $$ on a stroker manifold which I can't use as I already have on on my stroker shovelhead.

Finally, as an old Harley mechanic and now professional Engineer, I'm so dismayed about the poor tolerances, excessive tolerance stack-up, poor quality control, and inconsistent finished product that the factory used to put out and that we had to deal with back then. It would not have been that difficult to make these manifolds fit perfectly; well within the range of adjustment of the cylinders and heads - as many did. But working back then at a shop that did MANY motors each year, it was hit-and-miss. This was simply sloppy, even with the technology of the 1950's. This was a long-standing problem with more than a few Sportsters and Shovelheads and it was ignored by the factory for all those years. I even remember that during a visit by the factory rep. to the dealership that I worked at in 1970, we complained about this problem and were told to wrap the installed O-ring with 1-1/4 turns of Teflon tape before installing the clamp. Really?!? There was no excuse for it taking all those years to design a better (rubber band) setup.

I feel better, now that I've vented....

[T. Cotten]

Its even more dismaying, Bill,..

That a stylus was probably first put to papyrus about bubble-testing around 100AD by Heron of Alexandria. T'aint rocket surgery.

Even H-D mentioned it, reprinted on pg. 93 in the "Panhead Service Manual 1948-1957 Rigid" that is commonly available, although for oiling issues.
Did they just want their own servicepeople, and everyone else, to suffer in ignorance?

Vintage motorcycling has enjoyed a shot-in-the-arm thanks to the internet spreading the word. More machines are running with happier owners than ever before.
It shouldn't have taken this long, and a lot more machines should have survived un-molested.

....Cotten
PS: "Rubber-band" set-ups only added headaches, as modern fuels became more and more digestive.
PPS: My pinhole camera-obscura worked much better for the '79 eclipse. Diffusion from the clouds I'spose.
Good thing I've got my welder's helmet....

[billpedalino]

Well said...

And the historical reference is quite interesting!

[cdndewey]

"Perfect" vacuum isn't achieveable, Dewey!

But we can produce the same differential with pressure, as one atmosphere is just shy of 15psi.
So 15psi is overkill, but it will display, with bubbles, any and all possible places for a vacuum leak.

Even though the motor will never draw extreme vacuum, a tiny leak can still cause grief, or at very least a loss of performance.

....Cotten

Aye...and what I was getting at is this statement..."as the negitive pressure in the manifold increases and approaches -14.7 psi".

Since it says -14.7 psi...which is gauge pressure...that would be near 0 psi absolute...which like you say isn't achievable.

Like you say pressurizing to 15 psi is overkill as a manifold is highly unlikely to see 15 psid. I know the importance of manifold leaks as I had a piston fry on my Sportster 35 plus years ago because of an air leak.

[T. Cotten]

Once upon a time....

I had built had built a '39 OHV for an associate, and had become quite quite familiar with it, allowed to ride it often.
It was subsequently sold to a fellow who would return it to me for spring maintenance.

One year, he said it "ran fine", but it just didn't seem to me to jump like the scalded dog I remembered.

So a bubble-test showed just a little blister of tiny foam bubbles upon one inlet nipple rivet that took fifteen seconds or so to form.
I sealed it with my 'secret sauce', retuned, and the owner announced that it ran "like a brand new motorcycle!"
This leak only robbed performance, but many burnt motors brought to me "ran fine".

No two leaks behave the same, except those that prevent the machine from running at all.

....Cotten

[Pete539]

I have a question for Mr. Cotton.You stated in one of your posts that you might not get Viton O-rings to seal.Would you please expound on your theory has to why Viton O-rings might not seal.Viton FKM series O-rings are good for temperatures from approx. -40F to consistent temps.of 425F.They are as fuel proof as any seal produced and are a standard in the aviation industry.Presently I am fighting an intake leak on my freshly refurbished CH and have found the typical aftermarket O-rings to swell upon use indicating their instability when exposed to heat and fuels.Please reply to this question as I am extremely interested producing the best possible assembly with out the use of additional sealing compounds or other home brew methods.

[T. Cotten]

I'm confused Pete!

I have always endorsed JAMES viton O-rings wholeheartedly.
(And yes, common O-rings are a lit fuse.)

....Cotten
PS: The search function brings up nothing, so please link me to the discussion so I can clarify it.
(If it was my previous post to this thread, I was referring to the stroker geometry problem.)

[Pete539]

Viton O-Rings

I'm confused Pete!

I have always endorsed JAMES viton O-rings wholeheartedly.
(And yes, common O-rings are a lit fuse.)

....Cotten
PS: The search function brings up nothing, so please link me to the discussion so I can clarify it.
(If it was my previous post to this thread, I was referring to the stroker geometry problem.)

T.Cotton,I'm sure my pondering came from an out of context interpretation of a posted reply to Bill Pedalino on 8-15-2017 in reference to Intake Manifold sealing problems common to both Sportsters and Shovel heads.I'm on a learning curve here and I'm just wanting gather as much experienced advice as possible to help me produce the best possible end product.Thank you for short but concise reply,it keeps me on track.Peter