Specs for a foster fitting?

[Ribbonstone]

Just to mention.

Probes with Fosters machined onto one end  with an added to Foster adpator...are still basically Foster nipples that detach.  Have the same size/construction issues as any other Foster fitting. 

Would pretty much have to have the probe screwed directly onto the fill hose to avoid that..which might work for you if all your PCP’s took the same probe.  Doubt we’ll get makers to agree to standardized probe sizes.


Fosters, even steel ones, can get buggered up just by use.   The little lock-bearings in the QD can beat up the locking surface of the Foster.  Obviously this wasn’t hard steel, eventually raised a few burrs (little bright spots) that wouldn’t let the QD over the burrs.

WIN_20200609_13_05_36_Pro by Robert Dean, on Flickr


Taking apart (cutting open) a couple of QD’s...those little ball locks are hard.  Harder than the QD itself, are the same kind of little dents in the ball seats inside the QD...worse with the brass ones.

That would let the QD’s ball locks to kind of “dent” them selves wider apart….and move the lock more towards the outside edge of the Foster.

If you see something like the above picture...change BOTH THE FOSTER AND THE QD.

Hard to get air gun  (or any gun) makers to agree to much of anything.  Making a unique probe size, so that you have to buy THEIR probe, seems to be part of the profit margin.

[steveoh]

Just to mention.

Probes with Fosters machined onto one end  with an added to Foster adpator...are still basically Foster nipples that detach.  Have the same size/construction issues as any other Foster fitting. 

Would pretty much have to have the probe screwed directly onto the fill hose to avoid that..which might work for you if all your PCP’s took the same probe.  Doubt we’ll get makers to agree to standardized probe sizes.


Fosters, even steel ones, can get buggered up just by use.   The little lock-bearings in the QD can beat up the locking surface of the Foster.  Obviously this wasn’t hard steel, eventually raised a few burrs (little bright spots) that wouldn’t let the QD over the burrs.

WIN_20200609_13_05_36_Pro by Robert Dean, on Flickr


Taking apart (cutting open) a couple of QD’s...those little ball locks are hard.  Harder than the QD itself, are the same kind of little dents in the ball seats inside the QD...worse with the brass ones.

That would let the QD’s ball locks to kind of “dent” them selves wider apart….and move the lock more towards the outside edge of the Foster.

If you see something like the above picture...change BOTH THE FOSTER AND THE QD.

Hard to get air gun  (or any gun) makers to agree to much of anything.  Making a unique probe size, so that you have to buy THEIR probe, seems to be part of the profit margin.

Yup. I have a handful of male ”Foster” fittings that look like this. The Air Venturi brand fittings get to this point very quickly.

I think that If the various Airgun forums and groups team up, educate airgunners about these issues, and keep hammering away at the subject we might get traction. Also pointing out the issues when reviewing airguns will again keep the subject hot.

I’ve gotten my best paired male / female fittings from Best Airgun Fittings in England.  And that is key. Pair up the best higher pressure rated fittings for fill whips and Airgun fittings also.

Foster and Parker seem to work ok together mix and match but the sizes are different. Not optimal. Then you get into one manufacturer offering a four ball female fitting and another offering a five or six ball fitting. Me thinks more balls are probably better for distribution of stress.

[Ribbonstone]

A different Foster….about the same use (not that I’ve kept strict count over all the years).  Better steel would be my guess.

WIN_20200609_13_05_16_Pro by Robert Dean, on Flickr


Old stuff sometimes works “forever”, even if it’s an odd sized fitting.    I just keep using it because it works, not that it’s convenient or stylish.

So rather than a probe adapted to Foster….it's a QD adapted to foster.

DSCN0036 by Robert Dean, on Flickr

[bduares]

On my AV Nomad II I have noticed that one end of the fill hose has a connector slightly "looser" than the other. If I use the looser connector on my AEA probe I have to physically hold it in against the fill port hard until it has about 1k psi in the hose in order to be able to let go and let the fill continue. The other, tighter fitting, end I just pop it in and go.
So it looks like not just from the same brand but even on the same hose you can have differences big enough to matter

[Ribbonstone]

thinking it's contact area...that (the last picture) old big-azz quick connect spread the load over a wider area, which limited the "dents", and kept  it working "forever" (in PCP terms that would be 18 years).

[Privateer]

@Manny

Can you change the female part on your pump?
I can send you the one from my AV compressor which didn't seem to have a problem.
It's got 6 balls inside.
(So is that REALLY a female if it has 6 balls!)
 

[Nomadic Pirate]

@Manny

Can you change the female part on your pump?
I can send you the one from my AV compressor which didn't seem to have a problem.
It's got 6 balls inside.
(So is that REALLY a female if it has 6 balls!)
 

I did put the Nomad fill hose on the pump to see if it made a difference and it actually was worse,......I have a few female fosters so as soon as I get time I start try them all to see if I find one that seals.

[oldpro]

[Weaklink]

The fitting on my little-used Chinese pump would not seal with a few of my & friend's fittings.  Bought a couple of these and have been happy.  Your mileage may vary...

Amazon:  Outdoor Guy Paintball Remote Line Stainless Steel Quick Connect Disconnect 1/8 NPT

will not allow me to post direct link.... 

[Ribbonstone]

Being lazy...not going to do it.

Last China pump came with two QD’s.

Obvious by use, they were different sizes….so they must have expected some variation in Foster nipples.


 Just not ready to cut up what is working.


NOTiCE...not automatically blaming the foster nipples.....couod very well be the "slack" ion the quik detach.

[subscriber]

The problem is that the male connector has a straight cone that the balls in the female connector bear against.  This means point contact.  Point contact means infinite stress.   Even if there are 6 balls, that is a tiny contact area - until the male cone either elastically or plastically deforms to increase the contact area.

Permanent plastic deformation occurs when the contact pressure at the ball exceeds the yield strength of the material.  It would take very hard material to resist permanent deformation, unless the nominal contact area can be increased. 

The simplest way to increase the contact area is to increase the number of locking balls.  If the male fitting had a curved surface matching the ball radius, at least there would be line contact, rather than point contact.  Of course, that would require a standard size ball...  But then, hardened ball bearings are a commodity.  You just order them as 2.5 +- 0.01 mm  (assumed nominal diameter).

I decided to model fittings to demonstrate point contact VS line contact.  The standard fitting has a flat cone, while my proposal is a curved cone.  Note that fittings that have seen lots of use, end up with many dimples, approximating a curved cone.  I am simply suggesting the locking cone be machined to look like that from the start.

Also, the very edge of the locking cone should have a small fillet so that if the balls create simples, they do not raise sharp burrs right at the edge of the 8 mm OD.

[oldpro]

The problem is that the male connector has a straight cone that the balls in the female connector bear against.  This means point contact.  Point contact means infinite stress.   Even if there are 6 balls, that is a tiny contact area - until the male cone either elastically or plastically deforms to increase the contact area.

Permanent plastic deformation occurs when the contact pressure at the ball exceeds the yield strength of the material.  It would take very hard material to resist permanent deformation, unless the nominal contact area can be increased. 

The simplest way to increase the contact area is to increase the number of locking balls.  If the male fitting had a curved surface matching the ball radius, at least there would be line contact, rather than point contact.  Of course, that would require a standard size ball...

I decided to model fittings to demonstrate point contact VS line contact.  The standard fitting has a flat cone, while my proposal is a curved cone.  Note that fittings that have seen lots of use, end up with many dimples, approximating a curved cone.  I am simply suggesting the locking cone be machined to look like that from the start.

Also, the very edge of the locking cone should have a small fillet so that if the balls create simples, they do not raise sharp burrs right at the edge of the 8 mm OD.

Peter
 can I use this in the meeting next month?

[subscriber]

can I use this in the meeting next month?

Absolutely, yes, Travis.  Anything posted in public now belongs to the public.

I can even email you the 3D CAD assembly file so you can look at it from different angles.

[Back_Roads]

I run into the same issues with power takeoff shafts and couplers on the farm end of things also, hydraulic fittings.

[subscriber]

Travis,

I sent you an email with the 3D assembly files. 

Subject:  Foster fitting improvement suggestion (3D assembly model)


I used only four balls in the model, because I wanted the contact points to fall where the problem is most easily understood (easy to increase ball count to 5, 6 or 7 from inside the assembly model).  Cluttering the view with more balls would be a distraction, for the purpose of this exercise.

To show you how simple it is to change the ball count from within the SW assembly model, I have added 6 and 7 ball images, below.  7 balls is as many as can reasonably be fit in.  6 may be better, if one took a closer look at all of the mating parts...

[subscriber]

Did any of you notice how they made this male and female fitting pack so cheap?  By skipping a few details, such as deburring the holes the balls travel in - image below.

[Ribbonstone]

Worst to best Foster fittings so far.

Brass
Soft steel
What I think if bronze (no one left to ask what the alloy was)
Hardened steel.

If you can eyeball steel and tell if it is heat treated hard or soft...you got me beat.

[oldpro]

Travis,

I sent you an email with the 3D assembly files. 

Subject:  Foster fitting improvement suggestion (3D assembly model)


I used only four balls in the model, because I wanted the contact points to fall where the problem is most easily understood (easy to increase ball count to 5, 6 or 7 from inside the assembly model).  Cluttering the view with more balls would be a distraction, for the purpose of this exercise.

To show you how simple it is to change the ball count from within the SW assembly model, I have added 6 and 7 ball images, below.  7 balls is as many as can reasonably be fit in.  6 may be better, if one took a closer look at all of the mating parts...

Thanks Peter
 When I meet with the ASTM F15.06 Ill bring samples and explain the condition across the Industry and the lack of a common sizing

[subscriber]

If you can eyeball steel and tell if it is heat treated hard or soft...you got me beat.

I can eyeball how fast a file cuts it (in a non-critical area) 

I think the run-of-the-mill Foster fitting has a hardness around that of mild steel.

How hard should the body of the male and female fittings be?  Hard enough to resist indentation, but not to the point of being brittle.  I would vote for a spring temper, with a hardness of about 44 Rc.  The thing is that heat treatment after machining adds a huge amount of cost.  Unless it is done right, the parts can distort; or be left brittle.  Machining from prehardened and pre-tempered steel will slow the process down and cost even more...

There is about 350 lb trying to spit out an 8 mm Foster fitting at 4500 PSI (maybe less, depending on the sealing diameter of the O-ring).  Think what the contact pressure is, if each little ball has to bear 50+ lb.  Double the number of balls...

The amazon fitting above stated that the material is 304 Stainless.  That is nasty to machine because it work hardens drastically.  However, that could be used to advantage, if a surface rolling operation is included.  This could be applied selectively to lock-up surfaces, very quickly.

304 SS can only be annealed by heating; not hardened. 


My idea of shaping the ball contact seat on the male fitting to match the ball radius could easily be tested.  Just modify a stock part that has already been dented.  If will require a form tool, or a CNC machine to do it well.  If that reduces dimpling it may be an easy improvement when implemented in future products, even if not a complete victory.

Those who chamber and crown their barrels successfully by hand, would have the skill to run a dimpled male Foster fitting in a drill-press, and use a round file of appraise diameter to dress down the raised metal, following the shape already impressed by the balls as a guide.  As long as you are not flaking off plating, there is nothing inherently bad about the idea  (one member above spoke about dressing the dimples already). 

You can test the strength of a modified joint by hanging two people from it, if you have doubts.  (If you bend it, throw it away).    If the idea of modifying HPA fittings scares you, don't do it.  4500 PSI may be "a lot" of pressure; but 350 lb is not a lot of force for adequately sized metal parts to carry.  The problem is not the "size" of the metal parts, but the lack of size of the contact area at the balls...