How to fix a TFX 2nd-stage that keeps leaking slowly whenever it is purged?

[garywong007]

I got this set of Atomic TFX last November and have put about 40 dives on it. Recently the 2nd stage starts to leak slowly every time it is purged. I checked the IP today and it reads 135 psi on sweep, drifts a bit and locks at 137 psi. The 2nd stage does not leak when air supply is turned on and breathing normally from it but whenever it is purged, it will keep on leaking slowly and cannot be stopped until the air supply is turned off. The magnehelic reads 1.5 CIW.

I took it apart, check and clean every parts, lube them and put them back together. After reassembly, I then followed the adjustment procedures based on the manufacturer's service manual, which says : "A) 1. Pressurize regulator and cycle lever, 2. If leaking, first check and make sure the lever has slack, 3. If still leaking, tighten spring adjuster until leak stops. B) 1. Set lever height (with the lever set screw) so there is ~1/16 to 1/8" of slack (This is done easiest by tightening set screw until leak starts then backing out 1/4 to 1/2 turn". To my frustration, the same problems still occurs - slow leaking starts once the purge button is pressed and it cannot be stopped. I tried to increase the spring tension but even with the spring adjuster all the way turned in, the leaking still won't stop. Subsequent adjustments were made with the spring tension adjuster and lever position at many combinations of different positions but still cannot stop the leak.

What could possibly be the reason and what should be done to correct this? It seems to me that the 2nd-stage's sealing holds pretty well against the air supply at IP but once the purged button is touched, the Venturi effect is so strong that it won't allowed the valve to be sealed again.....

Cheers

 

[BoltSnap]

@rsingler

 

[rsingler]

There are two possibilities.

We saw exactly this problem with the Scubapro D420 when it was first released. Your comment from the manual about lever slack is the key, but it's not intuitive since the valves are so different from end balanced barrels. If (for whatever reason) the lever doesn't have a little slack when pressurized, the poppet is pinned in place at the bottom. And if that position leaves the seat just barely in contact with the knife edge, the valve may leak. No matter how much spring pressure you put on the poppet, it isn't translated into more pressure on the seat/knife edge valve seal. Instead, the tip of the poppet just pushes harder on the base of the lever and nothing moves. It could be that the lever is slack, but that your front cover mould is putting just a hint of pressure on the nipple that links the diaphragm to the lever, eliminating the slack.
To evaluate this, do all your tuning adjustment with the front cover off, and see if the problem goes away.

The second possibility is more prosaic.
The thing about light-spring center-balanced valves is that the spring is light. That means that the seal is critically dependent upon knife edge and seat condition.
1) disassemble the poppet and flip the seat to the fresh side
2) remove the knife edge with a plastic crochet hook and put it under a cheap USB microscope. Any machining marks or scratches may be enough to cause your issue. A little 4000-8000 micromesh on a pencil can fix that.
3) when you reinstall it, make sure the floating knife edge is absolutely flat in its pocket.

But the reg's resistance to increasing spring pressure makes me think the poppet/lever linkage needs to be looser than it is now. Obviously, too loose will raise cracking effort.

Finally, confirm that it's indeed a valve leak, rather than IP leaking around a faulty seal where the housing sits in the valve tube.

Good luck!

 

[NauticalNick]

Interested to hear about the solution for this as I was eyeing that regulator for myself. Everything I have read thus far has been pretty positive (initial reviews) so it would be disappointing if this was a design flaw in such a premium product.

 

[-JD-]

I don't know how Atomic judges warranty "voiding" where you are, but at 4 months and 40 dives, I would think it SHOULD be fully covered (parts and labor) and may be worth inquiring before digging too deeply into it. You may also get some of the new improvements that are being incorporated.

 

[rsingler]

As I reread your post, I was struck by the connection to purging. Can you just breathe from a pressurized reg with no onset of freeflow? The valve opens and closes normally? I think that's what you told us...

If it's just related to purging, when Venturi takes over, that's the time the lever is slammed fully into the case. There may be a little mechanical issue that might respond to a judicious whack on your thigh. If your lever is as wiggly as it's supposed to be, it might be jamming askew in the little oval above the hex screw during a purge on the surface. I'd be surprised if the same thing happened underwater.

To evaluate, try moving the Venturi lever to Minimum and purging again. That restricts the amount of flow-induced lever pull. Does it still happen in this condition?

Interesting issue...

 

[garywong007]

There are two possibilities.

We saw exactly this problem with the Scubapro D420 when it was first released. Your comment from the manual about lever slack is the key, but it's not intuitive since the valves are so different from end balanced barrels. If (for whatever reason) the lever doesn't have a little slack when pressurized, the poppet is pinned in place at the bottom. And if that position leaves the seat just barely in contact with the knife edge, the valve may leak. No matter how much spring pressure you put on the poppet, it isn't translated into more pressure on the seat/knife edge valve seal. Instead, the tip of the poppet just pushes harder on the base of the lever and nothing moves. It could be that the lever is slack, but that your front cover mould is putting just a hint of pressure on the nipple that links the diaphragm to the lever, eliminating the slack.
To evaluate this, do all your tuning adjustment with the front cover off, and see if the problem goes away.

The second possibility is more prosaic.
The thing about light-spring center-balanced valves is that the spring is light. That means that the seal is critically dependent upon knife edge and seat condition.
1) disassemble the poppet and flip the seat to the fresh side
2) remove the knife edge with a plastic crochet hook and put it under a cheap USB microscope. Any machining marks or scratches may be enough to cause your issue. A little 4000-8000 micromesh on a pencil can fix that.
3) when you reinstall it, make sure the floating knife edge is absolutely flat in its pocket.

But the reg's resistance to increasing spring pressure makes me think the poppet/lever linkage needs to be looser than it is now. Obviously, too loose will raise cracking effort.

Finally, confirm that it's indeed a valve leak, rather than IP leaking around a faulty seal where the housing sits in the valve tube.

Good luck!

Dear Rob, thanks for the advices. When I return to my workshop next week, I will look more closely into the possibilities that you have mentioned. Actually, the unit came in with a faulty 1st-stage too. The IP sweeps at 155 and climbed to and lock at 172 psi. But even at such IP, the 2nd-stage is holding its sealing well and not leaking when breathing normally (but start to leak and won't stop once it is purged). So initially I thought it was a problem with the 1st stage. I looked carefully and found tiny grooves on the piston shaft. Since the reg is only 4 months old and out of good will, instead of polishing the shaft, I replaced it with a new one and the IP is fixed (as mentioned in the 1st mail, it sweeps at 135 and lock at 137 psi) But then, just when I thought the root cause is identified and resolved, the same problem occurs with the 2nd-stage even with a well-tuned and steady 1st-stage.

And yes, I always do the tuning with the front cover off so I think that rules out the possibility that the front cover is interfering with the mechanism.

Anyway, I will try more different ways to fix that and keep you informed of the result....


Cheers

 

[rsingler]

I mentioned this issue in an email to the Chief Technician Trainer at Huish. We'll see if he has any thoughts.

 

[garywong007]

As I reread your post, I was struck by the connection to purging. Can you just breathe from a pressurized reg with no onset of freeflow? The valve opens and closes normally? I think that's what you told us...

If it's just related to purging, when Venturi takes over, that's the time the lever is slammed fully into the case. There may be a little mechanical issue that might respond to a judicious whack on your thigh. If your lever is as wiggly as it's supposed to be, it might be jamming askew in the little oval above the hex screw during a purge on the surface. I'd be surprised if the same thing happened underwater.

To evaluate, try moving the Venturi lever to Minimum and purging again. That restricts the amount of flow-induced lever pull. Does it still happen in this condition?

Interesting issue...

Hi, Rob,

yes, I can breathe from a pressurized reg with no onset of freeflow (and the cracking is at 1.5 CIW) and the valve opens and closes normally. I also did put the Venturi lever to the minimum but the free flow still occurs nevertheless and cannot be stopped.... It is on one hand frustrating but on the other intriguing... prompts me to rethink and rethink and rethink again....

cheers

 

[garywong007]

I mentioned this issue in an email to the Chief Technician Trainer at Huish. We'll see if he has any thoughts.

Are you referring to Glenn?

Rgds