A Technician reports their Bottle Shop has removed the valve assembly from their B777 composite crew oxygen cylinders on numerous occasions with the same basic complaint of oxygen leaking at valve stem weep hole or cylinder leaking at the valve body to cylinder. Valve handle can also be 'over-opened' and cause the valve stem to rotate inside the valve body and not allow the valve to be turned off.

Narrative

Crew oxygen bottle valve Manufacturer Part Number (P/N) #804216-01 has been subject to leaking problems at the valve body safety weep hole and has most recently been a problem for our company system for leaking complaints out in the field. The Company Bottle Shop has removed the valve from the cylinder on many occasions with the same basic complaints of 'leaking at weep hole' or 'cylinder leaking at valve.' The valve assemblies are then sent to the Oxygen Shop for Inspection/Repair. The Shop has determined several issues with the valve that could be safety risks:1) Deformation of the valve body has been visually observed on the valve neck threads. It has been determined that the OEM (Original Equipment Manufacturer) torque specs of (240-260 inch pounds) used to secure lock nut to drip tube is too high and causes the first threads of the valve neck to distort. We brought this issue to the attention of the Vendor. During a recent base visit; they [Vendor] allowed the reduction of the torque specs to 170-175 inch pounds. But even with this lower torque value; the Shop is still seeing approximately 30% failure rate in the assembly process. The OEM Manual contained no provision for checking the threads of the valve neck for distortion. Because of this; many valves have been sent back to service with threads that were distorted. The Shop has since incorporated a Go/No-Go ring gauge in the assembly process to screen for this distortion.2) The Bottle Shop has had problems with the valves not sealing totally on the cylinder necks. Deformation of the upper portion of the valve body neck has been determined after the valve has been torque one time on the cylinder. The OEM Specifications (Specs) for valve to cylinder torque is 45-55 FT pounds. But with the distortion of the threads from the lockout installation and the distortion from the first installation to the cylinder can cause friction to be present when the valve is installed to the cylinder for the second time! This can cause the torque valve of 45-55 FT pounds to be reached before the valve has completely sealed to the cylinder crush washer.3) The valve has a long history of complaints of 'leaking from the weep hole.' The OEM has stated to us (Company and Shop mechanics) that the valve is designed to leak from the weep hole when the valve is in the 'closed' position to relieve any pressure remaining upstream in the system. The Shop has argued that the weep hole leak is actually caused by a 'jack screw effect' when the valve is 'closed' and causing the internal seal to partially loose its ability for total pressure seal. Also; in the OEM Manual Description of Operation; there is no reference to such a leak when valve is in the 'closed' position. During a recent base visit; the Vendor allowed an increase in torque applied to the stem retaining nut to 125 inches pounds; up from 100 inch pounds.4) The Shop has been perplexed by the reoccurrence of the leaking weep hole complaint. The Shop would overhaul valve assemblies; send them to the Bottle Shop for mounting only to have them returned in a short time with the complaint of 'leaking at weep hole.' It was only while testing new assembled valves that it was discovered that if the valve handle is rotated to 'full on' position and then rotated slightly further to ensure that the valve was fully opened; the entire valve stem assembly (valve stem; sleeve; and guide) would rotate in the valve body. This would allow the stem seal to loose partial seal and leak from the weep hole with the valve open. And in one case this action caused the valve stem assembly to bind and rotate and not allow the valve to turn off! If this action occurred while mounted on a full or partially filled cylinder of oxygen it could present a very dangerous situation for aircraft and personnel. Both shops are very concerned over the overall safety of this valve and stand ready to give any further explanation that would be needed.[I would] suggestinspection of all B777 composite crew oxygen bottles and/or removal from fleet.