Unfortunately it seems this electrolyzer has died (fortunately I have 2 more). The replacement solenoid valve did fix the problem, but it stall wasn't producing deuterium at anywhere near the rate it should. It was producing 0.3l/h and should be producing 3l/h. Blanked off against a pressure gauge, the cell would still produce about 400psi output but at very low supply rate. Cell voltage was at the maximum 5v the board could supply and it was only drawing a few hundred mA. Connecting the cell directly to a power supply required 12-13V across the cell (normal voltage should be 2-3v) to drive the required ~8A. At this current the cell would produce the required 3l/h but would rapidly overheat due to the increased power dissipation. This effect points to PEM membrane degradation (see fig 2) in following paper.
https://h2tools.org/sites/default/files ... er_194.pdf
I decided to dismantle this unit for exploration an possibly future repair.
The fitting the hydride cartridge screws into also has a pressure relief valve on the back that was not previously visible.
The PEM cell has an M5 output thread, which will also fit a 10-32 face seal pneumatic fitting rather than the internal o-ring they designed. The bolts holding the steel plates (1/4" thick on the water side, 1/2" thick on the deuterium side) were torqued to about 16 in-Lbs as assembled.
Halves are sealed with a silicone o-ring
There is an array of channels to allow water access to the cell electrodes and membrane.
The electrode feed throughs are sealed with RTV
The output fitting is stainless and has a fiberglass backup washer
The metal mesh electrode is spot welded to another finer metal mesh and lays on top on another fine metal mesh. These electrodes have some oxidation (rust?) on them and the water in the electrolyser was starting to get a very fine rust particulate material in it when the electrolyser was run. The good ones don't have this issue.
Under this electrode is the polymer PEM membrane
The OEM membrane is bonded to a graphite cloth presumably impregnated with a catalyst. The membrane surface abuts an o-ring that steels it against the opposing electrode to trap the deuterium, a hole drilled through the stack to the other end allows deuterium to diffuse to the center and exit the electrolyzer
A rigid graphite plate contacts the graphite cloth on the membrane
The sealing o-ring is removed
The negative electrode is bonded to a graphite cloth that contacts the graphite plate
The hole through the electrode allows the deuterium to exit
and is sealed to the output fitting with an o-ring
Resistance measurements are as follows
Does anyone have a guess on the exact type of membrane?