The key issue we faced with the piping was an information vacuum. Plans were not readily available, and once we had plans we were unable to find a plumbing company willing to do the work. Oddly, ReliOn (the fuel cell supplier) were little help: They were of course familiar with their K-cylinder/manifold/fuel cell system, but they seemed to have no familiarity with any other piping configurations, nor could they direct us to “standard” plans. Their official ReliOn installers were too expensive and too far from our island. We had no experience with hydrogen, though plenty with other types of plumbing. We finally said, “OK, how hard can this be? We’ll do the pipe ourselves”. It worked out great.

Hydrogen plumbing turns out to be straightforward, though sometimes fussy. The problem, endemic to hydrogen technology, is not that the work is complicated, but that there’s no-one who will say, “Here’s how you do it.” We had Powertech Labs (Vancouver BC) draw up plans for us and spec out parts. For the most part these plans worked out well. We had to make some corrections. A schematic of our piping appears below.

Hydrogen is corrosive. Chemically, it has an active electron and behaves somewhat like reactive halogens such as chlorine. This means that all pipes and fittings must be stainless steel, which is not cheap (see Cost). The fittings, available from Parker or Swagelok, are all compression or iron pipe thread and are easy to assemble. The pipe comes in 20 foot lengths.

It is vital to isolate the storage tank with shut-off valves, and to add purge intakes on each branch of line. This is because habitual, careful purging with nitrogen is vital in hydrogen work. One can expect to do a fair amount of tinkering with the plumbing, adding gizmos and troubleshooting leaks etc. Purging a 3/8″ line is easy; purging a 500 gallon tank is not (see Purging).

We were surprised to discover that the “one-way” check valves spec’ed out by Powertech were not in fact one-way, either because of the nature of hydrogen or due to an error in the specification. It took some detective work to figure this out. We solved the problem by changing the internal springs on the valves. The moral of the story is, even well-engineered plans may contain error.

Hydrogen leaks can be tricky creatures, sometimes very slow and thus difficult to find using soap solution. Often pressure testing overnight is the only way to check them.

Two views of the “out” line running from storage tank to fuel cell. The valve at the tank is open, the purging inlet valve is closed. The thick tube is part of the main tank pressure relief valve. The line running off to the right in the right hand picture goes to a digital pressure transducer (see below). A pressure relief valve (with thin blow tube) and pressure gauge are both visible. In the right hand picture the one-way check valve is visible at the lower corner of the yellow label.

The pressure transducer in the “out” line. On the left, conduit takes wire to the building. The redundant fittings to the transducer’s right allow us to remove the gauge for calibration without replacing too much pipe. ┬áBoth Parker and Swagelok fittings are visible.
The “in” line running from the electrolyzers to the storage tank, at the tank. At bottom is a bleed valve. Also visible are a shut-off valve, a check valve, and a second bleed valve at the entrance to the tank. This second valve is an artifact, from the original plans which we had to modify.

For more pipe details and pictures, see Manifold & Wing, Storage Tank, and

Hydrogen Codes.