It's interesting to note that, depending on who you ask, electrolytic hydrogen is cheaper (Roth), more expensive (Ellis/Sander), or the same cost (Greenwood) as hydrogen from the steam-iron plant. I suspect that it turns on what the assumed cost of power is. Bear in mind that nowadays, electrolytic hydrogen is much more expensive than hydrogen from steam reforming.
This cost data (Tables 5A, 5B) is not available in Grantville, but they can calculate the materials requirements and cost them out separately.
(1) materials only, steam and water treated as free; refrigeration power cost of 60 cents/1000cf.
(2) Ellis 595, mostly based on Sander; additional prices of 18.75 for silicol (p523) and 32-38 for hydrogenite (521ff). Cp. 462ff for fractional refrigeration, 445, 458 for Griesheim-Elektron 472 Carbonium.
(3) Greenwood 213, 234. Assumes power cost of 0.25p/kwh.
Ellis (537ff) breaks down the operating cost for electrolytic production of 632 cubic feet compressed hydrogen/hour (4,550,400 cubic feet/year of 300 workdays, 24 hours/day) and half that of compressed oxygen as follows:
Hydrogen Purification
I talked about purification of carbon monoxide in the section on "water gas." In essence, carbon monoxide may be removed by treatment with cuprous chloride, or hot soda lime, caustic soda, or calcium carbide, or by liquefaction. Carbon dioxide is eliminated by washing with slaked lime, or water under pressure. Bog iron ore will extract hydrogen sulfide. (Greenwood 211ff).
Hydrogen Transport
Generally speaking, in the early-twentieth century, hydrogen was compressed for shipment to industrial customers. In 1904, figure that a gas compressor for compressing 100 cubic meters of hydrogen every 10 hours cost $1000, and a second compressor for the associated 50 cubic meters oxygen would be $625. (Engelhardt 39). Ellis (556) estimated that compressors for a 10 cubic meter/hour hydrogen system would be $2850.
According to Ellis (538), an electrolytic hydrogen plant would require 4 kilowatt-hours for compression of 632 cubic feet (17.9 cubic meters) hydrogen to 300 psi (20 atmospheres), and 12 kilowatt-hours to compress 316 cubic feet oxygen to 1800 psi. Engelhardt (113) says that for compression to 100-120 atmospheres, the total power required would probably be about 4 kwh for 1 m3 hydrogen and 0.5 oxygen.
The tanks were also a significant expense. The plant had to purchase enough so that it didn't have to wait for empties to be returned in order to keep up with demand. The steel tanks weighed 10 kilograms per cubic meter gas held, and a 40 kg tank cost $11.75 in 1904. (Englehardt 118ff).
You have to be careful; don't use the same compressor alternately for hydrogen and oxygen, and don't use a former oxygen cylinder to carry hydrogen, or vice versa, without being sure that you completely removed the old gas. (Ellis 592).
****
An alternative to compression is liquefaction. Hydrogen was first liquefied in 1898. Liquefaction requires bringing the hydrogen to a pressure above its critical pressure (12.8 atmospheres), and then cooled below its critical temperature (-239.95oC). Keeping it liquid requires keeping it pressurized and cold, even in transport. And if you fail, well, remember that liquid hydrogen is a rocket fuel. I think it will be decades before liquid hydrogen appears in the new time line.
****
Since military balloons had to be launched near the front line, where transportation options were likely to be limited, the tanks were moved by a variety of means. The first use of compressed hydrogen in warfare was possibly in the British expedition to the Sudan (1885); each camel carried two 66 pound cylinders, each carrying 140 cubic feet (after expansion) of gas. (AGLJ) In the Boer War, fifty horses were needed to transport cylinders (weighing 1 pound/cubic foot hydrogen) enough to fill a 14,000 cubic foot balloon. (Greenwood 223). Later, the Germans used railway wagons that weighed 30 tons and carried almost 100,000 cubic feet hydrogen. (233) The American military neglected the balloon after the Civil War, but in 1891-3, the Signal Corps decided to add a tethered balloon and fill it with hydrogen from pressurized (120 atmosphere) cylinders. (Crouch 519ff).
Airships have much greater mobility than military balloons, so we aren't limited to "front line" options, but the rail network is much less developed in the 1632 universe.
The total cost of compressing and shipping hydrogen to a remote airship facility can be high. For 12.5 cubic meters hydrogen, compressed and shipped 300 miles, and empties returned, Schmidt estimated (1900) 29.5 cents to produce the gas, 2.5 to compress it, 16.25 as interest on the purchase cost of the tanks, 2 for labor, and 62.5 for the two-way freight, for a total of $1.13-9 cents/m3. (Englehardt 129).
In 1915, Fourniols compared the cost of producing hydrogen at a cheap-to-operate hydrogen plant and shipping it in compressed form, to generating it on site using the hydrolith process. The former produced hydrogen at a cost of only 0.4 francs/cubic meter. But compression and transport of 50,000 cubic meters for an unstated distance increased the cost from 20,000 francs to 960,000. In contrast, the same amount of hydrogen could be produced by the field process for 324,000 francs, of which only 40,000 was transport-related (carriages for the apparatus and reagents). (Ellis 534).
At some point, high pressure hydrogen pipelines, like the early-twentieth century one from Griesheim to Frankfurt, might reduce transport costs. (Ellis 440).
There are two complications with storing hydrogen; its great capacity for diffusion through other materials, and its ability to embrittle metals, include steel (Kirk-Othmer 13:851). That may limit the useful life of storage cylinders.
Hydrogen Recycling
The contents of a gas cell will become corrupted as hydrogen escapes and, more slowly, air enters. The hydrogen in this "spent gas" may be recovered for re-use by an adaptation (Greenwood 233) of the Linde-Frank-Caro liquefaction method used to separate hydrogen from carbon monoxide in the water gas processes.
Hydrogen Testing
To avoid explosions and fire, and maximize lifting power, it's important to know how pure the produced hydrogen is, and what other gases it's contaminated with. Hydrogen may be measured by combustion with excess oxygen, or by measurement of the thermal conductivity of the gas. Carbon monoxide will blacken paper moistened with palladium chloride, or it can be quantified by measuring the carbon dioxide formed by its reaction with hot iodine pentoxide. Carbon dioxide, in turn, is detected by its reaction with lime water or barium hydroxide. (272). Oxygen is revealed by blueing if the gas is bubbled through a colorless cuprous salt solution.
We can measure arsenic with the "mirror test" beloved of early detective stories, and hydrogen sulfide by its reaction with a lead acetate paper. (Greenwood 235ff, 254, 272; Taylor 193ff).
I leave it up to the reader to determine the extent to which these detection methods would be known in Grantville Literature, and how soon the necessary reagents and apparatus could be produced.
Conclusion
In the 1630s, I believe that electrolysis, whether of water or alkali, should be the dominant method of hydrogen production in Grantville itself. There's ready access to electricity, which, for the reasons I set forth in Cooper, "Aluminum: Will O' the Wisp?" (Grantville Gazette 8), should be relatively cheap for several years despite its ultimate dependence on burning coal.
And we don't have to worry about compressing the gas if the airship station is in Grantville. If we electrolyze water, we have the further advantage that we are producing oxygen (which itself is valuable) and the hydrogen is going to be of extremely high purity (at least if we use distilled water).