Many commentators have suggested that hydrogen is the long-term fuel of choice for motor vehicles. Burning it releases no CO₂, and it can be (renewably) made from water using electricity. The hydrogen produced can then either be burnt in a conventional engine, or used in a fuel cell to directly produce electricity. In the very long term this is undoubtedly true, though it does demand a cheap, clean source of electricity to make the hydrogen in the first place.

Perhaps as a result of this interest, many individuals and companies have begun marketing devices and ideas for using hydrogen today in vehicles. There are two different schemes, with quite different technical considerations.

The first, and most straightforward, idea is to simply substitute hydrogen for petrol (gasoline) or diesel, either in whole or in part. Hydrogen is made by electrolysis, either on-board from the alternator electrical supply or at home using mains electricity, and then burnt in the engine. The energy released drives the vehicle in the usual way. This is commonly marketed as "Run your car on water!".

In principle this does of course work. The problem is that the whole process is very inefficient, and the energy you get from burning the hydrogen cannot possibly be enough to supply the energy required to make the hydrogen in the first place. With an on-board system, the engine is only about 40% efficient at best, and the alternator perhaps 90%, so even if you assume the electrolysis is 100% efficient you only get back about a third of the energy you put in.

More subtly, other systems claim that the hydrogen is used as a combustion enhancer. The idea is that adding quite small amounts of hydrogen to the fuel/air mixture cause improvements to the burn, which in turn leads to better economy and reduced emissions.

In principle this is a sound idea. Several reputable studies have been done into the idea, and there is no doubt that adding hydrogen to a petrol / air mix gives the following benefits:

• Better ignitability (leaner mixtures can be ignited)
• Higher flame speed (leaner mixture will burn reliably)
• Reduced tendency to knock (so higher compression ratio can be used)

So, this is a great technology and we should all rush out and bolt hydrogen generators onto our cars? Well, no. The problem is that the rest of the engine needs to be optimised as well to get the benefits. To take advantage of the higher octane rating, the compression ratio must be raised by redesigning the cylinder head or pistons. More critically, the ability to burn a leaner mixture is only useful if the engine ECU is configured to inject a very lean air/fuel ratio, and manage the required variations in spark timing and fuelling (which current production vehicles, without exception, are not). Simply speeding up the burn, even if theoretically advantageous, is likely to make economy worse if the ignition timing is not adjusted to suit.

As a result of these considerations, I am highly sceptical about the use of hydrogen generators as an aftermarket fuel "saving" device, and will remain so until the makers produce some good results. (Some proper scientific testing was conducted on one of these devices in October 2009 by the BBC, and found no significant improvement.)

Many devices that produce hydrogen via on-board electrolysis feed not pure hydrogen, but a mixture of hydrogen and oxygen, to the engine. This mixture (two parts hydrogen to one part oxygen) is commonly known as Brown's Gas or HHO. Various semi-magical properties are claimed for this gas, but these are widely disputed and should be considered with scepticism.

You can find some commentary on HHO devices at HHOinformation.info; I haven't checked all the site for technical accuracy, but there are some interesting comments on how people are making money from selling these type of devices despite the vanishingly small chance that they actually save useful amounts of fuel.

A note on alternators: a view often expressed regarding on-board hydrogen generation is that there is "free" electricity available for that purpose. The argument goes like this: a typical modern car alternator can produce 80A at maximum output (around 1kW), in order to meet peak demands, but since the electricity demand in normal conditions is much less (say 20A), the "excess" is freely available for purposes such as hydrogen generation without any additional loading on the engine.

In reality, the alternator is much smarter than that. If the vehicle electrical system only has a small demand, and the battery is fully charged, then the alternator will regulate itself to produce less electricity. Crucially, when it does this it also puts less mechanical load on the engine, and so fuel consumption is reduced. Adding an extra electrical load, such as a hydrogen generator, inevitably increases the mechanical work that the engine must supply to the alternator, and so increases fuel usage by a small but definitely non-zero amount.

Please also read the general comments on fuel "saving" devices, if you have not done so already.

If you found this page helpful, you may like to support my work. If you think I have made a mistake, or am talking complete nonsense, please take a moment to read the Response to Critics before saying so in public (on discussion Forums and the like).