If you have a few hundred million $ to put into stirling development then by all means have a go and see what you can do, but almost no-one who has actual experience of Stirling engines and their problems (and I know a fair number) would put their own money into developing them for automotive LENR. Phillips spent something like 1-2 billion over 30 years, NASA probably half that, GE, SES, STM, GE, USAB, Kockums, Sunpower, Stirling Biopower, Infinia, Microgen and lots of other smaller players have all also managed to each spend 10's - 100's of millions without every managing to create a product that was even half way to being commercially viable, primarily due to high manufacturing costs, but also intractable seal reliability issues.
The Philips (then USAB then Kockums) 4-95 engine you talk of never had exceptional lifetime - at best it would only operate for 2-4000 hours only before the seals failed (though often failed earlier), though that is not something it is easy to learn through reading the literature - for some insight into these issues read the 1980s MOD I and MOD II reports from NASA. For comparison IC car engines typically manage 4000 hours or more, truck engines 10's of thousands of hours. The 4-95 (a Rinia/Siemens type 4 cylinder kinematic Stirling) is in fact the engine that was further developed and productionised by recently bankrupted SES. They were talking of making 50,000 of them at one point, but their manufacturing costs were still in the $15-20k range, and that is after they had demounted and centralised all of the expensive hydrogen re-compression and storage to reduce their unit cost (these systems add a lot of cost to any hydrogen stirling engine). Infinia's 3kW free piston engine is in the $10k range, but uses Helium not hydrogen. Optimistic estimates for mass production of the NASA 30kW (60kW peak, but very short life at power) automotive MOD I and MOD II engine developed during the 1980s typically ranged around the $5k mark, but would be far more expensive now due to inflation and greatly increased nickel and cobalt prices - as the SES experience attests. In large scale mass production the cost of a product generally comes down to a multiple of 2-4 times the material costs, the problem is that unlike cars for big and heavy Stirling engines large quantities of those materials are very expensive and difficult to work nickel and cobalt superalloys, and for free piston engines lots of expensive permanent magnets are also required. In terms of specific components the regenerators are extremely expensive ($1k) to make being formed out of sintered stacks of 100's of layers of <50µm stainless steel wire mesh. The heaters are a nightmare to make being 100's of superalloy tubes that typically can't be welded so must be vacuum braised together at high temps all with zero leaks. The coolers comprise 1000's of 1mm tubes brazed into large parallel flow heat exchangers, and all of these must be assembled in casings with zero porosity, in extremely clean conditions with seals that will reliably seal against leakage of high pressure hydrogen. All of these components must also withstand temperature extremes between -40 and +80°C in dirty, corrosive, high vibration environments for 10-20 years while never releasing hydrogen in a dangerous manner when in a confined space like a garage, and somehow also not creating excessive dangers when involved in a crash. As for recirculating hydrogen from leaking conrod seals, that is done anyway, but separating the oil out is difficult, and you need to oil cool and lubricate those pumping leningrader (PL) conrod seals so the bigger problem is that as the seals leak they also transport oil into the engine from the crankcase, leading to blocking of regenerators and carbon particles destroying the seals. So take an (optimistic) $10k 30kW 200kg Stirling + generator + very large radiators (need much cooler temps than IC engines) to the already heavy and expensive $5-10k electric powertrain and you can perhaps begin to see why stirling is such a non-starter for vehicles compared to: Rankine turbine generator + condenser that weighs perhaps 100kg (turbine and generator are smaller as operate at much higher speeds) and costs probably $3-5k when mass produced. Recuperated Brayton (like capstone C30) that weighs 100kg and costs about $5-10k when mass produced. On 6 February 2012 18:50, Jones Beene <[email protected]> wrote: > *From:* Robert Lynn **** > > ** ** > > > I've also been involved in the development of hydrogen working fluid > stirling engines, and while they might look attractive there are big > problems:**** > > -Very expensive and heavy ($1000/kW, 5-10kg/kW for kinematic engines (ie > with crankshaft)**** > > ** ** > > That conclusion may be premature and short sighted, given the advantages.* > *** > > ** ** > > The high cost to date for Solar Stirling seems to more of a issue of mass > production (lack thereof). Certainly, it can cost 10 to 100 times more to > produce engines one-off or low volume now, compared to the typical > automotive, robotically enhanced, engine production line. However, I see no > ultimate impediment to this for the Stirling concept - once there is demand > for millions per year. That kind of demand would be guaranteed if mated to > a Ni-H heat source.**** > > ** ** > > The Solar 4-95 Stirling engine developed by United Stirling of Sweden, was > as an outgrowth of an automotive engine development program – and is > expensive due to low volume and the need for exceptional lifetime in > operation, far more than any car. But they would not have gotten into it if > there was a systemic problem that could not be overcome with higher demand > and better engineering.**** > > ** ** > > IOW all of the negativity seems to be a short-horizon issue that can be > resolved simply by high demand and a few “workarounds.” Except for hydrogen > seals and the nickel, the cost of a converted ICE should be in the range of > standard auto engines (if and when mass production is guaranteed). **** > > ** ** > > The easiest solution to the sealing problem (the workaround) is to “live > with it” in the sense of providing only the simplest solution – the best > O-rings, etc and then to utilize makeup H2 from onboard electrolysis. **** > > ** ** > > Only one liter of H2/min (or less) should be adequate for makeup of seal > leakage in a 50 kW engine operating as a genset for a Prius style battery > pack using standard sealing techniques for hydrogen. **** > > ** ** > > That amount of H2 used as a makeup would be parasitic for about 200 watts > from the genset and is of no risk as a slow leak, due to the extraordinary > mobility of H2. **** > > ** ** > > IOW this workaround solution is *de minimis* in terms of net value of an > installed engine which does not demand fossil fuel. Moreover – it sounds > like just the kind of objection that the OPEC petro-lobby would dream-up to > thwart Stirling development at this critical stage.**** > > ** ** > > Jones**** > > ** ** > > ** ** > > ** ** >
