For a long time I've been aware of this problem (mentioned in the article above) and have been trying to figure out some way of taking advantage of the regenerator while somehow compensating for the loss from the introduction of additional "dead air space".he primary effect of regeneration in a Stirling engine is to increase the thermal efficiency by 'recycling' internal heat .... In practice this additional power may not be fully realized as the additional "dead space" ...reduces the potential efficiency gains from regeneration.
http://en.wikipedia.org/wiki/Stirling_e ... egenerator
What I think might work is some kind of "elastic" regenerator.
Here is the theory.
When air in the engine is passed from the hot space to the cold space through the regenerator the regenerator gets Hot right? The opposite of what happens to the gas. When the gas is cooled the regenerator gets hot and vice versa. When the gas is heated the regenerator is cooled.
Certain metals like aluminum have a high coefficient of thermal expansion (expand when heated). There are probably some metals or alloys like zinc that would be more effective than aluminum.
If a regenerator were made out of some kind of solid aluminum (or other metal with a high CTE) cylinder with channels for air passage through it, and if it were coated on the outside with some elastic material like that high temperature silicone (RTV or similar) used for head gaskets, Or possibly some kind or elastic tubing could be used. then when the metal regenerator got hot it would expand. Therefore when the regenerator got hot it would increase the volume of the air space at the same time that the gas is cooled creating a higher vacuum to draw the piston in. Likewise when the regenerator was cooled it would shrink in size and reduce the air space at the same time that the gas was heated and expanding creating some additional pressure to drive the piston out, above that caused by the expanding gas.
I thought of this possibility quite some time ago but did not think aluminum would expand and contract enough to make a whole lot of difference, but it might be worth a try.
Recently however I became aware of "superelastic" metals. The kind used for medical "stints". The stints are narrow when cool but expand when heated by body temperature after insertion, so as to open up an artery or whatever.
So an "elastic regenerator" would act very much like a medical stint, enlarging the passage when heated and relaxing or contracting when cooled. I assume that this action would work against (or opposite to) the expansion and contraction of the gas in the cylinders and thereby compensating somewhat for the "dead air space" problem delivering more power to the piston.
Some other arrangement might be used, like having the expanding and contracting regenerator push and pull some additional pistons working against some metal springs or pneumatic "AIR springs", but the above, simply coating the regenerator with silicone seems like the easiest and simplest approach, for experimenting anyway.