Jim,
I like the idea of accumulators and tried to incorporate them into the system back in about ’99. My motivation wasn’t for the thermal expansion issue. Rather, I was trying to add some system tolerance for small internal leaks. Unfortunately, failure analysis revealed two unacceptable failure modes that didn’t have a simple solution and so I shelved the idea. Accumulators work by using a gas charge separated by a piston or bladder. The first and more critical failure mode was if the accumulator seal failed while the
aircraft was parked on the ground. It could unlock and collapse the gear. –small probability of occurrence, but catastrophic consequence. Possible mitigation options were to disconnect or isolate the ‘up-circuit’ accumulator when on the ground or to pin the gear in the down and locked position. –like the big boys do.
The second issue comes from the stored energy in the accumulator. The added potential energy released upon selecting gear-down would contribute to the “extension failure lock-up mode” by overwhelming the return side of the pump. The accumulator will increase the pressure pulse that propagates through the system.
Page 10 has more detail: http://www.n91cz.com/Hydraulics/Lancair%20Hydraulics.pdf
This issue could potentially be mitigated by one-way flow restrictor out of the accumulator or by perhaps increasing the down-side operating pressure substantially. I didn’t investigate these options because I was still stuck on issue one above.
A number of other options were explored using mechanical energy storage via heavy springs and pistons that would return to a zero energy state when not in use. All had their drawbacks, size, weight, cost, etc. Some were theoretically possible though and could be revisited.
Adjusting the “non-adjustable” thermal relief valves solves the thermal expansion issue, but not the leak sensitivity, so the concept is certainly worth exploring further. The important thing is to analytically explore all operating and, in particular, off nominal failure modes to discover undesirable side effects on paper rather in the plane. Unfortunately it is very easy to fix one problem and in the process introduce another. Let me know if you would like to discuss further.
Chris
Chris Zavatson
N91CZ
360std