A commercial pneumatic actuator on the valve it's designed for will be
more than fast enough unless you have unusual impulse bit
requirements. Indeed it's not uncommon to put an orifice on the
pneumatic inlet to slow such actuators down.
Seal friction (both ball seats and stem) is typically the dominant
force a valve actuator has to deal with. You can measure the force
needed by making an adapter for a beam style torque wrench, chilling
down the valve with LN2, turning it, and reading the static and
dynamic torques. But commercial pneumatic actuators are designed to
have an overabundance of force unless you do something really dumb
like fill the ball valve with water before freezing it.
For a throttled engine with a geared motor drive the inertia of the
motor shaft is the only inertia that matters much, but that's not your
situation. The pistons in a pneumatic actuator have much more inertia
than the ball and stem assembly, but it's not something you need to
calculate out unless you have an annoying member of the team you want
to give a pointless but important sounding job to.
Make sure you mount the ball valve with the bleed hole in the ball
pointing upstream.
Ben
On Thu, Aug 4, 2016 at 10:15 PM, Erin Schmidt <7deeptide@xxxxxxxxx> wrote:
Hi Everyone,
Thanks for the responses, it's nice to have such a knowledgeable community
on this list willing to dialog with interested engineering students. I'm
also helping out with PSAS's liquid propulsion project. This is planned to
be a restartable static test engine, hence the augmented spark igniter.
To expand a bit on Kristen's question, more generally speaking how do we
define design requirements for MOV/MFV actuators? The advice offered by
Huzel & Huang, and in NASA SP-8097 (Liquid Rocket Valve Assemblies) seems
broad and lacking in specifics. Sizing the actuator should be based on the
required response time and estimations of the effect of dynamic and friction
forces acting on the ball, stroke length, trapped fluid, the inertia of
moving parts (the ball and extension) inasmuch as they impact the valve
actuation timing, and how repeatable it is. However, I haven't seen much
information on determining the required response time for main valves a
priori. I assume much of the sequencing is detemined through testing,
however I'm still curious if there are any general guidelines or rules of
thumb. Based on anecdotes on this list, and elsewhere is seems like a rule
of thumb repeatable sequencing precision in the low 100's of ms is
required(thanks Lloyd). We are using a Worcester C44 cryoball valve with a
Worcester pneumatic Series 39 actuator (piston, double-rack and pinion
type). We expect the actuation time of the MOV to be about 2 seconds. I'm
not sure how precise or repeatable this timing is yet, though I assume we
will be able to have a ~200ms LOX lead (fully open post chilldown; the
pintle will connected to the main valve via braided flexline). We are not
sure if this is too slow or not. Can you guys offer any advice from
experience or point us towards references that might shed light on the
issue?
On Wed, Aug 3, 2016 at 12:54 PM Lloyd Droppers <ldroppers@xxxxxxxxx> wrote:
--
Looks like a reasonable first pass to me for timing.
-I presume the nitrogen valve pressurizes both the LOX and the ethanol
tanks. If so make sure you wait long enough that your LOX tank is
pressurized to at least ~90% of full pressure before going to the next step.
This may take a surprisingly long time depending on your exact set up.
-Are you doing a prechill at all before or during this autosequence? It is
not necessary, but it is nice to have, and it will change the amount of time
it takes LOX rather than GOX to enter the chamber quite a lot.
I like to think about timing in terms of when the propellant enters the
chamber and back out the valve timings from that using the valve opening
time and hydraulic lag from cold flow data. For a small LOX ethanol engine I
think you want a LOX lead of between ~0.0-0.2 seconds into the chamber if
you have a torch igniter running. The initial Oxidizer will be GOX / Fluffy
lox so having OX come first by a small amount makes sense.
Also if you have long fuel lines make sure you purge them the same amount
each test, as the tube fill times, especially on test stands, can be quite
significant.
Lloyd
On Tue, Aug 2, 2016 at 1:15 PM, Thomas McNeill <thomas.mcneill@xxxxxxxxx>
wrote:
I agree with Ben. Solid igniters are very easy to get going. You can
use model rocket motors, like I have done, or make your own grains. I used
Epoxy and KN for my home made ones.
On Tue, Aug 2, 2016 at 2:26 PM, Ben Brockert <wikkit@xxxxxxxxx> wrote:
On Tue, Aug 2, 2016 at 6:11 PM, Robert Watzlavick
<rocket@xxxxxxxxxxxxxx> wrote:
I wouldn't necessarily say an ASI or torch igniter implies a
restartable engine. I think of it as more insurance against a hard start
since you can verify igniter chamber pressure prior to opening the main
valves.
You can put an interlock in a pyro igniter. On an end-burning igniter,
embed a thin wire near, but not at, the start of the igniter grain.
When it is disrupted you have a reasonable expectation that the
igniter is burning in a compter-sensable way.
That said I'm all for spark torch igniters, but if I was trying to get
a liquid engine project working as simply and quickly as possible it
probably wouldn't be included. Solenoid valves to run the igniter can
easily cost as much as main ball valves.
Ben
Ad astra
Erin Schmidt
