Well, during the design phase especially I know it will be useful.
Throttling at Max-Q, during start up and at shut down some amount of
management should be useful. For monitoring engine health and efficiency
and for safety reasons as well.
I don't believe the weight is that much of a factor because a lot of the
controls are already in place they are just wide open. The sensors don't
weigh very much and the flight computer is already there. Some extra
power may be required forcing a little bit more battery weight.
During the turbopump testing phase I need more precise control without a
doubt to maintain limits and do automatic shutdowns if necessary. Record
data and allow for tuning the engine up on the fly.
So no other ideas about combustion monitoring I guess?
-------- Original Message --------
Subject: [AR] Re: Closing the loop on rocket engines
From: Andrew Burns <burns.andrew@xxxxxxxxx>
Date: Thu, December 10, 2015 11:16 am
To: arocket@xxxxxxxxxxxxx
There are several differences between rocket engines and car engines that
are key to this discussion:
- Rocket engines need only work for a few minutes at a time before being
thrown away (generally)
- Rocket engines operate with very tightly defined propellants at
controlled temperatures and pressures and within a very narrow operating
window
- Rocket engines are normally essentially binary, on or off, they're not
normally dynamically throttled
- Rocket engines have an absolutely immense thrust/power to weight ratio
compared to car engines and are accordingly so much more highly stressed
If there's one thing being party to a number of small (for an orbital
rocket) liquid rocket engine test fires has impressed upon me it's just how
quickly things can happen and how much energy release is involved. There
simply is nothing like a rocket engine for staggering amounts of energy
released in a steady state condition, they really are just barely
controlled continuous explosions. As has been mentioned there is already an
element of closed-loop control used for things like mixture ratio control
or startup/shutdown sequencing but anything beyond that hasn't really been
necessary and any additional complexity is generally to be avoided as it
represents weight and failure modes.
Andrew
On Fri, Dec 11, 2015 at 4:40 AM, Valtteri Maja <valtteri.maja@xxxxxxxxx>
wrote:
One big thing about ship diesels nowadays is good monitoring and control,
you can get good reliability, low emissions and low fuel consumption that
way. All meaning lower costs. Also engine starts don't belch huge amounts
of smoke (though that's not important for every operator).
The difference in rockets compared to most other internal combustion
machines is that rockets operate on a very narrow set of conditions that
are well known beforehand, so benefit from measurement and control are far
less.
Rocket landers have the most variable operation points so might benefit
the most.
What about things like the startup transient? Relatively minor performance
indications for this, but might be a big thing on engine loads - you can
avoid overbuilding the engine if you can guarantee that the startup is
gentle. On the other hand, it should be the same every time and shouldn't
need any sensors or feedback control.
Also if you can see that you accidentally ran lean at some point, there
might be a burnthrough of that engine on the next flight (or if you haven't
yet lifted off, on this flight), so better not do it.
--
Valtteri Maja at gmail com
10.12.2015, 16.57, Paul Breed kirjoitti:
Getting simultaneous propellant depletion at clost to optimum ISP is what
matters.... it does not matter if the earlier part of the burn might be
slightly off on O:F the steep part of the curve is at the end.
On Wed, Dec 9, 2015 at 9:21 PM, David Weinshenker <daze39@xxxxxxxxxxxxx>
wrote:
On 12/09/2015 08:59 PM, Monroe L. King Jr. wrote:
I'm not suggesting using car O2 sensors I know they wont work. I'm
looking for something that will work.
Those optical sensors are not the same as O2 sensors.
Engine management in cars is what I know I am speaking generally in that
direction based on what I know about that.
I see plenty of reasons to want better control over the engine.
Yes, mixture control is a good reason.
A valve can be controlled, pressure/rpm of the power turbine and many
other factors.
If you want to do closed-loop control of rocket O/F ratio, there may be
usable methods other than direct flame-chemistry sensing - e.g., using
the pressure drops across the injectors as a proxy for flow measurement,
and then "closing the loop" to maintain a desired relationship between
those signals.
-dave w