Barry:
This is not directly relevant to whether the increased density alone could
produce a 33% increase in altitude (as I said previously, that turns out to
have been confounded with a change in grain design that put more propellant
into the vehicle above the density effect).
However, I have a current baseline design that shows 181 km w/ a burnout
velocity of 5965 ft/sec; an alternative grain design shows 206km at 6374
ft,sec. The total impulse of the former is 222561 n-sec.; of the later
231793 n-sec. Thus a difference of about 4% in total impulse appears to be
adding 409 ft/sec (6.8%) to the burnout velocity and 13.8% to altitude.
Bill
On Fri, Jul 12, 2019 at 6:00 PM Barry Jolly <1bcjolly@xxxxxxxxxxxxx> wrote:
I was figuring total impulse would give a valid comparison of the
propellant/motor combination and maybe be a good yardstick to judge one
better than the other. I looked up what I could find about your rocket.
Your history compares somewhat to mine. On the East coast we don't have
the test areas you do so most of my experience is with propellant, not
flight testing. Managed to get a hold of some Thiokol rubber ball kits in
the 11th grade (1963) and made propellant out of them. Also managed to get
lead poisoning from the curing agent. Fun days back then. Best of luck
with your project.
On Friday, July 12, 2019, 7:50:05 PM EDT, William Claybaugh <
wclaybaugh2@xxxxxxxxx> wrote:
Barry:
Current baseline is at 222,600 newton-seconds (combining both stages).
I’ll have to run a model to figure out what the 150 km baseline was
at....total impulse isn’t something I use much.
Bill
On Fri, Jul 12, 2019 at 3:07 PM 1bcjolly <1bcjolly@xxxxxxxxxxxxx> wrote:
How does total impulse compare between the older motor and the present
design?
Barry Jolly
Sent from my Sprint Tablet.
-------- Original message --------
From: William Claybaugh <wclaybaugh2@xxxxxxxxx>
Date: 7/12/19 15:09 (GMT-05:00)
To: arocket@xxxxxxxxxxxxx
Subject: [AR] Re: Vacuum processing of solid propellant
A careful review of the history of design changes on my two-stage shows
that the increase in performance from circa 150 km to around 200 km was
caused both by the measured increase in density of the vacuum post
processed propellant but also by the near simultaneous discovery of a
modified BATES grain design that pushed volumetric loading to a little
above 86% from a previous about 80%.
After reviewing that history and some additional work I’ve identified a
monolithic grain design that gives 83.5% loading in the upper stage and 85%
in the lower without erosive burning concerns. That is now the baseline
and, using the slightly denser propellant, sims to about 200 km, including
7 lbsm of ballast.
Bill
On Thu, Jul 11, 2019 at 10:10 PM William Claybaugh <wclaybaugh2@xxxxxxxxx>
wrote:
David:
To be clear, I’m open to there being something wrong somewhere in the
simulations; I’m not claiming this two-stage will do more than 150 klicks
when asked by non-professionals.
If there is something wrong in the sim, I’ve yet to find it...but the
results are hotter than I expected during initial design. But then I’ve
managed over the last two years to stuff about 12% more propellant into it
than I initially thought possible.
Bill
On Thu, Jul 11, 2019 at 10:00 PM David Summers <dvidsum@xxxxxxxxx> wrote:
Btw, thanks for all this. It sounds like most of the performance gain is
design specific, as in given a fixed design this change makes a large
improvement. Still fascinating!
-David
On Thu, Jul 11, 2019, 5:54 PM William Claybaugh <wclaybaugh2@xxxxxxxxx>
wrote:
Troy:
It’s late on this side of the lake; I’ll shoot you the most current files
tomorrow my time.
Bill
On Thu, Jul 11, 2019 at 9:48 PM Troy Prideaux <troy@xxxxxxxxxxxxxxxxxxxxx>
wrote:
Bill,
Troy:
The two test to date were, as before, near 400 psia and near 750 psia; the
500 psia Isp is an estimate coming out of Burnsim based on the measured Isp
at the two pressures.
The “estimate” part is fine, so long as you appreciate that to claim that
the same propellant in a denser form produces a different specific impulse
than the lower density equivalent can only be truly verified by conducting
2 instrumented static tests measuring both delivered total impulse and
pressure curves that closely match for a true apples to apples comparison.
That implies that the Kn ratio profile likely needs to be adjusted between
the 2 tests to accommodate the reduced mass discharge rate for the reduced
density propellant, but also (conversely) for the difference in burn rate
(normally increased for lower density resulting in an increase in discharge
& flow rates). Suboptimal densities often influence the burn rate exponent
too which can further complicate the comparison or at least achieving
useful comparable pressure curves.
Where I’m coming from (in terms of specific impulse) is the ultimate
comparison is to say I can achieve x isp with this density and I can
achieve y isp with that density both in their optimum configuration ie.
expansion ratios matched as best as possible to the conditions and
structural margins likewise.
In a dynamic sense, it does make sense that the higher density / slower
burning propellant will ultimately achieve a higher specific impulse due it
working for a longer time at a higher altitude, but I’m not sure if this
was factored into the simulations?
The 7 second trajectory average increase for the second stage (compared to
the previous “air processed” propellant) is coming out of the trajectory
sim and the estimated 2.5 second gain (at 500 psia) that I assume reflects
the higher aluminum content; I don’t see how density can have an effect on
Isp.
Well, yeah, that makes more sense.
I’m happy to share the motor files and RASAero file if you’d like to look
them over.
I’m happy to look over them, up to you Bill.
Cheers,
Troy
Bill
On Thu, Jul 11, 2019 at 7:39 PM Troy Prideaux <troy@xxxxxxxxxxxxxxxxxxxxx>
wrote:
So, just to be clear, the comparison measurements were all done and
measured at **very** similar pressures (*measured pressures*) not implied
similar pressures based on identical Kn profiles?
Also (just to be clear), your implied increase in specific impulse of the
upper stage (7 sec) is entirely based on porosity reduction or is that also
incorporating the additional metal influence.
Regards,
Troy
*From:* arocket-bounce@xxxxxxxxxxxxx [mailto:arocket-bounce@xxxxxxxxxxxxx]
*On Behalf Of *William Claybaugh
*Sent:* Friday, 12 July 2019 11:22 AM
*To:* arocket@xxxxxxxxxxxxx
*Subject:* [AR] Re: Vacuum processing of solid propellant
Troy:
The slightly increased Isp is measured in static tests at 2.5 seconds.
The additional flight performance is increasing altitude in the sim.
The baseline 231 seconds (at 500 psia) is measured in over 100 static
tests of the “air processed” propellant. I have to date only two tests of
the slightly denser version but those tests imply 233.5 at 500 psia. (Note
that the vacuum propellant is 10% Al rather than the baseline 7.2%).
I’ll be doing more test w/ more vacuum degassing time once the desert
cools some (read: November); those tests will be at around 400, 700, and
1000 psia and should provide a better reality check. For now, that is the
data I have to plow into the sims.
I should add that changes in grain design—the effect of a longer first
stage burn having been noted—are also in play here. My very most recent
modeling has extended the booster burn to 11 seconds while adding 3% more
propellant (smaller core); that pushes altitude from 210km to near 230km.
I should also note that 90+ % of the dry mass is weighed; only the payload
and ballast are not currently measured.
Bill
On Thu, Jul 11, 2019 at 6:59 PM Troy Prideaux <troy@xxxxxxxxxxxxxxxxxxxxx>
wrote:
Bill, fair enough. It sounds like a pretty tenuous chain of assumptions,
but it does shed light on the difference. I’m curious on how you arrived at
the assumption that the denser propellant would deliver a noticeable
increase in specific impulse? I can understand how that can occur in small
motors where you might expect noticeable deviations in pressure from
varying surface area exposure, but for a large motor?
Troy
*From:* arocket-bounce@xxxxxxxxxxxxx [mailto:arocket-bounce@xxxxxxxxxxxxx]
*On Behalf Of *William Claybaugh
*Sent:* Friday, 12 July 2019 10:14 AM
*To:* arocket@xxxxxxxxxxxxx
*Subject:* [AR] Re: Vacuum processing of solid propellant
Troy, David:
Let me be the first to agree that sims are just sims and reality awaits.
I too was surprised by the seemingly large change and have spent some time
trying to understand it:
First this is a two stage rocket and the effect of higher propellant
fraction is accordingly magnified in a way we don’t see in a single stage.
Second, the propellant is not just denser, it is also slightly higher
performance, a trajectory average of about four seconds for the first stage
and around seven seconds for the upper stage (which lights at around 40k
feet and burns out above 75k feet); this appears to be a significant second
order effect.
Lastly, the propellant is slightly slower burning (because of fewer
voids). When optimized w/ my 2-DOF sim the longer burn in the first stage
(10 seconds instead of eight) produces a much lower MaxQ which—in the
sim—produces a notably higher burnout velocity and altitude.
When the coast between burns is adjusted (was 21 seconds, now 12) because
of the higher and faster booster burnout the upper stage gains almost a
Mach number in burnout velocity (was Mach 5.2, now 6.1).
Then it gets tricky: the upper stage violates the two-caliber rule above
Mach 5.25. It looks like seven pounds of ballast behind the nose tip take
care of that but it does not look like the aluminum fins survive the ride.
I’m currently looking at titanium leading edges for the upper stage fins.
I don’t know if the sims are reliable (but the program has a long history
of accuracy); I’ll probably be flying one of the upper stages early next
year and then we’ll get a better idea of what to expect.
Let me also address a favorite topic, the cost benefit: this has cost
about $350 in out of pocket infrastructure investment (a vacuum tight lid
for the mixing bowl w/ view port). Everything else—including the vacuum
pump, gauges, and lines—was already in hand.
Bill
On Thu, Jul 11, 2019 at 5:35 PM David Summers <dvidsum@xxxxxxxxx> wrote:
That is fascinating! Is your propellant mass fraction unusually high? I
can't think of another reason for that result, and I'd like to understand
it better.
Thanks!
-David Summers
On Thu, Jul 11, 2019 at 1:15 PM William Claybaugh <wclaybaugh2@xxxxxxxxx>
wrote:
David:
In flight simulations of my two-stage, stuffing 3% more propellant into
the same volume raises the peak altitude from a tad over 150km to 210km in
the most recent runs.
Bill
On Thu, Jul 11, 2019 at 4:47 PM David Summers <dvidsum@xxxxxxxxx> wrote:
Hoping to take this on a learning tangent:
Why take all the effort to get a few percent higher density? I'd imagine
that it translates to less than 3% higher performance, so wouldn't a
simpler process with a few percent larger chamber be more optimal?
What am I missing? Or is it just a why not go for perfect kind of thing?
Thanks!
-David Summers
On Thu, Jul 11, 2019 at 12:41 PM Troy Prideaux <troy@xxxxxxxxxxxxxxxxxxxxx>
wrote:
If I vaguely recall (I could be mistaken) it was Mark Spiegl (?) who 1st
mentioned this process here about maybe 15 odd years ago. If I recall he
was utilising this process as an amateur and achieving some pretty good
results.
Troy
*From:* arocket-bounce@xxxxxxxxxxxxx [mailto:arocket-bounce@xxxxxxxxxxxxx]
*On Behalf Of *Anthony Cesaroni
*Sent:* Friday, 12 July 2019 6:31 AM
*To:* arocket@xxxxxxxxxxxxx
*Subject:* [AR] Re: Vacuum processing of solid propellant
Not with traditional solid constituents and configurations and I’ll leave
it at that.
One technique that does have some merit is centrifugal casting. In this
example you have more energy at your disposal to consolidate the solid
loading in the absence of proper vacuum processing equipment. The
propellant mix is prepared and post degassed using vacuum, turnover and
vibration as usual. The prepared propellant is then transferred into the
motor case or casting tube with a cap on one end. The case is then closed
with a cap on the remaining open end and the whole affair is spun,
longitudinally in a spin fixture or a lathe equipped with explosion proof
electrics (I’m sure). The case is spun at moderate speed until the
propellant cures.
After the propellant has cured, a section through the propellant will show
very high solid consolidation with a resin rich condition on the ID
surface. The next step is to use your explosion proof lathe to bore out the
excess resin on the ID and you will have an extraordinarily high solids
loaded propellant grain remaining. Some experimentation is required to
optimize the particle morphology and the speed should be optimized to
produce the most consolidation while minimizing the migration of the
smaller diameter particles.
To optimize the process further, one end of the case should be supported
by a roller steady and the dam on that end should have an opening to allow
the slow and well placed transfer of the propellant into the case while
it’s spinning instead of putting the whole mess in there at the beginning.
This actually works and you can achieve some impressive densities using
this method.
Anthony J. Cesaroni
President/CEO
Cesaroni Technology/Cesaroni Aerospace
http://www.cesaronitech.com/
(941) 360-3100 x101 Sarasota
(905) 887-2370 x222 Toronto
*From:* arocket-bounce@xxxxxxxxxxxxx <arocket-bounce@xxxxxxxxxxxxx> *On
Behalf Of *Edward Wranosky
*Sent:* Thursday, July 11, 2019 3:43 PM
*To:* arocket@xxxxxxxxxxxxx
*Subject:* [AR] Re: Vacuum processing of solid propellant
Is there any sort of VARTM like process for manufacturing propellant?
Edward
On Mon, Jul 8, 2019 at 8:39 AM William Claybaugh <wclaybaugh2@xxxxxxxxx>
wrote:
Uwe:
Yeah, as Anthony observed it is air trapped on the solids that produces
the problem.
The AP is tri-modal (200, 400, 600 micron) and mixed via a sieve on-site,
lots of opportunity for air to entrain in that process.
The Al is 5 micron and picks up air rather like a sponge. Wetting it and
degassing might help some but the subsequent final mixing w: the AP is
going to introduce air.
I’m thinking the strategy is going to be to post processing degas a set of
samples (5 minutes, 10, 15) and compare density gain per step. That should
offer a guess as to what can be achieved before the mix sets up.
Bill
On Mon, Jul 8, 2019 at 12:18 AM Uwe Klein <uwe@xxxxxxxxxxxxxxxxxxx> wrote:
Am 07.07.2019 um 19:59 schrieb William Claybaugh:
Uwe:
For safety we can’t mix the Al & AP together;
reminds me of the special 3 outlet can used for
chemically silvering mirrors back when. :-)
Hmm.
create separate airless pastes with each component as a first step?
Delayed wetting of solids always introduces air.
processing usuallythe
proceeds by mixing the liquids (including catalyst), adding the aluminum
to the liquid mix w/ careful hand stirring), then folding into the AP
followed by final mixing.
Bill
On Sun, Jul 7, 2019 at 11:37 AM Uwe Klein <uwe@xxxxxxxxxxxxxxxxxxx
<mailto:uwe@xxxxxxxxxxxxxxxxxxx>> wrote:
Am 07.07.2019 um 18:39 schrieb William Claybaugh:
> Does degassing the liquid components before mixing help or does
> subsequent mixing just reintroduce air?
Use mixing machinery where the mixing implements don't break the
surface. i.e. avoid "whipped cream" effects.
Any fine grained solids will need degassing after they
have been immersed in but not fully wetted by fluid components.
IMU no way around it.
Mix solids superficially, evacuate, add liquids on top.
return pressure to normal.
Then start mixing ( top mentioned prerequisites apply :-)
Uwe
