So, I ask here: Can an H2O2/Methanol rocket achieve a specific
impulse of 300?
Mike Miller, Materials Engineer
-- Joe M.
"Mike Miller" <cra...@hotmail.com> wrote in message news:5dcb47db.03011...@posting.google.com...
> So, I ask here: Can an H2O2/Methanol rocket achieve a specific
> impulse of 300?
Yep. A quick PROPEP run (w/1500 psi chamber) yields the follwing Isp's
at an O/F of 766/233:
Sea level optimum expansion: 248 seconds
Vacuum (area ratio 100): 325 seconds
--
Scott Lowther, Engineer
"Any statement by Edward Wright that starts with 'You seem to think
that...' is wrong. Always. It's a law of Usenet, like Godwin's."
- Jorge R. Frank, 11 Nov 2002
98% peroxide with an alcohol can just barely hit 300 Isp in a vacuum
with a >100x expansion ratio nozzle and basically perfect combustion.
H2O2/kerosene is somewhat prefered from a safety standpoint, because a
spill of H2O2/alcohol will mix together into a sensitive high
explosive, while a spill of H2O2/kerosene will stay separated. It
will likely still start a fire, which can vaporize things enough to
make something explosive, but it is still a safer mix to have.
In more practical engines, we saw 200s Isp in ground tests on our
50lbf 90% peroxide / kerosene engine with a 250 psi supply pressure.
Moderate scaling, tuning, and pressure increases can get that up
another 10%-20%, but I doubt a peroxide engine has been fired with a
demonstrated SL Isp over 250.
John Carmack
www.armadilloaerospace.com
> Yep. A quick PROPEP run (w/1500 psi chamber) yields the follwing Isp's
> at an O/F of 766/233:
> Sea level optimum expansion: 248 seconds
> Vacuum (area ratio 100): 325 seconds
Cool, thanks for the info (all respondants).
Is 1500psi a bit "high strung" for an RCS and OMS? How
closely related is vacuum Isp to chamber pressure?
Mike Miller, Materials Engineer
I assume the chamber temp would be comparatively low, which might be a help in
some cases.
--
http://inquisitor.i.am/ | mailto:inqui...@i.am | Ian Stirling.
---------------------------+-------------------------+--------------------------
<Squawk> Pieces of eight!
<Squawk> Pieces of eight!
<Squawk> Pieces of eight!
<Squawk> Pieces of eight!
<Squawk> Pieces of eight!
<Squawk> Pieces of nine!
<SYSTEM HALTED: parroty error!>
It's awfully high for a pressure-fed system in particular. You'd need
pumps for that, almost certainly, and that's a complication people usually
prefer to avoid for RCS/OMS systems. (It's been done, mind you. The
Agena RCS used electric pumps, albeit at rather lower pressure.)
>How closely related is vacuum Isp to chamber pressure?
To a good first approximation, given high-expansion nozzles for efficient
operation in vacuum, it's not.
Higher chamber pressure suppresses dissociation, and gives more compact
(not lighter, but smaller) engines for a given amount of thrust, and
greatly improves Isp when operating against back pressure (i.e. in
atmosphere) which limits nozzle expansion. Only the first of those is an
issue for vacuum-only operation, and it's not a big issue.
As an example, the vacuum Isp of the RL10 is within a few seconds of that
of the SSME, despite an order-of-magnitude difference in chamber pressure.
--
Faster, better, cheaper requires leadership, | Henry Spencer
not just management. | he...@spsystems.net
Compared to LOX/hydrocarbon, H2O2 combusted with pretty much anythign is
low temp. A possible issue is the rather high O/F of H2O2 systems,
meaning relatively low fuel flow for the thrust produced; it's better to
use fuel for regen cooling than H2O2.
> Higher chamber pressure suppresses dissociation, and gives more compact
> (not lighter, but smaller) engines for a given amount of thrust, and
> greatly improves Isp when operating against back pressure (i.e. in
> atmosphere) which limits nozzle expansion.
OTOH, this propellant combination is going to be comparatively
cool, so dissociation shouldn't be a big problem anyway.
Paul
Beal's peroxide/kerosene third stage for the BA-2 was intended
to have a 300 Isp. Did they do any testings on that, or were the
engineers being optimistic?
Mike Miller, Materials Engineer
Actually, peroxide is an *excellent* coolant, and regen-cooled engines
using it are usually peroxide-cooled. Yes, it has an upper temperature
limit which must be respected, but that's true of any coolant. And it has
a very high specific heat; it can soak up a lot of heat within that limit.
As for Isp, there is a learning still between theoretical numbers and
a practical engine. Pressure drops and pumping losses would have
the catalyst section operating at 1745 psia. As an example, for a
5000 lb thrust Propargyl Alcohol / Peroxide engine it would require
a pump output of 140 horsepower. Efficiencies will drive that up
further for the turbine. The combustion area would more resemble
a gas-turbine combustor than the typical rocket and that would
require further study to reduce noise, even at the cost of thrust.
(Here I am assuming that a 5000 lb thrust demonstrator is
sufficiently large to be taken seriously, scaling is still an issue).
Still the construction and development would a sizeable task not in
keeping with the big dumb booster concept, but still a direction I
feel should be investigated. A 300 Isp seems unlikely.
--
Anvil*
In general, I defer to Beal's engineers in areas of their experience
(Mark Henry has been directly helpful to us on a couple occasions),
but 300 Isp does smack of a
nice-round-number-that-might-maybe-be-possible. They were using 92%
peroxide, and you don't usually make relatively low pressure ablative
nozzles with gigantic expansion ratios, so I am a little dubious. If
there are any Beal engineers lurking, I would love to hear about any
vacuum chamber tests that were done...
John Carmack
www.armadilloaerospace.com
> Compared to LOX/hydrocarbon, H2O2 combusted with pretty much anythign is
> low temp. A possible issue is the rather high O/F of H2O2 systems,
> meaning relatively low fuel flow for the thrust produced; it's better to
> use fuel for regen cooling than H2O2.
Why do you say that it is better to use fuel? H2O2 is a *GREAT*
coolant, with properties similar to water, and you have a lot more
mass of it than you would have of any fuel in any other combination.
You can't heat it nearly as hot as you might heat a fuel, but that is
more than balanced by the higher specific heat and mass flow.
We had a 50lbf regen engine that could run steady state with a rich
fuel mixture, and we have a 1000lbf regen engine mostly completed that
should run steady state at optimal mixture, but it has been sidelined
while we focus on X-Prize issues for the time being.
John Carmack
www.armadilloaerospace.com
This being a third stage means that it is vacuum Isp. Actually 300
seconds is a pretty modest goal (relatively speaking, of course).
Oren
Perahps I should have said "use fuel for regen cooling than oxidizers."
Just in general, as fuels are FAR less likely to react with the metal
walls or decompose than most oxidizers. However, I stand corrected on
the H2O2 as coolant bit.
Unless, of course, they happen to be hydrazine-based...
Or acetylene.
> We had a 50lbf regen engine that could run steady state with a rich
> fuel mixture, and we have a 1000lbf regen engine mostly completed that
> should run steady state at optimal mixture, but it has been sidelined
> while we focus on X-Prize issues for the time being.
Just a quick question. My understanding is that HTP has
real issues with most materials. If you used it in a regen
loop, would that mean that you could only use pure aluminum
or 306 Stainless steel? Just curious. Also, how reactive
is HTP with say a glass/epoxy or glass/polyester composite?
I know it gets along fine with Polyethylene, but I was curious
how well (or poorly) it works with composites.
Might it be possible to use a copper thrust chamber regen
cooled with HTP, and just include a pressure relief valve
on it for any trapped HTP?
Jon Goff
Try H2O2 with some of the metal hydrides in either a gel or Hybrid design.
Vacum ISPs greater than Lox/Hydrogen are possible with MUCH higher
Propellant densities. Try running AlH3 or LiH3 thru ProPEP. Then if your
feeling really spunky you can start to play with the zip fuels(boron based).
Nasa Lewis did some studies on AlH3 suspended in various hydrocarbons burned
with H2O2 also.
The air force built and tested some small peroxide/zip fuel motors back in
the 60's but I was only ever able to dig up general info on them back when I
was in the propellant grade H2O2 business.
Of all the stuff that I personally experimented with, I think that biprop
motors with H2O2 burning a hydrocarbon/AlH3 is probably the best performance
system that can be easily built. Personally, I found the Alcohol family of
hydrocarbons to give the smoothest combustion with H2O2. Isopropyl was my
favorite although there are several higher energy alcohols that I never got
to experiment with. When working with alcohol/H2O2 mixtures it is important
to make sure that all of the Tin has been removed from the H2O2. Tin is
normally added to H2O2 as a stabilizer. If a mixture of H2O2/alcohol is left
to stand, which has tin, it will self detonate at some point. Also, you can
get significantly smoother combustion of H2O2 with Hydrocarbon&Metal Hydride
fuel if the metal used to make the hyride is of a type called
"Electro-exploded". Electro-Explosion uses strong electrical curents applied
to the smallest commercially available metal particles causing then to
explode into even smaller particles. There use to be one company that made
electro exploded metals and their electro exploded Aluminum was studied as
an upgrade to adding regular Al powder to the solid feul mixture of the
SRB's.
As Always,
Jay
"Anvil" <r...@ms.spacebbs.com> wrote in message
news:5b3c90be.03012...@posting.google.com...
There are different degrees of catalytic activity in metals. There are
some metals you don't want to use in contact with HTP at all, but there
are plenty more which are unsuited for lengthy HTP storage but just fine
for brief contact in a rocket engine.
>Might it be possible to use a copper thrust chamber regen
>cooled with HTP, and just include a pressure relief valve
>on it for any trapped HTP?
If memory serves, copper is one of the "no contact ever" metals. However,
you could plate the cooling passages with something more benign. (People
have experimented with plated cooling passages for other reasons; it works
fine.)
> There are different degrees of catalytic activity in metals. There are
> some metals you don't want to use in contact with HTP at all, but there
> are plenty more which are unsuited for lengthy HTP storage but just fine
> for brief contact in a rocket engine.
Do you have a few examples of the sorts of metal that are
tolerable for use in an HTP rocket engine? I know that
pure aluminum and 306 stainless are pretty good for the
long haul, but is there some material that gets decently
high thermoconductivity, decent strength, and that is
decently tolerable to HTP?
> If memory serves, copper is one of the "no contact ever" metals.
Yeah, that's what I had heard too.
> However, you could plate the cooling passages with something more
> benign. (People have experimented with plated cooling passages for
> other reasons; it works fine.)
Problem is that I'm trying to work on a method for very low cost
fabrication of rocket engines, and anything that involves plating
or coating usually won't fit the bill. Of course, I could be wrong,
maybe an electroplating option might be worth looking into for HTP
engines. Hmmm...I'll have to look into that.
Thanks.
Jon
The inner wall of the early British peroxide/kerosene engines was
nickel-plated low-carbon steel. (They might have preferred chromium
plating, but chromium poisons typical HTP catalysts.)
Later on they went to tube-wall construction, using some unremarkable
stainless steel or other (but they had to weld joints, rather than brazing
them, because the usual brazing alloys are "no contact ever").
>I know that
>pure aluminum and 306 stainless are pretty good for the
>long haul, but is there some material that gets decently
>high thermoconductivity, decent strength, and that is
>decently tolerable to HTP?
If (dim) memory serves, most aluminum alloys which do *not* contain copper
or manganese are okay for brief contact. With copper itself ground-ruled
out, that is about the only obvious choice that has really high thermal
conductivity -- most metals don't. The one wart of aluminum alloys is the
relatively low melting point.
Since phosphoric acid is used as stabilizer for H2O2 wonder about phosphate
coating of metal surface in contact with it. Similar to method used for
corrosion resistance on guns. Figure tanks can be made of aluminium with
parts requiring heat resistance of stainless steel (low manganese).
> The inner wall of the early British peroxide/kerosene engines was
> nickel-plated low-carbon steel. (They might have preferred chromium
> plating, but chromium poisons typical HTP catalysts.)
Hmm...I think I'll look into nickel coated copper. Just make
sure that the nickel is thick enough to keep the HTP separated
from the copper. IIRC nickel can be electroplated (which is a
lot cheaper AFAICT than CVD or IVD methods).
> Later on they went to tube-wall construction, using some unremarkable
> stainless steel or other (but they had to weld joints, rather than brazing
> them, because the usual brazing alloys are "no contact ever").
If you went with a channel wall technique, could you get away
with copper as the main structure, but each channel coated
with nickel?
> If (dim) memory serves, most aluminum alloys which do *not* contain copper
> or manganese are okay for brief contact.
Yeah, that's more or less what I read (also aluminum alloys
with iron in them aren't so great either).
> With copper itself ground-ruled out, that is about the
> only obvious choice that has really high thermal conductivity
> -- most metals don't. The one wart of aluminum alloys is the
> relatively low melting point.
Yeah, which is a rather substantial wart. Could you really make
a coating thick enough on copper to allow it to be used for the
main chamber material?
Jon
How about a little tri-mix of CO/O2/H2O2
what's the best impulse?
What's the best mix?
What happens if the chamber pressure is raised significantly, like for
example operating at 3000 psi?
Regards, Brad Guth / IEIS http://guthvenus.tripod.com
It sounds plausible, if done carefully.
>> ...The one wart of aluminum alloys is the relatively low melting point.
>
>Yeah, which is a rather substantial wart.
They've got some substantial advantages, though. Armadillo Aerospace is
using aluminum in its peroxide and peroxide/kerosene engines successfully.
(You can gain a bit with an anodized layer on the inside, as thermal
insulation.)
>Could you really make
>a coating thick enough on copper to allow it to be used for the
>main chamber material?
I expect it's possible. Whether it is practical and cost-effective is
harder to assess.