-
#1340
by
JH
on 08 Apr, 2018 23:06
-
Maybe Rocketlabs might get into the small upper stage business? I.e. making a small upper stage like the Fregat with low thrust requirements that is compatible with most launchers and supplied by the launch customer rather than the launch company.
Leave cheap launch of multi-ton payloads to the big players and have them develop full reusability for their launchers. Use their existence to sell your own expendable product. It could allow players like SpaceX or Blue Origin to do two-stage with full reusability more easily, and give satellites direct injection capability into high orbits for just a couple million.
I was thinking of something similar, actually.
If fully reusable SHLV's become a reality and their costs are even close to current projections, Rocket Lab might be better served by commercializing their Curie third stage, or a derivative of it, rather than trying to sell an expendable small satellite launch vehicle with with a payload specific price hundreds of times higher.
Affordable third stages from a third party, like Rocket Lab, could solve the "last mile problem" of launching smallsats and cubesats in bulk, while preserving much of the benefits of larger, reusable launch vehicles.
-
#1341
by
john smith 19
on 09 Apr, 2018 06:58
-
If fully reusable SHLV's become a reality and their costs are even close to current projections, Rocket Lab might be better served by commercializing their Curie third stage, or a derivative of it, rather than trying to sell an expendable small satellite launch vehicle with with a payload specific price hundreds of times higher.
Affordable third stages from a third party, like Rocket Lab, could solve the "last mile problem" of launching smallsats and cubesats in bulk, while preserving much of the benefits of larger, reusable launch vehicles.
RL may have hit a particularly useful set of features for US use. Their battery electric pump technology is the only kind that (could) be recharged over time with a PV array (I'd suggest body mounted with the stage given a slow spin to avoid hot spots). LOX tanks can be pressurized by heating up some LOX and either battery or nozzle cooling tapoff can generate that heat. The only joker is the RP1 tank. This is
not Integrated Vehicle Fluids as such, but it's a quite a nice (possible) synergy.
Now if the US were to be switched to a cryogenic fuel as well...
Obviously RL are a
long way from this. I've no idea what sort of hit you take going from primary to secondary battery technology. Quite a lot I'd guess. But if you can manage LOX boiloff then anything but LH2 is not going to be too difficult for you. Now "all" you have to do is figure out how to refuel it on orbit. So you'll be needing a set of RCS thrusters as well.
-
#1342
by
speedevil
on 09 Apr, 2018 08:18
-
Obviously RL are a long way from this. I've no idea what sort of hit you take going from primary to secondary battery technology. Quite a lot I'd guess. But if you can manage LOX boiloff then anything but LH2 is not going to be too difficult for you. Now "all" you have to do is figure out how to refuel it on orbit. So you'll be needing a set of RCS thrusters as well.
They almost certainly are already using secondary batteries, in this case, there are no suitable primary batteries at least at their current rates.
For long durations, thermal managment is going to be hard - methane/oxygen might significantly ease it over either hydrogen/oxygen or kerosene/oxygen.
Hydrogen is both underdense, and very, very cold, meaning your insulation gets lots more complex, as well as the pump being very different. Methane is a comparable density to kerosene, and a comparable temperature to LOX.
For body-scale solar panels, recharging the batteries is likely to take several hours per minute of thrust.
-
#1343
by
JH
on 09 Apr, 2018 15:41
-
They've said that their third stage uses a "green monopropellant", which almost certainly means VLM (viscous liquid monopropellant), which they developed under a DARPA contract around five years ago. It's a high density solid when at rest, but becomes liquid when agitated or subjected to force (this allows for starting and stopping the motor, unlike traditional solids). They've stated that it has comparable performance to traditional solid propellants, so probably in the low to mid 200 second range.
-
#1344
by
Gliderflyer
on 09 Apr, 2018 16:35
-
They've said that their third stage uses a "green monopropellant", which almost certainly means VLM (viscous liquid monopropellant), which they developed under a DARPA contract around five years ago. It's a high density solid when at rest, but becomes liquid when agitated or subjected to force (this allows for starting and stopping the motor, unlike traditional solids). They've stated that it has comparable performance to traditional solid propellants, so probably in the low to mid 200 second range.
In the recent Reddit ama, Peter said that the kick stage doesn't use the VLM.
-
#1345
by
Davidthefat
on 09 Apr, 2018 16:45
-
They've said that their third stage uses a "green monopropellant", which almost certainly means VLM (viscous liquid monopropellant), which they developed under a DARPA contract around five years ago. It's a high density solid when at rest, but becomes liquid when agitated or subjected to force (this allows for starting and stopping the motor, unlike traditional solids). They've stated that it has comparable performance to traditional solid propellants, so probably in the low to mid 200 second range.
In the recent Reddit ama, Peter said that the kick stage doesn't use the VLM.
Also, I'm not sure how "green" VLM is; not a chemist.
-
#1346
by
JH
on 09 Apr, 2018 17:53
-
Also, I'm not sure how "green" VLM is; not a chemist.
It was described as non-toxic, which is pretty green for monopropellant.
In the recent Reddit ama, Peter said that the kick stage doesn't use the VLM.
Interesting, I thought I had seen all of the responses in that AMA. That does beg the question of what the Curie stage is running on, as it is described as using a "green monopropellant". Perhaps a derivative of VLM, which is technically not VLM?
-
#1347
by
john smith 19
on 09 Apr, 2018 18:11
-
Also, I'm not sure how "green" VLM is; not a chemist.
It was described as non-toxic, which is pretty green for monopropellant.
In the recent Reddit ama, Peter said that the kick stage doesn't use the VLM.
Interesting, I thought I had seen all of the responses in that AMA. That does beg the question of what the Curie stage is running on, as it is described as using a "green monopropellant". Perhaps a derivative of VLM, which is technically not VLM?
Since the name itself ("Viscous Liquid Monopropellant") doesn't actually name the chemical it could be pretty much anything to begin with.
The
obvious storable, non toxic liquid mon and bipropellant is actually Hydrogen Peroxide in high concentration. Tested for at least 6 years storage on orbit
if in properly engineered systems. Catalytic ignition available. Usable in either form and breaks down to O2 and Water, or superheated steam, so able to ignite most conventional fuels.
-
#1348
by
ringsider
on 09 Apr, 2018 18:21
-
No need guess about VLM, the patent is out there:
https://patents.google.com/patent/US20120234196(Link edited.)
The VLM mixture of claim 1 further comprising any combination of:
a. a high-energy oxidiser additive or substitute or combination thereof present in an amount ranging from less than 1% to about 90% by mass,
b. a thermic agent or combination of thermic agents present in an amount ranging from less than 1% to about 25% by mass,
c. a surfactant or combination of surfactants present in an amount ranging from less than 1% to about 20% by mass,
d. a bonding agent or combination of bonding agents present in an amount ranging from less than 1% to about 10% by mass,
e. a rheology modifier or combination of rheology modifiers present in an amount ranging from less than 1% to about 10% by mass,
f. a burn rate modifier or combination of burn rate modifiers present in an amount ranging from less than 1% to about 10% by mass, or
g. a dissolved gas within the carrier fluid to improve dispersion, atomisation and combustion present in an amount ranging from less than 1% to about 25% by massAn embodiment of this concept was prepared for use in a VLM rocket motor. The formulation was comprised of the following constituents, and was mixed in the following order:
# Compound/ingredient Proportion (%)
1 Polydimethylsiloxane (Carrier-fluid) 28.4%
2 Castor Oil Derivative (thickener) 0.6%
3 Amidomethicone (Bonding Agent) 1%
4 Aluminium powder, spherical, 20 μm, (Thermic Agent) 10%
5 Ammonium perchlorate, spherical, 200 μm (Oxidant) 60.0%
-
#1349
by
Steven Pietrobon
on 10 Apr, 2018 03:53
-
-
#1350
by
Kabloona
on 10 Apr, 2018 04:37
-
No need guess about VLM, the patent is out there:
https://worldwide.espacenet.com/publicationDetails/originalDocument?CC=US&NR=2012234196A1&KC=A1&FT=D&ND=3&date=20120920&DB=EPODOC&locale=en_EP#
That just gives me blank pages. I tried searching for US2012234196, but couldn't find anything. I can read the claims though. Here's the combination given in the patent. Uses the same fuel and oxidiser as your standard solid propellant, but with different additives.
1 Polydimethylsiloxane (Carrier-fluid) 28.4%
2 Castor Oil Derivative (thickener) 0.6%
3 Amidomethicone (Bonding Agent) 1%
4 Aluminium powder, spherical, 20 μm (Thermic Agent) 10%
5 Ammonium perchlorate, spherical, 200 μm (Oxidant) 60.0%
This link is easier to read the patent in, and it lists quite a broad range of VLM propellant ingredients and percentages, so apparently quite a range of ingredients and combinations are covered by the patent.
https://patents.google.com/patent/US20120234196The particular combination listed above, though, is quite similar to a typical AP/HTPB based solid propellant mix in its pre-cured state, ie 70% solids (as compared to low to mid 80's achievable in a similar solid, IIRC) and the liquid carrier agent replacing the HTPB binder (which starts out as a liquid in the mixing/casting process before curing). So it's quite like a solid propellant in its liquid pre-cured state. Just with a slightly lower solids loading to improve flowability, I would guess.
-
#1351
by
john smith 19
on 10 Apr, 2018 06:12
-
That just gives me blank pages. I tried searching for US2012234196, but couldn't find anything. I can read the claims though. Here's the combination given in the patent. Uses the same fuel and oxidiser as your standard solid propellant, but with different additives.
1 Polydimethylsiloxane (Carrier-fluid) 28.4%
2 Castor Oil Derivative (thickener) 0.6%
3 Amidomethicone (Bonding Agent) 1%
4 Aluminium powder, spherical, 20 μm (Thermic Agent) 10%
5 Ammonium perchlorate, spherical, 200 μm (Oxidant) 60.0%
So although it's technically a liquid it's got an Isp like a solid?
I think perspective is quite important here. IIRC monoprops get around 220s but solids go 250-260s so while that's not great as a liquid that's
excellent compared to most monoprop propellants.
-
#1352
by
Kabloona
on 10 Apr, 2018 13:55
-
So although it's technically a liquid it's got an Isp like a solid?
The Isp will approach a comparable solid, but can't quite get there because the solids loading can't be as high, so they can't add the last, say, 5% more aluminum and 10% more AP that a solid would have, which are the most energetic components. So, as a rough estimate, you could expect an energy deficit of 15-20% or so as compared to a solid with similar composition (for the AP/Aluminum VLM formulation listed above).
The resulting reduced solids loading required for flowability leaves you with a slightly "watered down" version of an uncured solid propellant mix that, instead of using R-45 monomer (the HTPB precursor) which is quite viscous, you may be able to use (I would guess) a less viscous carrier fluid that flows more readily.
-
#1353
by
john smith 19
on 10 Apr, 2018 17:28
-
So although it's technically a liquid it's got an Isp like a solid?
The Isp will approach a comparable solid, but can't quite get there because the solids loading can't be as high, so they can't add the last, say, 5% more aluminum and 10% more AP that a solid would have, which are the most energetic components. So, as a rough estimate, you could expect an energy deficit of 15-20% or so as compared to a solid with similar composition (for the AP/Aluminum VLM formulation listed above).
The resulting reduced solids loading required for flowability leaves you with a slightly "watered down" version of an uncured solid propellant mix that, instead of using R-45 monomer (the HTPB precursor) which is quite viscous, you may be able to use (I would guess) a less viscous carrier fluid that flows more readily.
I guess it depends what you want. Pressure fed hypergols are (IIRC) good for about 330x. OTOH they are more complex, need a pressurizer (although I think the French have done work with GN2 in the props, like 2 fizzy soda cans) and can have slosh concerns at larger sizes.
The usual rule of rocket design lore is the higher the stage the higher the Isp you want. ButBlack Arrow used the "Waxwing" solid to get something like 3-4000m/s of its needed delta v.
TBH I'm surprised no one has had a go at doing a small pumped US using the sort of technology John Whitehead's team developed at LLNL.
-
#1354
by
Asteroza
on 12 Apr, 2018 07:15
-
So although it's technically a liquid it's got an Isp like a solid?
The Isp will approach a comparable solid, but can't quite get there because the solids loading can't be as high, so they can't add the last, say, 5% more aluminum and 10% more AP that a solid would have, which are the most energetic components. So, as a rough estimate, you could expect an energy deficit of 15-20% or so as compared to a solid with similar composition (for the AP/Aluminum VLM formulation listed above).
The resulting reduced solids loading required for flowability leaves you with a slightly "watered down" version of an uncured solid propellant mix that, instead of using R-45 monomer (the HTPB precursor) which is quite viscous, you may be able to use (I would guess) a less viscous carrier fluid that flows more readily.
I guess it depends what you want. Pressure fed hypergols are (IIRC) good for about 330x. OTOH they are more complex, need a pressurizer (although I think the French have done work with GN2 in the props, like 2 fizzy soda cans) and can have slosh concerns at larger sizes.
The usual rule of rocket design lore is the higher the stage the higher the Isp you want. ButBlack Arrow used the "Waxwing" solid to get something like 3-4000m/s of its needed delta v.
TBH I'm surprised no one has had a go at doing a small pumped US using the sort of technology John Whitehead's team developed at LLNL.
You mean a non electric pumped bipropellant US, or more specifically a (electric?) reciprocating pump bipropellant US?
-
#1355
by
Katana
on 13 Apr, 2018 05:27
-
No need guess about VLM, the patent is out there:
https://worldwide.espacenet.com/publicationDetails/originalDocument?CC=US&NR=2012234196A1&KC=A1&FT=D&ND=3&date=20120920&DB=EPODOC&locale=en_EP#
That just gives me blank pages. I tried searching for US2012234196, but couldn't find anything. I can read the claims though. Here's the combination given in the patent. Uses the same fuel and oxidiser as your standard solid propellant, but with different additives.
1 Polydimethylsiloxane (Carrier-fluid) 28.4%
2 Castor Oil Derivative (thickener) 0.6%
3 Amidomethicone (Bonding Agent) 1%
4 Aluminium powder, spherical, 20 μm (Thermic Agent) 10%
5 Ammonium perchlorate, spherical, 200 μm (Oxidant) 60.0%
200um solid particles are terrible to valves. Slurry propellants are experimental since 1940s but not operational up to now.
Operational green monopropellants (aside of HTP) are usually ADN or HAN based solutions, e.g. AF-M315E.
-
#1356
by
CameronD
on 13 Apr, 2018 07:40
-
200um solid particles are terrible to valves. .....
Not necessarily, depending entirely on flow rate and construction material. At fairly low flow rates (less than tens of meters/sec, whatever that may be in your language, and then some) most stainless steels will cope long enough for the associated rocket motor to run for however long it was designed to run for. How long is that? Minutes??
-
#1357
by
Katana
on 13 Apr, 2018 13:06
-
200um solid particles are terrible to valves. .....
Not necessarily, depending entirely on flow rate and construction material. At fairly low flow rates (less than tens of meters/sec, whatever that may be in your language, and then some) most stainless steels will cope long enough for the associated rocket motor to run for however long it was designed to run for. How long is that? Minutes??
When valves are shut off after opening, particles could be trapped between seal surfaces and cause leak.
Especially for low thrust engines, the whole valve orifice diameter is less than 5mm...
Compared to outstanding performance of AF-M315E, it's not worth the trouble.
-
#1358
by
TrevorMonty
on 13 Apr, 2018 17:34
-
-
#1359
by
john smith 19
on 13 Apr, 2018 18:44
-
You mean a non electric pumped bipropellant US, or more specifically a (electric?) reciprocating pump bipropellant US?
I was thinking of a GG driven reciprocating pump but, now you mention it, if you've got the battery technology a simple reciprocating pump (which seems to be a better fit for systems < 5000lbs of thrust) would be a good choice.
One thing I've not seen considered would be a reciprocating pump with 2 chambers driven by a single solenoid, possibly as a resonant system. I think modern power electronics is up to the job.