Author Topic: Neutron vs F9R and SS  (Read 59397 times)

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 9855
  • Everyplaceelse
  • Liked: 2292
  • Likes Given: 12743
Re: Neutron vs F9R and SS
« Reply #380 on: 09/29/2022 12:08 pm »
Well itís easy to say Neutron has less hardware problems when Neutron has zero hardware in test. A PowerPoint rocket will always be better.
Actually I was trying to get a handle on what issues are holding up SS flights. 

But Neutron can be expected to have less hardware problems because a)RL are not switching materials b)They are not doing any alloy development c)Their goal is somewhat less ambitious than full reusability.

Of course RL could make a complete hash of things and it doesn't see first launch before the latter half of the decade.  :( If I was to think about weak spots in the team it would be the shift from electric pump drive to ORSC. AFAIK turbines are a new thing for RL.

I look forward to a very interesting few years.  :)
MCT ITS BFR SS. The worlds first Methane fueled FFSC engined CFRP SS structure A380 sized aerospaceplane tail sitter capable of Earth & Mars atmospheric flight.First flight to Mars by end of 2022. Forward looking statements. T&C apply. "Extraordinary claims require extraordinary proof" R. Simberg."Competitve" means cheaper ¨cheap

Offline edzieba

  • Virtual Realist
  • Senior Member
  • *****
  • Posts: 4746
  • United Kingdom
  • Liked: 6876
  • Likes Given: 36
Re: Neutron vs F9R and SS
« Reply #381 on: 09/29/2022 12:26 pm »
Along with ORSC development (and if worst comes to worst, there are now methalox engines on the market or close to available on the market) there is stage construction: Electron is sufficient small that tanks can be continuously wound and the tanks and other structures (interstage, engine bay) bonded end-to-end such that thrust is pushing the assembly together, and tank pressure is not trying to force them apart. For Neutron, their moulds show they will be constructing the tanks in a segmented manner. That means bondlines in the pressure vessel walls themselves, which is nontrivial. Certainly not impossible - the infamous BFR test tank had a big circumferential joint and did not fail there when tested - but a new technique for Rocketlab.

Offline TrevorMonty

Re: Neutron vs F9R and SS
« Reply #382 on: 09/29/2022 03:35 pm »
Along with ORSC development (and if worst comes to worst, there are now methalox engines on the market or close to available on the market) there is stage construction: Electron is sufficient small that tanks can be continuously wound and the tanks and other structures (interstage, engine bay) bonded end-to-end such that thrust is pushing the assembly together, and tank pressure is not trying to force them apart. For Neutron, their moulds show they will be constructing the tanks in a segmented manner. That means bondlines in the pressure vessel walls themselves, which is nontrivial. Certainly not impossible - the infamous BFR test tank had a big circumferential joint and did not fail there when tested - but a new technique for Rocketlab.
Tanks can be  printed as one piece see link.

https://images.app.goo.gl/qRq86C9Yyz9DKXCo9


Offline edzieba

  • Virtual Realist
  • Senior Member
  • *****
  • Posts: 4746
  • United Kingdom
  • Liked: 6876
  • Likes Given: 36
Re: Neutron vs F9R and SS
« Reply #383 on: 09/29/2022 05:28 pm »
Along with ORSC development (and if worst comes to worst, there are now methalox engines on the market or close to available on the market) there is stage construction: Electron is sufficient small that tanks can be continuously wound and the tanks and other structures (interstage, engine bay) bonded end-to-end such that thrust is pushing the assembly together, and tank pressure is not trying to force them apart. For Neutron, their moulds show they will be constructing the tanks in a segmented manner. That means bondlines in the pressure vessel walls themselves, which is nontrivial. Certainly not impossible - the infamous BFR test tank had a big circumferential joint and did not fail there when tested - but a new technique for Rocketlab.
Tanks can be  printed as one piece see link.

https://images.app.goo.gl/qRq86C9Yyz9DKXCo9
I did not claim they could not be, but Rocketlab have already shown their moulds: they're sectional with internal layup, not monolithic with external layup.

Offline TrevorMonty

Re: Neutron vs F9R and SS
« Reply #384 on: 09/29/2022 06:36 pm »
There will still be lots of individual pieces that need moulding individually and bonded together, no different than aircraft like 787.

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 9855
  • Everyplaceelse
  • Liked: 2292
  • Likes Given: 12743
Re: Neutron vs F9R and SS
« Reply #385 on: 09/30/2022 08:32 am »
There will still be lots of individual pieces that need moulding individually and bonded together, no different than aircraft like 787.
True, but wasn't that part of why "Rosie" was such an innovation? Bonding them all on to the tank at the same time?

Beck did show the sectional tank molds at the presentation. This suggests it will be a 2 half tank sections . This is quite important as longitudinal joints are under 2x the stress of ring joints and that does look like major new territory for RL. Those joints will therefor need considerable care both in production and NDT.

Risk list so far
1) ORSC turbines
2) Tank longitudinal joints.
3) Ignition?

On the upside their GNC work with Electron should be fairly transferrable. Likewise all their knowledge on resins and fibers. Also stage separation.
MCT ITS BFR SS. The worlds first Methane fueled FFSC engined CFRP SS structure A380 sized aerospaceplane tail sitter capable of Earth & Mars atmospheric flight.First flight to Mars by end of 2022. Forward looking statements. T&C apply. "Extraordinary claims require extraordinary proof" R. Simberg."Competitve" means cheaper ¨cheap

Offline TrevorMonty

Re: Neutron vs F9R and SS
« Reply #386 on: 09/30/2022 09:06 am »


There will still be lots of individual pieces that need moulding individually and bonded together, no different than aircraft like 787.
True, but wasn't that part of why "Rosie" was such an innovation? Bonding them all on to the tank at the same time?



Rosie did drilling and machining of assembed tanks and fairings, more of glorified CNC machine.

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 9855
  • Everyplaceelse
  • Liked: 2292
  • Likes Given: 12743
Re: Neutron vs F9R and SS
« Reply #387 on: 09/30/2022 12:24 pm »
Rosie did drilling and machining of assembed tanks and fairings, more of glorified CNC machine.
Still a very good idea. Not something I think we'd seen WRT to CFRP. More a thing with metal tanks. I think it was Armadillo Aerospace who'd machine the tank end from a solid billet with the various flanges, test holes etc in place.

Still impressive. Doing all of those operations in one go.

I think this is a recurring issue people have with LV construction. They come up with some clever way to do the basic tank but completely ignore the fact that there is so much more to it. Mounting brackets, flanged connectors, thermometer pockets, stiffeners etc.  :( Without considering all of these items your new tank can take longer to construct (for the same size) than the design it replaces.
MCT ITS BFR SS. The worlds first Methane fueled FFSC engined CFRP SS structure A380 sized aerospaceplane tail sitter capable of Earth & Mars atmospheric flight.First flight to Mars by end of 2022. Forward looking statements. T&C apply. "Extraordinary claims require extraordinary proof" R. Simberg."Competitve" means cheaper ¨cheap

Offline trimeta

  • Full Member
  • ****
  • Posts: 1160
  • Kansas City, MO
  • Liked: 1577
  • Likes Given: 48
Re: Neutron vs F9R and SS
« Reply #388 on: 09/30/2022 02:12 pm »
I think this is a recurring issue people have with LV construction. They come up with some clever way to do the basic tank but completely ignore the fact that there is so much more to it. Mounting brackets, flanged connectors, thermometer pockets, stiffeners etc.  :( Without considering all of these items your new tank can take longer to construct (for the same size) than the design it replaces.
Of course, when Relativity makes this argument, everyone says they're dumb for using additive manufacturing on a tank, all a tank needs is sheet metal bent into a cylinder.

Offline TrevorMonty

Re: Neutron vs F9R and SS
« Reply #389 on: 09/30/2022 04:22 pm »
I think this is a recurring issue people have with LV construction. They come up with some clever way to do the basic tank but completely ignore the fact that there is so much more to it. Mounting brackets, flanged connectors, thermometer pockets, stiffeners etc.  :( Without considering all of these items your new tank can take longer to construct (for the same size) than the design it replaces.
Of course, when Relativity makes this argument, everyone says they're dumb for using additive manufacturing on a tank, all a tank needs is sheet metal bent into a cylinder.
Relatively tanks will still need drilling and some machining for engine bay assembly and  fitout.

Given low volume of Neutron boosters the fitout is likely to be labour intensive. Being reuseable extra build cost isn't big deal. Expendable US will benefit the most from robotic assembly and is where RL will invest most. Reduce US build cost enough and there is very little benefit to reuseable US.

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 9855
  • Everyplaceelse
  • Liked: 2292
  • Likes Given: 12743
Re: Neutron vs F9R and SS
« Reply #390 on: 10/01/2022 07:06 am »

Given low volume of Neutron boosters the fitout is likely to be labour intensive. Being reuseable extra build cost isn't big deal. Expendable US will benefit the most from robotic assembly and is where RL will invest most. Reduce US build cost enough and there is very little benefit to reuseable US.
Interesting point.  I'd picked up on this on the CNES/USAF concepts for a SABRE powered 1st stage but the point is applicable to all semi-reusable systems which are starting from a clean sheet, like Stoke Space for example.

The classic "cheap" second/upper stage is to use a solid, but in truth these tie you hand-and-foot to the solids supplier (unless you make your own) and their pricing.

For RL an obvious question would be if they can use the machinery for Electron to build the S2 as quickly and simply as possible.

Since the S2 is single use making it as small as possible (not necessarily light. I mean low volume) would lower mfg costs, so sub-cooling seems a good idea.

IMHO the other "big ticket" items are TVC and GNC. Historically engine gimballing TVC is a highly specialised, low volume field, calling for high power levels to cope with high thrust and fast response. Likewise fiber laser gyros are high cost items. So for real cost reductions you need to eliminate TVC and IMU based GNC.

 That means a) Having S1 deliver S2 to a very tight state vector in terms of velocity and altitude, or measure them accurately using on-board sensors post seperation (so on further forces applied to the S2) and upload it to S2. b) using sun/star/geomagnetic sensors on S2 to establish its altitude and velocity. RL have already said that Electron is good at delivering its payload to a tight orbital box and this is one of the reasons their customers come to them.

Without RCS or TVC the key parameter is engine start time and run duration. Work by the DLR during testing their facetted RCC TPS showed tighter control of even the start time of a solid S2 on a sounding rocket (coupled with extensive monte carlo preflight simulation)  could substantially improve mission success by narrowing the final burnout altitude and velocity parameters. With S1 being liquid fuelled the altitude/velocity parameters should be much tighter to begin with.

This can simplify the S2 control processor to basically running a series of counters and driving various control lines at different times, possibly running on the EMU for the engine itself. This would leverage a piece of hardware you're going to have on the stage anyway. Note one degree of freedom liquids give you is the ability to do multiple restarts.

Not having this ability in the Ariane 5 upper stage definitely limited their ability to dual launch commsats to different parts of geosynchronous orbit and was a required feature for Vinci.

If this seems like too austere a stage (engine, tanks, payload, frame) then cold gas and/or magnotorquers would be the obvious fallback for controlling the stage prior to engine ignition and ensuring precision attitude.
MCT ITS BFR SS. The worlds first Methane fueled FFSC engined CFRP SS structure A380 sized aerospaceplane tail sitter capable of Earth & Mars atmospheric flight.First flight to Mars by end of 2022. Forward looking statements. T&C apply. "Extraordinary claims require extraordinary proof" R. Simberg."Competitve" means cheaper ¨cheap

Offline TrevorMonty

Re: Neutron vs F9R and SS
« Reply #391 on: 10/01/2022 12:18 pm »
Electron low thrust Curie kick stage helps with accurate orbit insertion. Going lot harder with Neutron high thrust US unless they add some thrusters.

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 9855
  • Everyplaceelse
  • Liked: 2292
  • Likes Given: 12743
Re: Neutron vs F9R and SS
« Reply #392 on: 10/01/2022 01:35 pm »
Electron low thrust Curie kick stage helps with accurate orbit insertion. Going lot harder with Neutron high thrust US unless they add some thrusters.
That's one approach.  Another is to tighten up the ascent specs at S1 MECO. This lets you amortise the cost of any special sensors (or higher quality sensors) needed to give that accuracy.

[EDIT Reusable S1's totally change launch vehicle economics.

Historically you put all the GNC on the last stage, because that was going to clean up any failings of the lower stages velocity and altitude targets.

 But if the S1 is recoverable you can "spend" mass and money if it makes for a measurably cheaper S2, ideally by eliminating the standard IMU and TVC entirely,  but in a pinch going with cold gas thrusters and/or magnetorqurers for attitude control and star cameras for navigation.

If you have multiple S2 engines then you have the option of hard mounted engines steering by differential throttling (which AFAIK has never atually been tested in rockets, outside of possibly the DC-X). Bigger actuators than valves on cold gas canisters, but still smaller than gimbaling a whole engine.   ]
« Last Edit: 10/02/2022 09:22 am by john smith 19 »
MCT ITS BFR SS. The worlds first Methane fueled FFSC engined CFRP SS structure A380 sized aerospaceplane tail sitter capable of Earth & Mars atmospheric flight.First flight to Mars by end of 2022. Forward looking statements. T&C apply. "Extraordinary claims require extraordinary proof" R. Simberg."Competitve" means cheaper ¨cheap

Offline edzieba

  • Virtual Realist
  • Senior Member
  • *****
  • Posts: 4746
  • United Kingdom
  • Liked: 6876
  • Likes Given: 36
Re: Neutron vs F9R and SS
« Reply #393 on: 10/03/2022 10:30 am »
The core premise that avionics are too expensive and purely open-loop guidance would be an acceptable alternative may have been true a few decades ago, for fixed-nozzle solid stages.
But today, for a liquid propellant stage, the sensors required are not of extreme cost - particularly as you still need the rest of the avionics suite for open-loop control, engine management, TVC, and other stage functions such as spinup, spindown, and seperation - and even if you decide that ring-laser gyros are too pricey, then your fallback is not 'no IMUs' but use of MEMS IMUs.

You're giving yourself a lot of new headaches, making your vehicle less reliable (the first stage now needs to perform absolutely perfectly as your upper stage has no capability to compensate), for a very tiny cost saving.

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 9855
  • Everyplaceelse
  • Liked: 2292
  • Likes Given: 12743
Re: Neutron vs F9R and SS
« Reply #394 on: 10/03/2022 11:20 am »
The core premise that avionics are too expensive and purely open-loop guidance would be an acceptable alternative may have been true a few decades ago, for fixed-nozzle solid stages.
But today, for a liquid propellant stage, the sensors required are not of extreme cost - particularly as you still need the rest of the avionics suite for open-loop control, engine management, TVC, and other stage functions such as spinup, spindown, and seperation - and even if you decide that ring-laser gyros are too pricey, then your fallback is not 'no IMUs' but use of MEMS IMUs.

You're giving yourself a lot of new headaches, making your vehicle less reliable (the first stage now needs to perform absolutely perfectly as your upper stage has no capability to compensate), for a very tiny cost saving.
You might like to read what I wrote again. Slowly.  :(

Sensors are not a problem. If they come from established mass market products.
 Specialised sensors, such as IMU's that update fast enough to deal with orbital speeds are expensive and have to be thrown away.
Specialised actuators like the kind to gimbal a whole rocket engine (and the gimbals themselves) are also an issue.
Again it's about the economics of making the expendable components as cheap as possible, because you have to make them for every launch. If the baseline S2 can deliver the necessary accuracy without them then that's a clear win.  Photon would then be available if super-tight orbital parameters or additional delta v are needed.

You'll notice I called out 2 specific areas. IMU's and TVC. Those are the most specialized and highest power elements. They are expensive to make and every expendable stage (unless they are designed out) has to have a new set of them.

If you've got the luxury of a clean-sheet design (as RL have) it makes sense to look at all ways to lower recurring costs. The cheapest, lightest and most reliable parts are the ones the design does not include to begin with.

Time will tell how RL approach this question.
« Last Edit: 10/03/2022 11:25 am by john smith 19 »
MCT ITS BFR SS. The worlds first Methane fueled FFSC engined CFRP SS structure A380 sized aerospaceplane tail sitter capable of Earth & Mars atmospheric flight.First flight to Mars by end of 2022. Forward looking statements. T&C apply. "Extraordinary claims require extraordinary proof" R. Simberg."Competitve" means cheaper ¨cheap

Offline edzieba

  • Virtual Realist
  • Senior Member
  • *****
  • Posts: 4746
  • United Kingdom
  • Liked: 6876
  • Likes Given: 36
Re: Neutron vs F9R and SS
« Reply #395 on: 10/03/2022 12:28 pm »
Specialised sensors, such as IMU's that update fast enough to deal with orbital speeds are expensive and have to be thrown away.
These are neither as specialised nor as expensive as you believe.

If you insist on buying only 'space rated' components from companies that specialise in space hardware, they will happily charge you whatever they think they can. Rocketlab already don't do that. They already (through internal development and external acquisition) manufacture components for both their stages and spacecraft, and have successfully produced a three stage vehicle that throws away between 100% and 66% of it's avionics and still hits a price point well below Neutron's expected price point.
If Rocketlab can source avionics cheaply enough to hit Electron's target price, whilst throwing away double to triple what Neutron will, they will have no problem keeping upper stage costs down for Neutron without the enormous sacrifices to capability and reliability required with an unguided upper stage.

Online LouScheffer

  • Senior Member
  • *****
  • Posts: 3032
  • Liked: 5329
  • Likes Given: 670
Re: Neutron vs F9R and SS
« Reply #396 on: 10/04/2022 02:05 am »
Specialised sensors, such as IMU's that update fast enough to deal with orbital speeds are expensive and have to be thrown away.
These are neither as specialised nor as expensive as you believe.
You don't want cheap-but-crappy orbital injection.   The customer has to correct for large dispersions, which costs them money.  In return they will demand discounts that far exceed any money you might save on second stage cost.

So instead of getting rid of IMUs and closed loop control, instead you should figure out how to do them cheaply.  This should be entirely possible.  You can buy IMUs that are orders of magnitude cheaper than navigation-grade units.  These cheaper units have quick and high resolution response but bad drift characteristics.  You could combine this with GPS attitude determination, which has low angular resolution but no drift.  No one has tried this for rockets, to my knowledge, since even the pure INS is not that big of a fraction of second stage costs, and the technology is mature and reliable.  But if IMU costs start being important there's a lot of room for improvement.

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 9855
  • Everyplaceelse
  • Liked: 2292
  • Likes Given: 12743
Re: Neutron vs F9R and SS
« Reply #397 on: 10/04/2022 07:58 am »
Quote from: LouScheffer [/pre
link=topic=55413.msg2415336#msg2415336 date=1664849106]
You don't want cheap-but-crappy orbital injection.   The customer has to correct for large dispersions, which costs them money.  In return they will demand discounts that far exceed any money you might save on second stage cost.
Agreed. And as I noted the DLR were able to substantially improve the chances of hitting their targets with something as primitive as a 2 stage sounding rocket solely by controlling S2 start timing.

The historical approach has always been basically "get enough velocity more-or-less where it needs to be and we'll tighten up the parameters in the final stage." Splitting the errors, 50/50 between the usual two stages, or worse 70/30. IMHO this needs to be reconsidered. The optimisation parameters have changed.

The reason I'm pushing this subject is part of the thread title.  SpaceX has either succeeded in goals that it has set itself or replaced them with more ambitious ones. Given the critical importance to Musks goals we must assume that they will eventually achieve a fully reusable vehicle.

At this point a semi-reusable vehicle can have substantial cost reduction (to the company operating it) relative to a fully expendable vehicle.  And (in princple) the fully reusable vehicle can have substantial cost reductions relative to semi-reusable.

So if you know you're going to be  facing a fully reusable vehicle in the market, and you cannot or will not develop a fully reusable vehicle, but you're developing a clean-sheet semi-reusable launch vehicle, you should make every possible move to cut recuring costs.
Quote from: LouScheffer
These cheaper units have quick and high resolution response but bad drift characteristics.  You could combine this with GPS attitude determination, which has low angular resolution but no drift.
You've certainly cut the price of the IMU.

Now the question is what kind of GPS receiver do you need to do this?
Systems that can deliver updates at orbital rocket rates are also ITAR controlled. TBH I suspect a system that can achieve GPS lock while accelerating up to about 8000m/s (even if it only outputs an update at the standards once-a-second) is also ITAR controlled.

It looks like you've exchanged one military grade subsystem for another.  :(

Of course you could build a software defined radio, as INMOS engineer Philip Mateos did in the early 90's. Back then a processore with 5-10MIPS was seriously high performance, but not today. The key challenge (there were patents and I suspect it's explained in them) is to create a "synthetic Gold" code that can be tested agains all GPS satellites simultaneously which can then be refined to pull the datastream out of the signal. OTOH if you know where  you're launching from you could upload currently visible constellation from that location, along with local time, to accelerate lock.  But I digress.

I'm not against tighter control. I'm suggesting much more can be achieved without the need for the specialised sensors and actuators, or at least not mounting them on the expendable stage.
MCT ITS BFR SS. The worlds first Methane fueled FFSC engined CFRP SS structure A380 sized aerospaceplane tail sitter capable of Earth & Mars atmospheric flight.First flight to Mars by end of 2022. Forward looking statements. T&C apply. "Extraordinary claims require extraordinary proof" R. Simberg."Competitve" means cheaper ¨cheap

Offline Asteroza

  • Senior Member
  • *****
  • Posts: 2233
  • Liked: 791
  • Likes Given: 29
Re: Neutron vs F9R and SS
« Reply #398 on: 10/04/2022 10:40 am »
Quote from: LouScheffer [/pre
link=topic=55413.msg2415336#msg2415336 date=1664849106]
You don't want cheap-but-crappy orbital injection.   The customer has to correct for large dispersions, which costs them money.  In return they will demand discounts that far exceed any money you might save on second stage cost.
Agreed. And as I noted the DLR were able to substantially improve the chances of hitting their targets with something as primitive as a 2 stage sounding rocket solely by controlling S2 start timing.

The historical approach has always been basically "get enough velocity more-or-less where it needs to be and we'll tighten up the parameters in the final stage." Splitting the errors, 50/50 between the usual two stages, or worse 70/30. IMHO this needs to be reconsidered. The optimisation parameters have changed.

The reason I'm pushing this subject is part of the thread title.  SpaceX has either succeeded in goals that it has set itself or replaced them with more ambitious ones. Given the critical importance to Musks goals we must assume that they will eventually achieve a fully reusable vehicle.

At this point a semi-reusable vehicle can have substantial cost reduction (to the company operating it) relative to a fully expendable vehicle.  And (in princple) the fully reusable vehicle can have substantial cost reductions relative to semi-reusable.

So if you know you're going to be  facing a fully reusable vehicle in the market, and you cannot or will not develop a fully reusable vehicle, but you're developing a clean-sheet semi-reusable launch vehicle, you should make every possible move to cut recuring costs.
Quote from: LouScheffer
These cheaper units have quick and high resolution response but bad drift characteristics.  You could combine this with GPS attitude determination, which has low angular resolution but no drift.
You've certainly cut the price of the IMU.

Now the question is what kind of GPS receiver do you need to do this?
Systems that can deliver updates at orbital rocket rates are also ITAR controlled. TBH I suspect a system that can achieve GPS lock while accelerating up to about 8000m/s (even if it only outputs an update at the standards once-a-second) is also ITAR controlled.

It looks like you've exchanged one military grade subsystem for another.  :(

Of course you could build a software defined radio, as INMOS engineer Philip Mateos did in the early 90's. Back then a processore with 5-10MIPS was seriously high performance, but not today. The key challenge (there were patents and I suspect it's explained in them) is to create a "synthetic Gold" code that can be tested agains all GPS satellites simultaneously which can then be refined to pull the datastream out of the signal. OTOH if you know where  you're launching from you could upload currently visible constellation from that location, along with local time, to accelerate lock.  But I digress.

I'm not against tighter control. I'm suggesting much more can be achieved without the need for the specialised sensors and actuators, or at least not mounting them on the expendable stage.

the question becomes, the cost from the added weight of a COTS embedded board with SDR sufficient to roll your own high end GPS setup, versus buying an ITAR controlled COTS space rated GPS sensor. If the board can also take care of other upper stage management, then the mass hit becomes more of a sensor vs SDR thing.

ultimately you can't make your second stage too dumb, and there may be merit in using the same board with both stages.

Offline john smith 19

  • Senior Member
  • *****
  • Posts: 9855
  • Everyplaceelse
  • Liked: 2292
  • Likes Given: 12743
Re: Neutron vs F9R and SS
« Reply #399 on: 10/04/2022 11:31 am »
the question becomes, the cost from the added weight of a COTS embedded board with SDR sufficient to roll your own high end GPS setup, versus buying an ITAR controlled COTS space rated GPS sensor. If the board can also take care of other upper stage management, then the mass hit becomes more of a sensor vs SDR thing.

ultimately you can't make your second stage too dumb, and there may be merit in using the same board with both stages.
True. TBH the way processor capability has risen over the decades the days when the processor would be a significant weight (like the roughly 200lb of combined processor/IO processor of the shuttle GPC for a 400KIPS capability) are long gone. Including the industry fondness for putting stuff in metal boxes I'd expect you can put 200MIPS in a 1lb box.

Keep in mind once you've done all the paperwork for your ITAR controlled GPS sensor you have to do it all over again for your next launch, and the launch after that.  That's the cost of the hardware and the time spent on the paperwork. :(

I believe it's possible to fly a mission as accurately (with the same margins on altitude, velocity and attitude) without S2 IMU or TVC if the S1 parameters are tight enough and the S2 state vector is accurate enough, IE it's updated after the stages have separated based on sensor readings on S1.

However I recognise that is an ideal position that might not be posible IRL. AFAIK no one has actually tried to do this.  :(

What I'd like people to think about is wheather those highly expensive and specialized peripherals (IMU and TVC) are necessary, or would cheaper options deliver near equivalent capability?

 For example in the 60's the DoD developed a surveying system based on a mini computer and telescope for survey work that could locate the site to +/- 6m. In daylight. No GPS. Equivalent (but mostly superior) hardware is basically in every modern phone. Magetic compas sensors would be another option, again enabled by the vastly cheaper cost of processing their raw information.

On the TVC side it's the combination of speed X load that makes actuators heavy and power hungry. 
Historically people have gone with the gimbled single S2 engine. The gimbals and actuators have just been the "price of doing business"  :(
 But if you're going to stay with a single S2 engine how about getting it perfectly lined up before ignition? Given the stage has the high mass of the engine at one end, and the high mass of payload at the other the stage is likely to be quite well balanced, allowing cold gas thrusters (at both ends) to line up the burn quite tightly.
BTW astronavigation systems were the SoA in cancelling drift on IMU for ICBMs before GPS became available.
MCT ITS BFR SS. The worlds first Methane fueled FFSC engined CFRP SS structure A380 sized aerospaceplane tail sitter capable of Earth & Mars atmospheric flight.First flight to Mars by end of 2022. Forward looking statements. T&C apply. "Extraordinary claims require extraordinary proof" R. Simberg."Competitve" means cheaper ¨cheap

Tags:
 

Advertisement NovaTech
Advertisement SkyTale Software GmbH
Advertisement Northrop Grumman
Advertisement
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
1