Project Roskosmos "тайга" Lunar Lander

[SiberianTiger]

Any reasons for naming the lander "Taiga"? :hmm:

It's a very nice name! :thumbup: I guess it also sounds well in most languages (at least, Western ones).

 

[N_Molson]

Any reasons for naming the lander "Taiga"?

It is somehow in the line of "Angara" or "Dnepr" which are names of rivers, it seems to be the modern russian fashion in terms of nomenclatura. Also, "Taïga" is a class of vegetation, and the lander is a class of spaceship. Then each vessel of the serie could have specific names from space personalities like "Y. A. Gagarin", "S.P. Korolev", etc... Like each US LEM had a name : "Spider", "Eagle"...

 

[N_Molson]

Polished the "Killrot program" ; cleared bugs caused by time acceleration. Can now be only used when time accel = x1, enabling time acceleration disables "Killrot". Now it's a proper navigation tool to manoeuver the tug/lander assembly.

Also fixed a bug that spawned the tug adapter each time the engine was fired. Stupid variable typo.

 

[N_Molson]

I made a small modification of TransX that allows to perform precisely the first part of the lander stack TLI burn, which is a breakthrough.

Block-DM2 ignited (first orbit) :



Block-DM2 discarded :



The orbit is 115 x 3200 km at this point ; 700 out of the required 3200 m/s of delta-V have been achieved. It is time to set the lander in sleep mode and set the focus to the KVTK (oh erm... and to overwrite the TransX.dll with the standard one... annoying...)

TLI burn part 2 (second orbit) :



Still, the gain of efficiency in comparison to my previous attempts is impressive.

 

[N_Molson]

Finally got all the stuff in Lunar Orbit after 5 launches. Now the challenge is to align the planes of the 3 spacecrafts (Soyuz, Lander, Fregat).



First lunar dawn.



Separation and avoidance manoeuver.



SoyuzTMA-L in lunar orbit.

 

[Kelly Wright]

What is the link to the latest beta ? The last one in the thread seems to be dead....

 

[N_Molson]

Sorry, I didn't noticed that. Stay tuned for updates


I'm currently coding the Soyuz->Taïga crew transfer. It is a bit complex given that I want to implement a "health" system and track those values. But its going its way.

I managed to dock, make a crew transfer, and almost landed on the Moon. In fact, I would have, if I had not stupidly hit the "kill engines" key 40 meters above the surface. The worst moment to do that, I crashed before understanding what happened ! :facepalm:

But I had like 500 kilograms of fuel left at this moment, so the landing was really possible. I guess it demonstrates that the whole 5-launches-mission concept works

---------- Post added 05-15-12 at 02:23 AM ---------- Previous post was 05-14-12 at 12:48 PM ----------

I did it ! On lunar surface ! :bananadance:



I had 10 kg of propellant left, out of 5 tons. It was an Apollo11-like landing !

So, after 5 launches and 1 month of sim time, here we are. It works ! Still have to test the return trip, but its much simpler.

 

[Urwumpe]

Do you plan including a factory module or factory vessel in your add-on to produce the needed launch vehicles for every mission?

 

[N_Molson]

That's would not be a priority, but why not...

 

[Urwumpe]

That's would not be a priority, but why not...

Maybe you need assistance by Thorton there, regarding the Proton launcher, it is absolutely no priority, since it could open a pretty nasty can of worms.

A tool to automatize the scenario editor by Lua would be a good general option.

 

[sputnikshock]

Congrats for this project! Most exciting, most fantastic! Thank you very much.

However, we are facing one serious weak spot in the flightplan which could make the whole mission inherently unsafe. Namely, what happens if the TMA for one reason or another is unable to rendezvous or dock with the Fregat in LLO? This would inevitably lead to a crew stranded in lunar orbit waiting to die, wouldn't it?
Therefore, may I suggest mating the Fregat and TMA before TLI, so no matter what might go wrong, it would still be possible to perform TEI (or various other abort profiles on the way to the moon). Now docking the TMA to the Fregat 'nose on' makes it impossible to dock with the lander without undocking from the Fregat first, which doesn't really solve the problem. IMHO, the best solution would be to integrate the Soyuz and the Fregat in a fashion similar to this former ACTS proposal, but with a standard orbital module. So something along the lines of this:

This would of course render the KTDU-80 Soyuz engine useless, so it could be taken out, which would save another 310kg. (http://www.astronautix.com/engines/ktdu80.htm).
I would keep the Soyuz fuel tank for RCS thrusters, and since Fregat and Soyuz use the same fuel, it should be possible to feed propellant from the Soyuz to the Fregat engine (as was done with Fregat-SB where extra fuel was added in a torodial drop tank).
After doing a bit of research and math I came up with the following figures for the Soyuz+Fregat spacecraft:

Dry mass:
6340kg (Soyuz) - 310kg (KTDU-80) + 100kg (lunar return heat shield) + 930kg (Fregat) = 7060kg

Propellant:
880kg (Soyuz) + 5250kg (Fregat) = 6130kg

Fuelled mass:
13190kg

With an ISP of 333sec this leads to a spacecraft delta-V of 2041m/s.
(All specs for Fregat taken from here: http://www.laspace.ru/rus/fregat_construction.php).
In other words, if you assume a hybrid trajectory to the moon with 15m/s for MCCs, 880m/s for LOI and 925m/s for TEI - a total of 1820m/s - you will have a pretty comfortable dV safety margin, which could be traded for up to 1157kg of extra payload/equipment if needed!!

So far so good, but now two new problems arise:

1.
Instead of sending the TMA and a Fregat to LLO separately, you now would have to inject 13 tons in one go. According to the weight figures presented here I calculated that the KVTK in development for Angara-A5 would produce a dV of 3517m/s with a 13190kg payload. This should be fine for TLI. Of course since the 23.5 ton KVTK would need to be lofted by a Proton, it remains to be checked if this is possible or if it needs to do a small orbital insertion burn. One way or the other I think it's not unrealistic.

2.
The 13 ton Soyuz+Fregat spacecraft needs to be launched into LEO itself. This is no big deal with Proton or Zenit, however it would be a non man-rated launch. Sending another Soyuz to ferry the crew is inacceptable in terms of mission planning IMHO, and also adds considerable cost to the mission.
Here is my proposal:
Fill the (unmanned) Soyuz+Fregat spacecraft with dry cargo and launch it to the ISS, thus replacing a regular Progress resupply launch (and thus refinance a bit of the lunar mission). The crew would have spent some time on the ISS as part of a regular expedition already, and instead of going straight back home, would add a lunar excursion to their mission. Even better, the Soyuz+Fregat could remain docked to the ISS for quite some time which adds additional scheduling flexibility to the lunar mission.
Now, a typical flight ot the ISS needs around 110-120m/s of dV. Let's assume one would need the same amount of propellant to return to a lower orbit and rendezvous with the Proton launched TLI KVTK. And this happens to be exactly the excess dV calculated earlier. After docking with the TLI stage Soyuz+Fregat would still have the 1820m/s dV available which is needed for MCC+LOI+TLI! Edit: Typo, should read: TEI
Even better, spacecraft mass would be down to 12310kg by now. For which the KVTK could now produce a comfortable 3649m/s TLI.


I thought about this for quite a while and finally decided to propose it. I hope I didn't get my math wrong somewhere on the way. And I hope even more that you don't regard this as a criticism of your work. As I said, I think you are doing an oustanding job on an outstanding project, which will be great as is. I just thought I might share my ideas for discussion.

Best regards and keep up the good work
sputnikshock

 

[N_Molson]

Your ideas are indeed interesting, but there are some points that also raise potential problems, I think. Also I thought about this mission for more than two years now, and I'm confident I've found a solution that is an acceptable tradeoff between safety, costs, schedule and complexity. :tiphat:

Namely, what happens if the TMA for one reason or another is unable to rendezvous or dock with the Fregat in LLO? This would inevitably lead to a crew stranded in lunar orbit waiting to die, wouldn't it?

No. You forgot the Taïga. It has more than enough Dv to rendez-vous with the Soyuz (via a computer program which can be adjusted from the ground when the spacecraft passes on the Earth side), dock, perform the TEI, and guide the crippled spacecraft to the reentry corridor (call this a "LOR Apollo 13 scenario"). In the Lunar Orbital only scenario, I think it is safe to have a Proton ready in Baïkonur to launch another Fregat, in case of. Proton rocket has the advantage to use a non-cryogenic fuel, which allows more flexibility in terms of readiness (and if it isn't used, it will roll back to the MIK and be refitted for the next commercial satellite launch).

This would of course render the KTDU-80 Soyuz engine useless, so it could be taken out, which would save another 310kg.

Not so idea I think, remember that the Soyuz can jettison the BO and get in a "Zond" config, which gives more Dv. Very useful for adjusting the reentry corridor, and a safeguard if something goes wrong at this point.

I would keep the Soyuz fuel tank for RCS thrusters, and since Fregat and Soyuz use the same fuel, it should be possible to feed propellant from the Soyuz to the Fregat engine (as was done with Fregat-SB where extra fuel was added in a torodial drop tank).

This would require a major modification of the Soyuz-TMA-M, a proven man-rated spacecraft. A new propulsion system would require a serie of reliability tests (to be sure that the crossfeed thing won't leak, fail or just blow up in real space conditions).

Instead of sending the TMA and a Fregat to LLO separately, you now would have to inject 13 tons in one go. According to the weight figures presented here I calculated that the KVTK in development for Angara-A5 would produce a dV of 3517m/s with a 13190kg payload. This should be fine for TLI. Of course since the 23.5 ton KVTK would need to be lofted by a Proton, it remains to be checked if this is possible or if it needs to do a small orbital insertion burn. One way or the other I think it's not unrealistic.

Actually I already use the KVTK. An orbital insertion burn is indeed required (the Proton-M can loft around 21.5 tons in LEO at 51.6° (the only launch corridor available)), which leaves us with roughly 3200 m/s of Dv, just enough for the TLI.

Fill the (unmanned) Soyuz+Fregat spacecraft with dry cargo and launch it to the ISS, thus replacing a regular Progress resupply launch (and thus refinance a bit of the lunar mission). The crew would have spent some time on the ISS as part of a regular expedition already, and instead of going straight back home, would add a lunar excursion to their mission. Even better, the Soyuz+Fregat could remain docked to the ISS for quite some time which adds additional scheduling flexibility to the lunar mission.

This adds complexity to an already complex mission plan. Getting a launch window to the Moon isn't easy from the 51.6° inclination. Plus this involves the ISS, which means again a serie of test-flights for the modified spacecraft (see what happens with the Dragon).

After docking with the TLI stage Soyuz+Fregat would still have the 1820m/s dV available which is needed for MCC+LOI+TLI!

I guess you mean MCC+LOI+TEI. Another problem is that you need extra Dv for Lunar Orbit manoeuvers. There are the docking manoeuvers themselves, of course, but also plane alignements. Because performing the TLI so that a spacecraft enters LLO at the same Inclination than one of the targets isn't enough, time and synchronization matters. With the tools we have, I think it's pretty impossible to make such a complex calculation (especially when 3 spacecrafts are involved). So we have to deal with some LLO plane changes (which require much less Dv than in LEO).

---------- Post added at 11:19 AM ---------- Previous post was at 10:46 AM ----------

Wrote again the depressurization / repressurization code. More flexible now.

A VC or a panel will soon become a priority ; as there are more and more shortcuts and more and more values to track. Without gauges, labels & buttons, it is a little complicated.

Update of the to-do list & priorities on the first post.

 

[sputnikshock]

Thank you very much for your reply. I know you have been thinking about this missions for a very long time (as I have been following your thread) and please don't get me wrong I do not see anything negative in your plan at all. Just think that I have a few ideas that could maybe contribute to making this whole enterprise even better. Also, I know how it is when some smart%$$ jumps on something one has spent a long time thinking about... Therefore I don't want to argue or start a lenghty discussion. Nevertheless, I do want to throw in some more figures/facts/thoughts one last time so all my arguments are on the table and you can judge whether or not to consider it for your project. No matter how decide, I am looking forward to flying your add-on and mission. Actually I can't wait! :thumbup:

You forgot the Taïga.

Indeed I did! :facepalm: This is good news, my worst fears about a stranded crew are resolved. Still I think an integrated Soyuz+Fregat s/c could offer many advantadges for the mission.

remember that the Soyuz can jettison the BO and get in a "Zond" config, which gives more Dv. Very useful for adjusting the reentry corridor, and a safeguard if something goes wrong at this point.

Considering what happened with TM-5 I think this can only be regarded as a last ditch effort - after TEI and/or just before reentry. Anyway dropping the BO for more dV would be possible in an integrated Soyuz+Fregat config as well.

This would require a major modification of the Soyuz-TMA-M, a proven man-rated spacecraft. A new propulsion system would require a serie of reliability tests (to be sure that the crossfeed thing won't leak, fail or just blow up in real space conditions).

Well, the step from 7K-T to Soyuz-T and again from Soyuz-T to TM involved engine changes. One unmanned and one manned testflight was performed when these were introduced. The second testflights were done as parts of nominal missions IIRC. Anyway, one could do most of the testing through modified Progress or pseudo-Progress flights (and still get something useful out of these missions).

Alternatively one could drop the KDTU-80, keep the tanks partially fuelled (pretty much as in Apollo Skylab) with RCS propellant, and use the larger 1050kg dry Fregat config = no need to crossfeed, no need for testing a modified engine. Instead: Proven S/C minus engine, plus proven bus.

This would lead to:

Dry S/C weight: 7180kg
20% Souyz propellant: 176kg (= 40m/s of RCS dV at full gross weight!)
7100kg Fregat propellant
Fuelled weight: 14456kg

After going to the ISS and back to KVTK rendezvous, fuelled weight would be
13512kg. dV available: 2065m/s. Which means an extra 245m/s dV for LLO maneuvers.

All this is on the conservative side of things.
For example potential weight savings due to removal of redundant RCS thrusters and piping either on the PAO or Fregat - and the fact that Fregat guidance avionics and comm equipment are not needed - are not considered yet.

Actually I already use the KVTK. An orbital insertion burn is indeed required (the Proton-M can loft around 21.5 tons in LEO at 51.6° (the only launch corridor available)), which leaves us with roughly 3200 m/s of Dv, just enough for the TLI.

Well, I can't argue about this as I have not flown a KVTK in orbiter yet. But the official Khrunichev data for the stage says:
Gross weight: 23530kg
Fuel weight: 19600kg
Final weight: 3330kg

Now there's a 600kg discrepancy of course, as 23530-19600=3930. Drop-away ullage motors? I don't know, but will assume the higher dry weight of 3930kg.
RD-0146 ISP is reported to be 470sec.

With these figures a partially fuelled KVTK with a full weight of 21.5tons (= no insertion burn needed) still yields 3212m/s dV with the 13512kg payload!
(With a 12310kg payload - as in my original calculation; Soyuz->Fregat crossfeed config - it would be 3380m/s.)

An ISP of 470 seems very high. If calculated with a more conservative 461 (as quoted in many sources) we get:
21.5 ton KVTK with 13512kg payload -> 3150m/s
21.5 ton KVTK with 12310kg payload -> 3315m/s

This adds complexity to an already complex mission plan. Getting a launch window to the Moon isn't easy from the 51.6° inclination.

I actually believe the complexity added by an ISS stopover is a good thing in terms of mission planning, as it reduces overall mission complexity.
1. You win time. If any of the other launches go wrong, get delayed etc. you have a crew and s/c in orbit with plenty of time buffer (months) which adds alot of flexibility.
2. Even if you had to abandon the lunar mission entirely, your crew and s/c would have contributed to another mission (ISS) which turns a full failure into a partial failure and limits your losses cost-wise.

In addition mission complexity is further reduced by using the Soyuz+Fregat integrated s/c.
3. It eliminates the TEI-Fregat TLI and LOI
4. It eliminates the LLO rendezvous maneuver between Soyuz and the TEI-Fregat
5. It eliminates the need to modifiy the TEI-Fregat for long-duration LLO stay (i.e. solar panels)
6. It adds safety by incorporating the possiblity of post-TLI abort profiles for the manned flight

Here is a comparison of potential flight plans. I hope I didn't mix up your info from post #187.

not necessary Launch 1 : Soyuz-FG / TMA to ISS

Launch 1 : Proton-M / KVTK / Fregat not necessary
-> TLI not necessary
-> LOI (72 hours later, Fregat in LLO) not necessary not necessary Launch 2 : Proton-M / Soyuz-Fregat cargo flight to ISS

Launch 2 : Proton-M / Fregat / Lander (1-3 months later) Launch 3 : Proton-M / Fregat / Lander (1-3 months later)
Launch 3 : Proton-M / KVTK#1 (in the following 24 hours) Launch 4 : Proton-M / KVTK#1 (in the following 24 hours)
-> Docking between the Lunar Lander and the KVTK#1 -> Docking between the Lunar Lander and the KVTK#1
-> TLI -> TLI
-> LOI (72 hours later, Lunar Lander in LLO) -> LOI (72 hours later, Lunar Lander in LLO)

Launch 4 : Soyuz-FG / TMA not necessary Launch 5 : Proton-M / KVTK#2 (in the next 24 hours) Launch 5 : Proton-M / KVTK#2
-> Rendez-vous TMA with the KVTK#2 -> Rendez-vous Soyuz-Fregat with the KVTK#2
->TLI ->TLI
-> LOI (72 hours later) -> LOI (72 hours later)
-> TMA / Lander rendez-vous -> Soyuz-Fregat / Lander rendez-vous
-> Lunar Landing -> Lunar Landing
-> TMA / Lander rendez-vous -> Syouz-Fregat / Lander rendez-vous
-> TMA / Fregat rendez-vous (still orbiting in LLO) not necessary
-> TEI -> TEI
-> Recovery/Landing -> Recovery/Landing

Launches shared with other missions: 0 Launches shared with other missions: 2
Lunar landing mission launches: 5 Lunar landing mission launches: 3
Proton-M launches: 4 Proton-M launches: 4
Soyuz-FG launches: 1 Soyuz-FG launches: 1

Total launches: 5 Total launches: 5

Best regards
sputnikshock

 

[N_Molson]

I also don't think that the crew should spend a long duration aboard the ISS before going for the Lunar Landing. There will have a lot of work to do there, maybe including EVAs, and won't be in perfect condition prior to the Lunar Landing Mission. Even worse, it will be one month or so since they won't have trained for the Lunar Mission, which I consider as an absolute no-go. Cosmonauts don't improvise unless something goes wrong, they follow a very tightly planned script.

Also, the Fregat has already been modified for long durations. While I know it is not a proof of reliability (to say the least), the Fregat designed for Phobos-Grunt should have been able to survive a 6-months-long trip to Mars.

Concerning the figures, the KVTK and the Fregat we will be using have extra dead mass, that accounts for the docking apparatus. And, when I simulated the mission, I quickly discovered that the rendez-vous have to be taken into account (sync orbits, small orbital plane corrections).

You can find the figures used by Thorton on this post. I trust him very much, as he seems to have access to information (including schematics...) that cannot be found on the Web...

 

[sputnikshock]

I rechecked my numbers and need to make a few corrections.
1. I based my calculations on a Soyuz TMA. The TMA-M model - which would be used - is of course 70kg lighter.
2. According to International Launch Service's official Proton Mission Planner's Guide (http://www.ilslaunch.com/sites/default/files/pdf/PMPG Section 2.pdf)
Proton-M payload to 180km LEO orbit is infact 23t. :thumbup:

Thanks for the link to Thorton's KVTK figures. I have no clue what his sources are, but I am convinced they are very good. Recalculated with his numbers, and it still works fine. Also did a calculation with a TMA-M without any modifications except the heatshield (that is, with KDTU-80) mated to a Fregat (heavy version). So two flightproven off-the-shelf s/c (again except for the heatshield of course). With 540kg fuel in the Soyuz tanks it still works fine.

As far as LLO maneuvering fuel is concerned, depending on s/c configuration, up to 380m/s would be available. Remember, only 2 instead of 3 R&D maneuvers are necessary in LLO when using the integrated Fregat solution.

So in the end it comes down to wether or not one considers an ISS stopover as an advandtage, disadvantadge, or even a no-go. Since you made your position on this very clear I give up...

 

[N_Molson]

1. I based my calculations on a Soyuz TMA. The TMA-M model - which would be used - is of course 70kg lighter.

We estimated the thicker heat shield to additional 250-300 kg. So final mass of the Soyuz is ~= 7500 kg (the R-7 has enough power to lift it to LEO).

2. According to International Launch Service's official Proton Mission Planner's Guide (http://www.ilslaunch.com/sites/defau...ection 2.pdf)
Proton-M payload to 180km LEO orbit is infact 23t.

I noticed that figure too, but I think it includes a full Block-DM or Briz-M burn, which we can't do there (that would be too heavy (nearly 40 tons !) associated with the KVTK).

As far as LLO maneuvering fuel is concerned, depending on s/c configuration, up to 380m/s would be available. Remember, only 2 instead of 3 R&D maneuvers are necessary in LLO when using the integrated Fregat solution.

I think that it would be advisable to modify the number/power of the Soyuz-Fregat thrusters if an approach to the ISS is planned. Else, the "braking power" will become dangerousely low. Again, flight testing. Also, the whole set of translation (and rotation) RCS has to be moved, because the CoG would significantly shift backwards.

So in the end it comes down to wether or not one considers an ISS stopover as an advandtage, disadvantadge, or even a no-go.

On a realistic point of view, I'm convinced that the training problem and crew fatigue is a major issue. Crews always train until the last day before any mission. And staying 1 month in the ISS, while it is a tremendous experience, is exhausting for the human body. Astronauts/Cosmonauts often resort to medication because of headaches and other little health problems caused by the lack of gravity... The muscles that usually help you to walk become quickly lazy, which is a problem when it comes to a lunar EVA, especially since we are using very heavy and quite "rigid" Orlan suits.

Of course, those issues would concern any long duration manned mission. If we are ever going to Mars, spaceflight will have to deal with that (with maybe a lander simulator in the mothership, and probably a spinning part of the hull or similar centrifugal device that can generate artificial gravity, to keep the crew in full health).

 

[sputnikshock]

We estimated the thicker heat shield to additional 250-300 kg

That's interesting. Maybe could you elaborate a bit on how this is calculated/estimated?

I am only asking because the Zond reentry capsule (7K-L1 SA), which was basically a 7K-OK SA modified with an improved heatshield for lunar return, was reported to have been just 65kg heavier than than a 7K-OK SA. (2700 vs 2765kg).

Or does the 250-300kg figure already include other things like enhanced lunar comm gear, consumables etc.?

I noticed that figure too, but I think it includes a full Block-DM or Briz-M burn, which we can't do there (that would be too heavy (nearly 40 tons !) associated with the KVTK).

No. It clearly says 23tons to 180km is for a 3-stage Proton-M. No Briz or DM involved.

 

[N_Molson]

That's interesting. Maybe could you elaborate a bit on how this is calculated/estimated?

I am only asking because the Zond reentry capsule (7K-L1 SA), which was basically a 7K-OK SA modified with an improved heatshield for lunar return, was reported to have been just 65kg heavier than than a 7K-OK SA. (2700 vs 2765kg).

Or does the 250-300kg figure already include other things like enhanced lunar comm gear, consumables etc.?

The Zond was basically stripped down of everything that wasn't absolutely necessary. Also remember it was designed for 2 cosmonauts, and the 7K-OK for 3.

And anyway, according to Thorton's figures, the KVTK weights 24 tons (not taking into account the docking apparatus), so even 23 tons is too low.

 

[sputnikshock]

The Zond was basically stripped down of everything that wasn't absolutely necessary. Also remember it was designed for 2 cosmonauts, and the 7K-OK for 3.

Agreed.
To make this crystal clear: I do not intend to suggest that 65kg would be the number or even close (it feels way too light to me). Neither do I inted to suggest that 250-300kg is wrong or unrealistic. To be honest, I simply do not know. But again I am asking how this number came about as I'd like to get a better understanding of it. I assume some kind of calculation, rule-of-thumb, comparison figures etc must have been involved.

And anyway, according to Thorton's figures, the KVTK weights 24 tons (not taking into account the docking apparatus), so even 23 tons is too low.

Obviously. But I don't see your point. Since KVTKs are needed for your mission they need to be brought to orbit somehow. Which leaves two choices:

a) do an orbital insertion burn with the KVTK
b) launch it partially fuelled, to a gross weight of 23000kg

For b) and with Thorton's numbers I get:

Gross weight in LEO: 23000kg
Dry weight: 4400kg
Propellant: 18600kg
Isp: 463

Which yields a maximum dV of 7509m/s.
In other words, a 3150m/s TLI would allow a payload of 14177kg
Subtract your additional dead mass on the KVTK - how much mass did you calculate for your docking collar? 250kg? This would still yield a TLI payload of 13927kg.

 

[N_Molson]

a) do an orbital insertion burn with the KVTK
b) launch it partially fuelled, to a gross weight of 23000kg

Believe me that the 2 options have been thoroughly tested, and that a) is the right choice.