Kerbfleet: A Jool Odyssey-END OF CHAPTER 21! (and hopefully not so many talking heads in 22!)

[Mister Dilsby]

1 hour ago, Aperture Science said:

@Kuzzter

Please. We need a "Probes Interlude". A special episode dedicated to The Probes.

I'm not saying I will... but if I did, I know exactly how I'm going to do it. I even have a song picked out. 

59 minutes ago, Foxbat said:

Now I don't feel quite so inept landing my replica Derpstar!  I call mine Freightliner since I'm partial to that brand of big rigs.  Still have some tweaking to do but having a full sized cargo bay in a relatively compact form is a far sight better than my previous design.

Oh no, what have I done! Foxbat, there are so many better cargo SSTOs out there than the Derpstar! When my kids see it they both laugh and go "Derp derp derp". But I'm glad it worked for you  Maybe the Mark II will be more successful.

[Guest]

27 minutes ago, Kuzzter said:

I'm not saying I will... but if I did, I know exactly how I'm going to do it. I even have a song picked out. 

PLEASE KUZZTER PLEASE

I beg you!

Edited February 26, 2016 by Guest

[Starhawk]

54 minutes ago, Kuzzter said:

I'm not saying I will... but if I did,

Part of your post says 'Maybe..'.

52 minutes ago, Kuzzter said:

I know exactly how I'm going to do it. I even have a song picked out. 

And part of your post says 'Yes I will!'

Happy Concerned landings!

[Mister Dilsby]

8 hours ago, Starhawk said:

Part of your post says 'Maybe..'.

And part of your post says 'Yes I will!'

I probably will. I have a concept for it but it can't be executed for some time, because the concept involves things that can only happen after the Intrepid arrives in the Jool system. So, remember that time when everyone was going CROSSOVER CROSSOVER DO A CROSSOVER! and I said SHADDAP ALREADY I HEARD YOU! and then everyone did because you're all wonderful and then a couple of months later Kappa Jeb showed up and we were all like OHMAGERD CROSSOVER!? Yeah. But then there was also the time when people were going DRES DRES DRES LOVE WE DEMANDS ZEE DRES! and I was like SHADDAP ALREADY MY COUNSEL WILL I KEEP! and we still haven't been to Dres. But then again I never said I had a concept for a cool Dres story so read into that whatever your tinfoil-shielded brains tell you

Speaking of probes, I did test @Rune's concept for the Scansats and I like it:

This succeeded (barely) in my "A" criterion of being able to go from a 500km orbit around Jool to a polar orbit around Tylo in its own power. A design with fewer tanks can meet the less difficult criterion "B" which is to go from a 100km equatorial orbit around Tylo to a polar one. 

Also speaking of probes, while I was landing the Derpstar and doing Tedus's EVA inside the hangar deck to take those screenshots last update, I kind of forgot about the Tugbot:

Whoops. Fortunately the little guy has plenty of mono and its own RTG, we'll get that stowed properly in the aft ramp before we do anything else. Am also playing with some designs for Derpstar Two (The Second Derping) to try and cut the number of flights needed from three to two. But it would be worth three if I also bring a cute little rover for Sarge to drive on Bop, right? Stay tuned  

[Geschosskopf]

11 hours ago, Kuzzter said:

Agree completely. Just tested the new scansats, and converted the Jooldivers to mono and tested them as well. Simpler and better all around, so that's what will go up in Samantha's next Derpstar flight!

Just be sure to lock off the probe mono tanks once they're docked, or the main ship might drain them during maneuvers en route.

[Mister Dilsby]

4 minutes ago, Geschosskopf said:

Just be sure to lock off the probe mono tanks once they're docked, or the main ship might drain them during maneuvers en route.

Yep. All probes and boats have an AG0 to disable their RCS, which we trigger during the docking maneuver. Intrepid actually doesn't use any RCS at all--her OMS is on vernors. 

Edited February 26, 2016 by Kuzzter

[Geschosskopf]

24 minutes ago, Kuzzter said:

This succeeded (barely) in my "A" criterion of being able to go from a 500km orbit around Jool to a polar orbit around Tylo in its own power. A design with fewer tanks can meet the less difficult criterion "B" which is to go from a 100km equatorial orbit around Tylo to a polar one. 

Why hold onto the probes so long?  Instead, release them in stages during Intrepid's capture burn.  First Intrepid burns just enough to capture with its Ap out in the vicinity of Pol and Bop.  Pause the burn and quickly release any probes going to those places.  Start burning again until the Ap is just a bit beyond Tylo.  Pause the burn, release the Tylo probes.  Continue the burn until Intrepid's Ap is set for Vall or Laythe, wherever it's going first.

By doing things this way, the probes don't have to burn to raise their Aps to their ultimate destinations, and Intrepid doesn't have to burn extra fuel dragging their Aps down beforehand, so a double savings.  Thus, each probe will only need a small-medium burn to hit its target, then a very small burn to tweak the encounter to come in already on a polar orbit.  Thus, the only big burn the Tylo probe would do is to capture itself at Tylo.

I would have suggested releasing the probes a bit before or just after entering Jool's SOI.  Jool's SOI is so big it takes months to get from the edge to a low Jool Pe, so any tweaks done to the probes at this point would space their Jool Pe passes out an hour or so at least, so no problems.  But then each probe would need to do its own capture burn and I doubt they have the fuel for that.

See, I'm trying to convert your mothership into a flotilla

 

[Rune]

59 minutes ago, Kuzzter said:

Speaking of probes, I did test @Rune's concept for the Scansats and I like it:

This succeeded (barely) in my "A" criterion of being able to go from a 500km orbit around Jool to a polar orbit around Tylo in its own power. A design with fewer tanks can meet the less difficult criterion "B" which is to go from a 100km equatorial orbit around Tylo to a polar one.

Great to be of help! Now let's minmax this sucker:

-RCS ports. Quads have a weight similar to the linear ones, and you only need four to get full three-axis rotation and translation. You save 160kg of empty weight that way, If I assume correctly that you are using twelve linears there..

-Electrical system: running on RTGs, you not only don't need the battery, you only need the one to keep the probe happy. You save another 130kg of empty weight that way.

-Main engines. The O-10s are 90kg each, but their increased Isp (+10s) if probably worth it with such a high mass ratio. Maaaybe you could use just one, but that means cheaty clipping and finding out that their thrust vector is slightly inclined by default. Probably not worth it.

-Docking port: too late now, right? Still, a small one would have netted you another 30kgs in weight reductions, if you are up for an Intrepid "redesign". You could do it with KAS, just bolting a Jr. docking port over the old one...

In a big ship, those things are rounding errors, but here we are talking about an initial empty weight only a little over 1mT! With those changes, you get an empty weight about ~300kg less, which is something more than 20% of your mass. Meaning, 4/5ths the fuel will give you the same dV. And of course, as @Geschosskopf says, fancy navigating can decrease the dV requirements quite a bit... There, I removed a tank, at least.

 

Rune. The lands around Jool are ripe for gravity assists!

Edited February 26, 2016 by Rune

[Geschosskopf]

37 minutes ago, Rune said:

-Electrical system: running on RTGs, you not only don't need the battery, you only need the one to keep the probe happy. You save another 130kg of empty weight that way.

Well, you do need enough battery to transmit the science data without interruption.  The actual SCANsat biome and altimetry maps aren't big things (about 500-1000 should do it), but if @Kuzzter hasn't disabled stock resource scanning in SCANsat, the M700 will need about 2500-3000 EC in the battery to do its survey.

[Commander Zoom]

56 minutes ago, Geschosskopf said:

Why hold onto the probes so long?  Instead, release them in stages during Intrepid's capture burn.

Was going to say, this assumes a capture burn... and then I remembered that, post 1.0, aerobraking at Jool from interplanetary speeds is a much more problematic thing.  Bye bye to all that "free" (firey-hot) delta-V at the arrival end.

During my expedition, back in the Before Time, I held onto my probes until after the first braking/capture pass, so that they wouldn't need to burn to keep from flying back out of the system entirely.  May or may not be relevant now.

Edited February 26, 2016 by Commander Zoom

[Mister Dilsby]

50 minutes ago, Rune said:

Great to be of help! Now let's minmax this sucker: [snip]

-RCS ports. Quads have a weight similar to the linear ones, and you only need four to get full three-axis rotation and translation. You save 160kg of empty weight that way, If I assume correctly that you are using twelve linears there..

-Electrical system: running on RTGs, you not only don't need the battery, you only need the one to keep the probe happy. You save another 130kg of empty weight that way.

-Main engines. The O-10s are 90kg each, but their increased Isp (+10s) if probably worth it with such a high mass ratio. Maaaybe you could use just one, but that means cheaty clipping and finding out that their thrust vector is slightly inclined by default. Probably not worth it.

-Docking port: too late now, right? Still, a small one would have netted you another 30kgs in weight reductions, if you are up for an Intrepid "redesign". You could do it with KAS, just bolting a Jr. docking port over the old one...

Really good points--let's see: I think where I ended up is 10 total linears, not 12: if I miss the bay going in, I can just puff the main engines to back out, right? But yeah,4xblocks might be more optimal I actually did to that on the Jooldipper probes (not pictured) and it works perfectly. However, I might just like the look of the linear ports better. Maybe I can get the number down to 6, I have plenty of torque installed to handle the imbalance of the single off-center RTG.  

Regarding the electrical system--the reason for the battery is to have enough charge to transmit the scan. in the Bill Space Program test just concluded, about 700 charge (net) was consumed to get the scan picture. So I think that battery mass may be necessary.

And regarding O-10s etc: no, I don't think I want to go with a single off-center one. TWR is already pretty low, and I don't want to do cheaty clippy things  Port will stay as-is, I'm not going to refit Intrepid at this point, and I also am concerned about the stiffness of a junior port under ship maneuvers.

1 hour ago, Geschosskopf said:

Why hold onto the probes so long?  Instead, release them in stages during Intrepid's capture burn.  First Intrepid burns just enough to capture with its Ap out in the vicinity of Pol and Bop.  Pause the burn and quickly release any probes going to those places.  Start burning again until the Ap is just a bit beyond Tylo.  Pause the burn, release the Tylo probes.  Continue the burn until Intrepid's Ap is set for Vall or Laythe, wherever it's going first.

Actually that's kind of the plan, because "ripple firing" the probes at the moons from just within Jool SOI would be crazy awesome. But if I screw up or any of that or it otherwise doesn't work, I want to build in the capability to shoot them from a local position.

Edited February 26, 2016 by Kuzzter

[Geschosskopf]

7 minutes ago, Commander Zoom said:

Was going to say, this assumes a capture burn... and then I remembered that, post 1.0, aerobraking at Jool from interplanetary speeds is a much more problematic thing.  (I haven't been back since before.)  Bye bye to that "free" (firey-hot) delta-V at the arrival end.

During my expedition, back in the Before Time, I held onto my probes until after the first braking/capture pass, so that they wouldn't need to burn to keep from flying back out of the system entirely.  May or may not be relevant now.

Ah yes, the Before Time.  I miss it badly.  Back in those days, you could send "probe bombs" to Jool, set to fragment in the vicinity of its SOI edge.  Then each probe could aerocapture on its own to a different Ap all for free

 

[Commander Zoom]

I would say you've been reading my Jool album, but I haven't put it up yet...

(I really should get on that.)

EDIT:  @Rune, you say the linears aren't a mass-savings over the quads?  well, dang... they should, IMO.  That's what I get for not actually looking.

Edited February 26, 2016 by Commander Zoom

[Rune]

4 hours ago, Geschosskopf said:

Well, you do need enough battery to transmit the science data without interruption.  The actual SCANsat biome and altimetry maps aren't big things (about 500-1000 should do it), but if @Kuzzter hasn't disabled stock resource scanning in SCANsat, the M700 will need about 2500-3000 EC in the battery to do its survey.

 

3 hours ago, Kuzzter said:

Regarding the electrical system--the reason for the battery is to have enough charge to transmit the scan. in the Bill Space Program test just concluded, about 700 charge (net) was consumed to get the scan picture. So I think that battery mass may be necessary.

Are you guys 100% sure on that? Because last I gathered science (which admittedly, I haven't needed in a long time), if the charge run out mid-transmission, the antenna would just get into "trickle mode" and send the packets as soon as it builds enough energy for them (screwing your SAS in the process). So if you have enough E's to send one packet, and patience, you will eventually get 100% of the science. Batteries are for when you are time-limited (Jooldivers, for example). Then again, I have been wrong before, and I surely will be again.

3 hours ago, Kuzzter said:

Really good points--let's see: I think where I ended up is 10 total linears, not 12: if I miss the bay going in, I can just puff the main engines to back out, right? But yeah,4xblocks might be more optimal I actually did to that on the Jooldipper probes (not pictured) and it works perfectly. However, I might just like the look of the linear ports better. Maybe I can get the number down to 6, I have plenty of torque installed to handle the imbalance of the single off-center RTG. 

And regarding O-10s etc: no, I don't think I want to go with a single off-center one. TWR is already pretty low, and I don't want to do cheaty clippy things  Port will stay as-is, I'm not going to refit Intrepid at this point, and I also am concerned about the stiffness of a junior port under ship maneuvers.

Actually that's kind of the plan, because "ripple firing" the probes at the moons from just within Jool SOI would be crazy awesome. But if I screw up or any of that or it otherwise doesn't work, I want to build in the capability to shoot them from a local position.

Yeah, I just suggest stuff 'cause I can't help myself, but of course, you are the master of this ship! And we are discussing, literally, minute details...

3 hours ago, Commander Zoom said:

I would say you've been reading my Jool album, but I haven't put it up yet...

(I really should get on that.)

EDIT:  @Rune, you say the linears aren't a mass-savings over the quads?  well, dang... they should, IMO.  That's what I get for not actually looking.

Individually, yeah, they are lighter (30kgs vs 50). But if you don't use the bare minimum for translation (6), and need them to do attitude control, then you need so many of them (10 minimum, I think) that the whole system is actually heavier, and by quite the fraction (300kgs vs 200kgs). Of course, this is peanuts on bigger ships, and the linear ones have twice the thrust...

 

Rune. Professional nitpicker here.

[sdj64]

If the initial goal was to save part count and mass, I don't think the group has succeeded on either. Liquid fuel has the tank and engine variety to save part count, and the ISP to save mass.

Here are a few more options:

Disposable RCS package on a decoupler attached to each scan sat

Girder cradle on the tugbot that holds the sat in place without docking to it, maybe a landing leg to hold it in

Target the docking port with the sat, line up in front, burn 0.5 m/s, flip around before you get in the bay and hope

Right angle claw tugbot, like you thought about before

[ExplorerKlatt]

11 minutes ago, Rune said:

 

Are you guys 100% sure on that? Because last I gathered science (which admittedly, I haven't needed in a long time), if the charge run out mid-transmission, the antenna would just get into "trickle mode" and send the packets as soon as it builds enough energy for them (screwing your SAS in the process). So if you have enough E's to send one packet, and patience, you will eventually get 100% of the science. Batteries are for when you are time-limited (Jooldivers, for example). Then again, I have been wrong before, and I surely will be again.

 

Rune. Professional nitpicker here.

A recent update (1.0.5, I think) changed antenna behavior. If you run out of EC the antenna cancels the transmission and stores the data back in the experiment 

[Rune]

Just now, ExplorerKlatt said:

A recent update (1.0.5, I think) changed antenna behavior. If you run out of EC the antenna cancels the transmission and stores the data back in the experiment 

Isn't that what the "require complete" button is there for? Or is that one of Claw's bugfixes and I am looking like a bigger fool with this comment?

 

Rune. Right now I give myself even odds. I've been known to be an optimist.

[Mister Dilsby]

I love how you guys are splitting kilograms for a probe that will have something like 3600-3800 vacuum dV, depending on whether it's 5 1/2 linear ports or two thruster blocks, carried on a 1200 ton monster with several km/s on full tanks which I've never even bothered to calculate because ISRU. No, really! No sarcasm! Really! This is among other things an engineering game, so why not be the best, most optimalist engineer you can possibly be?

[soundnfury]

1 hour ago, Rune said:

Individually, yeah, they are lighter (30kgs vs 50). But if you don't use the bare minimum for translation (6), and need them to do attitude control, then you need so many of them (10 minimum, I think) that the whole system is actually heavier, and by quite the fraction (300kgs vs 200kgs). Of course, this is peanuts on bigger ships, and the linear ones have twice the thrust...

Actually, you can get full three-axis control (only two-axis translation, but that should be enough with a little skill) using 8 linear thrusters.  I discovered this when building Mariner-alike probes in Realism Overhaul, which have a convenient hexagonal frame.  Label the corners of the hexagon ABCDEF, then you put thrusters prograde at A & D, retrograde at B & E, and pairs of roll thrusters at C & F.  (If you're _really_ patient, you can use single thrusters at C & F, rolling in opposite directions, but then rolling causes translation.  That can get tricky.)  Now, as long as your CoM is at the centre of the hexagon, you can rotate around any axis with a suitable pair of thrusters.  For the axis through CF, use A & E or B & D; for the other horizontal axis, use A & B or D & E.

In Realism Overhaul, this works out as a net win, but in stock, 8 linears is still heavier than the 4 4-blocks for full 3-axis control.  Also, in stock you have overpowered reaction wheels, so you often only care about translation; that is not generally the case in RO.

So none of the above is of any use to you here, but I thought I'd share anyway!

[Deddly]

Since @Kuzzter just sanctioned us talking about various ways to optimise tiny probes

I could never understand why anyone would use four or more thruster blocks on a small craft when all you need is a combination of two thruster blocks at or near the CoM and two linear RCS ports placed directly on top of them using offset (assuming you have reaction wheels).

Edited February 26, 2016 by Deddly

[FyunchClick]

59 minutes ago, Deddly said:

Since @Kuzzter just sanctioned us talking about various ways to optimise tiny probes

I could never understand why anyone would use four or more thruster blocks on a small craft when all you need is a combination of two thruster blocks at or near the CoM and two linear RCS ports placed directly on top of them using offset (assuming you have reaction wheels).

Well, that gives you full translation, rotation an yaw but not pitch. whether or not that is a problem depends on having reaction wheels to compensate (not all probe cores do). It IS the way I set up RCS on my mk2 based SSTO's though (taking to heart the many fine lessons of the master ssto builders at the forum which usual begin with 'what can you strip?').

Edit: but reading your post again, you did mention reaction wheels. Was that edited in or am I still that drowsy coming out of a nasty flue? In any case please consider this an endorsement on the efficiency of that method you propose, whether or not it meets Kuzzter's aesthetic requirements.

Edited February 26, 2016 by FyunchClick
oops.

[Mister Dilsby]

1 hour ago, Deddly said:

I could never understand why anyone would use four or more thruster blocks on a small craft when all you need is a combination of two thruster blocks at or near the CoM and two linear RCS ports placed directly on top of them using offset (assuming you have reaction wheels).

Even for not-tiny spaceplanes often just use 4 RCS blocks mounted in pairs fore and aft, with a place-anywhere next to each. Skimmeroo does this. That gives translation in all directions, reaction wheels take care of rotation, and it avoids the problem of finding spots on the top rear of the ship in and around where the tail assembly wants to go.

[Geschosskopf]

1 hour ago, Kuzzter said:

Even for not-tiny spaceplanes often just use 4 RCS blocks mounted in pairs fore and aft, with a place-anywhere next to each. Skimmeroo does this. That gives translation in all directions, reaction wheels take care of rotation, and it avoids the problem of finding spots on the top rear of the ship in and around where the tail assembly wants to go.

We really need 5-way RCS blocks as stock.  I use the above in my stock games and begrudge having to add those extra 4 parts, given that spaceplanes are part-count monsters to begin with and then usually they have to dock to something that itself is already a part-count monster.

The only way I've found to minimize the number of RCS thrusters is to use RCS Build Aid to make sure the CoM of the rest of the ship doesn't move at all regardless of resource consumption, then put 4x 4-way thrusters exactly on the CoM angled at 45^above and below the centerline.  Unfortunately, this only works with cylindrical ships, so no Mk2 spaceplanes, and you need reaction wheels for pitch rotation.  But that's the fewest parts.

[FlyingPete]

19 hours ago, Rune said:

-Docking port: too late now, right? Still, a small one would have netted you another 30kgs in weight reductions, if you are up for an Intrepid "redesign". You could do it with KAS, just bolting a Jr. docking port over the old one...

I've found a lot of use for the humble docking port adapter. Just needs a mid-sized port, that trashcan-lid-thing and a junior port. Fly it up attached to the scansat (with struts bracing the junior port connections) and leave it attached to Intrepid. I've done the same thing with the large docking ports and it's a useful bit of hardware to have lying around a space station.

Alternatively, assuming the scansat doesn't need to be retrieved, just add a decoupler between the docking port and the scansat. Then it can be undocked as normal, then the docking port jetissoned before the probe is despatched on its mission.

[FyunchClick]

13 hours ago, Geschosskopf said:

The only way I've found to minimize the number of RCS thrusters is to use RCS Build Aid to make sure the CoM of the rest of the ship doesn't move at all regardless of resource consumption,

I love RCS Build Aid, Apart from allowing you to tweak radially asymmetrical ships to handle well even off SAS, it's great to help make sure your space planes stay stable in atmosphere wet or dry by showing both COMs (and allowing to include/exclude resources) allowing you to get the COL, position landing thrusters if you have them, and of course to help place RCS to not translate when you change attitude, and vice versa. It's become as indispensable to my builds as KER and it makes hitting an asymmetric design that handles not just well but great a lot easier. 

Sorry for the rant, just had to plug RCS for a bit.