Jump to content

DER (Duna Emergancy Relay) Writeup


Carl

Recommended Posts

I'll be making a challenge out of this in the near future all being well, but with imgur hating me at the moment, (can view images hosted there, but actually using it myself is beign a bit temperamental atm), i want to get the writeup done ASAP.

 

My problem was i'd totally overestimated the antenna capabilities of a probe i sent out, it had allready done a quick pass over minimum before going out for an even flyby, missed gilly on that run but then setup for a Duna intercept, as MJ was doing the the hohman burn i noticed my come line was red, a quick glance at the signal indicator, a lot of swearing, and some furious looking at the orbits of everything. Followed by more swearing.

 

The problem was that the probe was inside eve's orbit, in front of Kerbin in it's orbit whilst Duna was behind kerbin and falling backwards, it would be almost on the opposite side of kerbal by the time of intercept. That created a real conundrum. There was no way a probe launched the normal way could beat my existing probe to duna, it would be there in a little under 200 days and the best solution i could come up with still looked like a 100 days more. But there was an answer. I could launch on a retrograde intercept trajectory so i went around Kerbol the opposite way to all the planets, and my probe.

Eventually, (as in when i got attempts that got that far) i determined that would take some 166 days. Doable. I won't detail my failures here except to say that roughly a dozen configurations boiled down into 5 successive designs, (most intermediate configurations where getting to orbit problems), the first attempt fell so far short of the dv requirements it wasn't even funny. To be fair i had never tried this before, (if anyone else ever has please pipe up now, or have i got a KSP first?), save with mun/minmus missions back when MJ wasnt available just after releases, and the dv comparisons aren't even close. The second attempt was more serious, i'd done some proper looking into the numbers, but between a few mess ups still fell short of the amount needed for a duna intercept. Back to the drawing board, (these were all sim flights, i.e. revert to VAB).

This led to DER-3#, my first true BFR, at 4.6 kilotons in mass and a little under 700 parts. it finally had what it took to get me onto a duna intercept. In case your wondering, whilst i had an imperfect transfer window adding a bit to the dv it totaled up to about 19 kilometers per second. No you didn't misread that, kilometers per second. Let that sink in for a moment, thats how seriously dv hungry retrograde trajectories are. I knew as soon as i finished the intercept burn i wasn't going to have enough dv to decelerate at the other end, but i went out there with the simulation to find out how accurate my now well researched prediction of the dv required was going to be. Turned out spot on.

This resulted in DER-4#. 7.1 kilotons in mass and just over 900 parts. It made the intercept, it even had enough dv to decelerate, but then i found i didn't have enough solar power to run the ions at full, so i flew through duna SOI before i could make the 14 kilometer per second decel burn required. Yes 14kps dv to decelerate. These retrograde trajectories are a killer on the dv requirements.

So DER-5# was born. 7.35 kilotons in mass and a mind boggling 1007 parts. Early test flights ran into problems with unplanned disassemblies of the ion stage, (i thought i'd built it with the biggest round tanks then clipped radials to boost capacity per stage,but clipping issues created stresses which had predictable results when enough mass was above them and a decoupler was fired below them). Thats when i discovered what i'd mistaken for an ore container was actually a 1.25m xenon canister. Rebuilt with that with no clipping and joy it worked. Sort of. The rebuild of the ion stage meant that whilst peak TWR values where similar to the old form, with more mass in each stage it stepped up at less frequent intervals. (Initial duan decel attempt resulted in leaving duna SOI with 3.1kps still to kill). However DER-5#, unlike 4 didn't have to have the rotation angle babied every couple of minutes to retain decent thrust so i was willing to do somthing that would have just been too much hassle, (and created other issues too), with that form. Namely start the decel burn outside duna SOI. Doing so would allow me to shed enough velocity to increase my transit time of Duna enough to complete the burn. Worked like a charm, and despite expending somthing like 35kps i even had 6 and a bit kps of dv left over for duna maneuvering.

 

For anyone at squad who's remotely interested in things that made this a problem build in some way, (mostly what inflated the part count the worst), here's a breakdown of worst parts.

 

Sepetrons, with such large boosters i find you need sepetrons top and bottom to stop the engine on the bottom hitting things or getting shoved into the exhaust but you need more on the top so it yaws at the desired rate and in the desired fashion so aero then keeps carrying it away. Thats 6 sepetrons per booster. on smaller designs thats not an issue, but this has 5 radial mammoth stages followed by an in line mammoth stage (2 each and i'd have preferred more because 2 where too anemic but hand placing that many was too much of a pain), followed by an in line core rhino stage (8 sepetrons here), followed by another 6 on each radial rhino stage. The LV-N fuel stages i found got away without sepetrons but the switchover from LV- to Ions needed 8 more because they tended to be sticky. Thats a total of 200 sepetrons. Most on radial stages. A better solution is needed for BFR's. Also sepetrons really need to autostage position properly in the VAB, repositioning that many sepetrons was a huge pain.

2.5 and 3.75m LF only tanks and 2.5m LV-N engines. As it was each engine "pod" for the LV-N's was 5 parts, (girder, LF tanke, Nosecone, Engine, Fuel line), but to get enough TWR i needed 12 of the buggers. Being able to use a smaller number of 2.5m parts here would have saved 10's or parts, and if i'd had 3.75m LF only tanks i could probably of ditched all the radial mounted LF tanks and still got more LF fuel carriage. That would have ditched me another 168 parts. meaning in combination with 2.5m LV-N's i could have saved a potential 200 parts there too with the use of certain things i didn't have.

The ion stage is a bit more awkward to critique. Mostly because whilst i can think of several ways it could be improved, i'm not sure i would have had the tech for them. More girder lengths at the tiny cubic strut scale would have helped as each ion engine mounting arm has somthing like 8 tiny cubics thrown together. Ultimately 1.25m Ions with the significant reductions in the number of engines needed would have provided the most benefit. The tankage was part heavy, but for an odd reason. After discovering the 1.25m xenon tank i found that the higher weight of the 1.25m as opposed to 0.625m decoupler necessitated a total xenon carriage increase. I could have used 2 1.25m tanks at each step for this, but the ion stage was allready rather too long, doing that would have created serious issues. So i felt forced to use radial tanks. I don't think 2.5m xenon tanks would have helped, (though they'd be nice overall if we got 1.25m ion engines), as they would probably have carried too much fuel and especially the decoupler penalty would be even greater. IMO the issue is that xenon fuel is insufficiently dense. It makes carrying large amounts of it really volume intensive and to keep single axis craft dimensions under control on the launchpad your thus forced to adopt part count inefficient solutions.

After that you've only got the niggles, (see below), and the potentially major savings that could have come with 5m stock parts, but again i doubt i'd have had those unlocked even if they'd existed.

Beyond that only a few minor niggles. Not having the longest girder locked behind an endgame tech would be you know, cool. and 3.75m nosecones sop i could drop the part count from 2 to 1 there would be nice.

 

Now for those pictures detailing the stages of the successful run, some will have comment underneath, some won't. Be warned there's a lot though so i am spoiler tagging this for obvious reasons.

 

]WgGoXLH.png

On the pad. BFR indeed.

7poOXos.png

This angle lets you see more of the central core of the design with the Rhino bottom Center stage.

1DALXrq.png

Initiating Main Engine Sequence. We have liftoff.

wkHmbdD.png

This is why i also call this Atlantis, don't worry, if your monitor is too low contrast ratio to see anything through the smoke you'll get a clearer shot after the first set of boosters go.

TbYAbdn.png

First radial mammoth boosters away.

ySI1CF8.png

Here you go, better Atlantis Effect shot.

iOU3yDr.png

Second radial mammoth boosters away.

vIYc2UB.png

Third radial mammoth boosters away.

3lyTlKJ.png

24,000m in altitude and it still masses 4Kt's. As i said, BFR, Atlantis, Insanity, e.t.c.

e58fPDf.png

Third radial mammoth boosters away. Also the Fairing goes and we get our first look at the Ion Stage. Also note between this and the last screenshot i added a positive final burn angle to the ascent profile, i'll explain why in a bit.

s7CrRKE.png

With the fourth radial mammoth's gone where on the in-line mammoths below the radial Rhino's. The use of Mammoths here is to keep the  TWR high or you need an even steeper and less efficent final burn angle setting.

qSQ7BZq.png

In Line Mammoths Away.

Hk0zYq4.png

Center Rhino is exhausted and can now be separated.

PPW2ZD8.png

And just after that i set the burn angle back to zero. he problem is without that positive burn angle for most of the in line mammoth and early rhino radial burn you'll find your apoapsis goes up slower than you close on it, and you'll actually pass it during the burn resulting in a failed orbit attempt. Setting a positive burn angle slows the rate at which you approach the apoapsis enough that eventually you get enough velocity it starts going back up again, at that point you can zero it and not pass apoapsis on the way up.

riFFUtj.png

Coasting to Circularisation Burn.

WvXYhYv.png

Beauty Shot.

4FL2cMl.png

Circularisation burn, quite low dv, shame the accent is so inefficient in other ways.

JjVnPgH.png

We have orbit, final mass to orbit, 1.4kt's, and with 40kps of dv in the tanks. Again, BFR, Atlantis, Insanity, yadda, yadda, yadda...

On the other hand with that much dv the only thing likely to keep up with us over the long term is a whitestar on full burn. So there's that:D.

kpVDCTq.png

Setting up the Kerbin Exit burn. The main point is to expend the LFO stage doing it and get rid of them.

O0533nc.png

As you can see i setup the ejection burn final vector so it was pointing directly retrograde along Kerbin's Orbit around Kerbol, that lets me carry the maximum of velocity in the desired direction out of Kerbin.

LgVEekm.png

Lined up for the Burn.

sosHYQG.png

Burn in Progress.

WwdlfTI.png

Burn complete, jettisoning radial Rhino stages. And we get our first really clear look at the LV-N fuel core.

wgnsyJ8.png

Better angle.

E2LpW7z.png

Final Post Burn Trajectory.

rwDnm00.png

And here i am with my Kerbol orbit as a result of the last burn in kerbol orbit. Don't try doing the whole burn in Kerbin SOI, it isn't big enough for a burn of this magnitude. You'll leave it the maneuver node will go messed up and you'll end up flying of in a random direction. Note i'm still traveling prograde here. You can also see the red comm line going to the probe i'm trying to save.

RUqE1yW.png

Solar system view of the planned burn to not only send me into retrograde orbit around kerbol, but also put me on an intercept trajectory with Duna.

aoRxDSW.png

Aligning for burn.

EkuIwMU.png

Ready to go

jp8ybtA.png

Beauty Shot

5XcBbcf.png

And Ignition.

IIcSYiL.png

First LV-N Fuel Stage Separation.

3rzk5RA.png

Second LV-N Fuel Stage Separation.

RBVaJ6L.png

Third LV-N Fuel Stage Separation.

LwKyYLU.png

Fourth LV-N Fuel Stage Separation.

SMTowlW.png

Now enough stages are out of the way a cool cosmetic shot showing what that engine ring looks like.

nVShI4p.png

My Kerbol orbit around this point, note i'm still traveling prograde at this stage of events.

Y3s0ZrN.png

Another Beauty Shot.

PFpqR0p.png

Fifth LV-N Fuel Stage Separation.

bGWn1R6.png

Point of minimum orbital velocity, this is where the orbit goes from prograde to retrograde.

9S0K3IG.png

This is what the orbit, (or non-orbit really), looks like a minimal orbital velocity.

QZSNXWF.png

My orbital Debris trail, i had a lot of cleanup to do later on. Though some of it disposed of itself into Kerbol

S3hTIRi.png

Sixth LV-N Fuel Stage Separation, just the LV-N Drive Core stage to go and then i'm onto the Ion's.

BtSTWD7.png

LV-N Drive core a few seconds before burnout. Glowy, trippy.

tcDbEVu.png

Separation of LV-N Drive Core, Ion's are lit.

DdGBqYZ.png

So thats where the space Kraken got to, it was chasing my ion stage retrograde around kerbol. Huh, who knew :D.

zcjzmMU.png

Our first closeup of the Ion Stage.

uB3clI6.png

A look at the orbit shortly after Ion Stage ignition, i was a few minutes late taking this, but as you can see whilst i'm firmly retrograde around Kerbol now, my periapsis is still inside Kerbin's orbit.

bn8UOiK.png

Ohhh, X-Wing Ion Engine array, quick where's the nearest thermal exhaust port.

izDDJih.png

First Ion Engine Fuel Stage Separation.

72RYBuT.png

Closer...Closer...

XJ0kHRl.png

Phew, we have an intercept trajectory.

E39bTog.png

Here's the Debris Trail.

u0szvw4.png

Here's what it looks like on the map at duna. I did make a correction burn prior to this but after the main burn of some 1.4 m/s but it was so miniscule i didn't document it.

52qUNaf.png

Start of the Pre-Duna SOI Deceleration Burn

MJgiDWj.png

As you can see just over an hour out.

n3OuLFN.png

Second Ion Engine Fuel Stage Separation.

CPZuTbA.png

During this Decel Burn i briefly got an Ike Encounter, kinda pointless as i kept decelerating and it went away but cool.

qIbQhvu.png

See, Ike went away.

F6SOjYt.png

Third Ion Engine Fuel Stage Separation.

dxGD59h.png

And here's the end of the pre-SOI burn. Note i'm still traveling retrograde around Kerbol here, just at a much lower velocity.

qBI50WV.png

Now in Duna SOI i plotted the deceleration burn.

K61fCBk.png

Note the 2 and a half hour transit time of Duna SOI, whilst decel means it will allways take a good bit more than listed, that 30 minutes greater transit time compared to my pure in-Duna-SOI decel attempt coupled with the 4KPS lower decel burn required is what made this attempt with the pre-SOI burn work when the all-in-SOI burn didn't.

bifLNaN.png

Fourth Ion Engine Fuel Stage Separation.

RAMmXiP.png

Fifth Ion Engine Fuel Stage Separation.

x3TO8Tz.png

A quick look at my Duna trajectory shows that in spite of burning for well over an hour i'm now nearly 30 minutes longer from exiting duna SOI than when i started the burn, the trajectory has deflected a bit due to sideways velocity added by my decel burn, but it's still straight as a ruler otherwise.

vTlVbHj.png

Sixth Ion Engine Fuel Stage Separation.

hC26sJK.png

More deflection but still ruler straight.

0JlA2kL.png

A few minutes later, i've killed over 7kps of velocity since i hit duna SOI and despite some trajectory deflection it's still straight as a ruler, thats what decel burns from retrograde look like.

DFoD7HG.png

Finally with 400 m/s of velocity left to kill we have some serious curve.

7hXM97s.png

Seventh Ion Engine Fuel Stage Separation.

6N9XPmV.png

With that we just barely have a super elliptical orbit.

iSy97JH.png

100m/s to go, nearly there.

UQEupY2.png

We have Orbit.

Yay. And phew. [/spoiler]

 

Ok now thats out of the way, is there anything i could have done to make this more efficient that either involved tech i didn't have or which i've thought of since completing the mission? One thing, really basic too, but only came to me while uploading the screenshots. The way i fitted 12 LF only tanks around the 3.75m on the LV-N fuel stages could also have been done with one set around the bottom tanks for the first Radial Mammoth stage and the Rhino stage, to which i could then have mounted vectors, and with a bit more tech on the rhino stage, aerospikes. That would have doubled my Rhino stage thrust without hurting ISP, (thus allowing my in line mammoth stages to be ditched for more fuel on the rhino stage whilst still having better TWR), and whilst i suspect a third radial Mammoth stage would still have been needed, (to haul all the fuel tanks without going too long), and the part count wouldn't change much, (fewer sepetrons but more fuel, nosecone and engine parts), it would have greatly simplified many aspects of the design and given me much better TWR on the pad as each of the initial radial mammoth stages would have had 4 times the thrust.

Given access to Clampotron Seniors and more setup time i'd also have taken the LV-N stage and the Ion stage up separately with the former empty of fuel and used a rhino LFO stage attached to the bottom of the LV-N stage and then hauled the fuel up separately. Unfortunately time constraints and lack of clampotron seniors meant it had to be an "everything in one shot" deal.

 

I'm sure Slashy and co will critique the hell out of my design even worse.

Edited by Carl
Link to comment
Share on other sites

I'd suggest some large LF tanks and an LV-N.  I'm sending a crewed rescue chaser to catch my Dres colonizer, and 10k LF + LV-N goes pretty darn fast even with 2 tons of payload.

 

Using Mk3 fuel tanks, that comes in at only about 10 parts for the interplanetary bit (including lifesupport) to get 14km/s, and a pair of drop tanks to stage off could do the rest to get you to 21km/s for the cost of 6 more parts and 20k more LF.

You'd of course be swapping out the pod for a probe core, and the lifesupport for large relay antennae.

Link to comment
Share on other sites

34 minutes ago, suicidejunkie said:

I'd suggest some large LF tanks and an LV-N.  I'm sending a crewed rescue chaser to catch my Dres colonizer, and 10k LF + LV-N goes pretty darn fast even with 2 tons of payload.

 

Using Mk3 fuel tanks, that comes in at only about 10 parts for the interplanetary bit (including lifesupport) to get 14km/s, and a pair of drop tanks to stage off could do the rest to get you to 21km/s for the cost of 6 more parts and 20k more LF.

You'd of course be swapping out the pod for a probe core, and the lifesupport for large relay antennae.

Ummm, thats exactly what i did, did you look at the screenshots? A bunch of LF fuel tanks and a bunch of LV-N's to get the initial kick, and ions just finished the job, (it could be done with pure LV-N's but the weight to do so is ridiculous, far more than i could have lifted off Kerbin). And it's 19km/s just to get the intercept, it's another 14km/s to slow down at the other end. It's basically just shy of 35km/s for the full trip.

 

EDIT: or am i totally not understanding somthing here?

 

 

Edited by Carl
Link to comment
Share on other sites

Ok been moseying over the wiki i think i get what your saying. That i could have replaced all those LV-N fuel tanks, (baring those used as LV-N mounting points), with Mk3 LF fuselages. You mentioning some design of your own that i have no clue about, (and form your descriptions has way too few parts to work), threw me, sorry.

 

Your not entirely wrong, 9 of those plus the existing engine "nacelles" would have given me the same fuel over a slightly greater number of stages improving my overall dv slightly. However there are 2 key issues with that. 1. as i reiterated at a few points i was short tech in a lot of places, i don't even have access to Mk2 plane parts yet. 2. Even if i had that many Mk3 Long LF fuselages would if what i've seen of them in video's is true, have been much too long a design to be practical. The LV-N stage alone would have been quite a bit taller than the VAB which would have made it impossible to build somthing to lift it. 3. More minor as i could workshop round it several ways, but the thinner design would likely have made mounting 6 symmetry booster configurations i used much harder.

Edited by Carl
Link to comment
Share on other sites

Ok... ummm wayyyyyy to many pictures at once... put many of them in a spoiler box, or link the album...

Would occlusion by KerbOl be the problem, or is it just a matter of range?

If its just a matter of range, the easiest solution is to realize that the range depends upon antenna strength at both ends.

Assuming stock settings and a lvl 3 facility, DNS strength is 250 G. Increasing your start location strength to 1000 will double your range... an array of 10 RA-100 dishes will do this.

If you only need periodic contact, you can just place this array on the ground at KSC... contact 50% of the time, 3 hours on 3 hours off.

If line of sight is a problem, put it in orbit and send it to escape velocity so that it doesn't get blocked at the same time as kerbin.

0BStZOE.png

Should be much easier than going into a retrograde orbit around the sun with a relay, and then it will serve you well for missions to the outer planets, particularly with OPM

Link to comment
Share on other sites

9 minutes ago, KerikBalm said:

Ok... ummm wayyyyyy to many pictures at once... put many of them in a spoiler box, or link the album...

 

They are in a spoiler box...

 

As for the rest, didn't have high end antenna, The probe had a single communotron on it and all i had relay wise was 2Gj RA-2's. Still your probably right i could have more easilyl built a comm relay in kerbin orbit. if i'd realised that would work. The wiki makes it sound like that wouldn't work.

Link to comment
Share on other sites

1 hour ago, KerikBalm said:

Ok... ummm wayyyyyy to many pictures at once... put many of them in a spoiler box, or link the album...

Would occlusion by KerbOl be the problem, or is it just a matter of range?

If its just a matter of range, the easiest solution is to realize that the range depends upon antenna strength at both ends.

Assuming stock settings and a lvl 3 facility, DNS strength is 250 G. Increasing your start location strength to 1000 will double your range... an array of 10 RA-100 dishes will do this.

If you only need periodic contact, you can just place this array on the ground at KSC... contact 50% of the time, 3 hours on 3 hours off.

If line of sight is a problem, put it in orbit and send it to escape velocity so that it doesn't get blocked at the same time as kerbin.

Should be much easier than going into a retrograde orbit around the sun with a relay, and then it will serve you well for missions to the outer planets, particularly with OPM

 

Quote

 

Range Calculation

Each antenna has a "power rating" (measured in unitless number). It ranges from 5k for the integrated Command-Module antenna available in every spacecraft, to 100G for the largest dishes. A connection can be established between two stations (craft to ground, or craft to relay) when the square root of the product of the strength ratings is greater than the distance between the two objects. When a craft has multiple, combinable antennas, the antenna's power ratings stack in a way that provides a diminished return using the following formula:

  • Vessel Antenna Power = Strongest Antenna Power * ( Sum of Antenna's Powers / Strongest Antenna Power ) ^ ( Average Weighted Combinability Exponent for Vessel )

The range can then be calculated as stated above by taking the square root of the product of the two Antenna Strength ratings

  • Range = SQRT ( Antenna Strength 1 * Antenna Strength 2 )

The Average Weighted Combinability Exponent for the Vessel can be thought of as the average of the sum of the combinability exponents of all combinable antennas on the vessel when weighted proportionally against their respective power level. This can be shown in the following way:

  • SUM (( Antenna 'n' Power * Antenna 'n' Exponent ) : ( Antenna 'n+1' Power * Antenna 'n+1' Exponent )) / SUM ( Antenna 'n' Power ) : ( Antenna 'n+1' Power )

i.e. A vessel with a Comm 88-88 (100e9 @ 0.75) and also a Comm 16 (500e3 @ 1.00) would have the following:

  • (( 100e9 * 0.75 ) + ( 500e3 * 1.00 )) / ( 100e9 + 500e3 ) = 0.75000125 ... being the Weighted Average Combinability Exponent for the Vessel.

Please Note: All Antennas and Relays have a Combinability Exponent value of 0.75 except for the Communotron 16 having a value of 1 and the Communotron 16S having a value of 0. These values can be found in the part config files.

 

you need 23 RA100 to get 1000G range 10RA100 will only give you 560G

Link to comment
Share on other sites

Hmmm, I must have missed when they made the combinability exponent 0.75 instead of 1.

Anyway, an antenna 4x as powerful will give you 2x the range. An array with an effective strength of 560 will give you sqrt (560/250) = 1.4967x the range... is a 50% range extension enough?

Is it a range problem, or occlusion? If its occlusion, you don't need to go full retrograde, just put your array into a lower orbit, and it will get ahead of kerbin, and your array and kerbin wont both have their LOS blocked at the same time.

I would try and solve this problem in some other way than going into a retrograde orbit around the sun... heck, even a trip to an eliptical orbit with the PE inside Moho's orbit could probably work for far less mass and dV.

Link to comment
Share on other sites

This mission would have been more fun to read about if someone a little more humble had done the write-up. Opening lines blaming imgur for "not working" made for a cringeworthy start, but you lost me with the three full paragraphs of why it's SQUAD's fault that this was hard to do. In fact, I'm still not really sure why you wanted the retrograde intercept, or why you considered it a more prudent option than a cheaper and less complicated solution--the complaints and the self-aggrandizing are the only things I got from this post. Would have been more appropriate in my opinion to cut out all the excuses and bragging and touting of minor details and preface this extensive report with a more humble approach: "I'm sure this could have been done with fewer parts if I was a more experienced mission designer, but I had fun doing it this way anyways and learned a lot."

You don't have to make up reasons for every little move you make--it's fine if you just wanted to do the retrograde intercept for its own sake. No right way to play KSP, after all. Just... maybe try not to be so high and mighty. Glad you got your probe connected. Hope you had some fun.

Link to comment
Share on other sites

2 hours ago, MitchS said:

You don't have to make up reasons for every little move you make--it's fine if you just wanted to do the retrograde intercept for its own sake. No right way to play KSP, after all. Just... maybe try not to be so high and mighty. Glad you got your probe connected. Hope you had some fun.

No offense, but someone complaining about other people's writing style and trying to dictate other people's writing and play style don't have much of a leg to stand on when it comes to advising others "not to be so high and mighty".

Link to comment
Share on other sites

33 minutes ago, suicidejunkie said:

Sorry, the high part counts and mentions of Xenon tanks got me on the wrong idea, and the images are blocked at work.

No worries, might have helped if you'd explained that at the start, but eh, where fine so no worries. With these kinds of dv requirements a single engine type design was never going to work anyway. The onyl drive that dosen;t hit it's dv cap before completing the mission is ion, and i didn't want to do burns that insanely long.

 

3 hours ago, KerikBalm said:

Hmmm, I must have missed when they made the combinability exponent 0.75 instead of 1.

Anyway, an antenna 4x as powerful will give you 2x the range. An array with an effective strength of 560 will give you sqrt (560/250) = 1.4967x the range... is a 50% range extension enough?

Is it a range problem, or occlusion? If its occlusion, you don't need to go full retrograde, just put your array into a lower orbit, and it will get ahead of kerbin, and your array and kerbin wont both have their LOS blocked at the same time.

I would try and solve this problem in some other way than going into a retrograde orbit around the sun... heck, even a trip to an eliptical orbit with the PE inside Moho's orbit could probably work for far less mass and dV.

It's a range issue, as i explained in the OP though the orbital positions make even a low sun pass inside moho take a 300 days mission to do it, (MJ has an ASAP intercept function on it's advanced transfer planner). I had approx 200 days. I might have been able to beat MJ a bit by building an initial trajectory that wasn't an intercept so i closed with the sun faster, then burned hard there, but that would likely still have required a lot of dv as most visions i have of it still mean a lot of acel at the start, lots of decel at the sun, and still needing a high velocity relative to duna at intercept meaning a lot to kill there, it may just have been technically capable on technically less dv, but i doubt the savings would have been massive even if it works. Image of the relative orbital positions of Duna, Kerbin, the Probe, and the Probe's, i circled the prober, it's duna intercept, and kerbin and duna in red, i circled my approx remembered kerbin position at time of probe reaching duna in blue.

gKC2dZz.png

If you want to have a mess i do have a backup save from just before mission launch btw. I was tempted to include it in the Op, but after fighting imgur for 2 days straigth where i did nothing else much to relax i wasn't in the mood to mess very much.

 

@MitchS: I honestly feel like you never even read 2/3 of what i wrote and instead skimmed and made completely erroneous assumptions. As noted above i'd been having serious issues with Imgur for days before this, as have other forum members who've been complaining in the "what did you do in KSP today" thread about it. At no point did i rag on SQUAD or say it was their fault. Instead i provided feedback on what made the whole thing the biggest pain to build and fly. At no point did i demand immediate instant action. In fact i expected but maybe i should have spelled this out, people to take the inference from me calling it a BFR in a little doom joke, and referencing SG-A with another name and calling it insanity that i didn't expect people to build designs of this size because designs of this size are way outside the scope of the game at the moment. The feedback is thus as an inference from that, not something i expect much action on for the sake of rockets this big, (though some are issues even at smaller sizes ofc). But since i thought fellow forum members might like to hear the tough points on the design and since most devs love feedback i thought i'd provide it, at no point did i rag on squad for it being hard, it's a BFR, an Atlantis, Insanity, it was never going to be easy. I knew that, i figured i didn't have to explicitly state that in big glowing neon letters. Asd for why a retrograde shot, i explained it once allready, but a bit more detail with specific screen shot is above. Am i acting a bit liquidy now, yeah. But then i spent 2 and a half days building that OP doing only a little relaxing and someone has then come in and started outright attacking me after clearly not having read the OP in full. I think i'm allowed to be a bit upset about that.

Link to comment
Share on other sites

12 hours ago, Carl said:

It's a range issue, as i explained in the OP though the orbital positions make even a low sun pass inside moho take a 300 days mission to do it, (MJ has an ASAP intercept function on it's advanced transfer planner). I had approx 200 days. I might have been able to beat MJ a bit by building an initial trajectory that wasn't an intercept so i closed with the sun faster, then burned hard there, but that would likely still have required a lot of dv as most visions i have of it still mean a lot of acel at the start, lots of decel at the sun, and still needing a high velocity relative to duna at intercept meaning a lot to kill there, it may just have been technically capable on technically less dv, but i doubt the savings would have been massive even if it works. Image of the relative orbital positions of Duna, Kerbin, the Probe, and the Probe's, i circled the prober, it's duna intercept, and kerbin and duna in red, i circled my approx remembered kerbin position at time of probe reaching duna in blue.

perhaps this is all the result of over reliance on MJ.

Kerbin has an Orbital period of 426 days- that is to go all the way around the sun. In 213 days, it gets halfway around the sun.

If you drop your PE outside Moho's orbit, your SMA is now ~0.7x what is was before.

https://en.wikipedia.org/wiki/Orbital_period#Small_body_orbiting_a_central_body

as SMA is the only thing that changes here, the relative orbital period is thus sqrt (0.7^3) = 0.586 -> so it takes 249 days to do a complete orbit, and 124.5 days to reach perapsis.

You don't care if moho is there or not when you reach perapsis, so don't bother having mechjeb give you a moho intercept.

Duna is about 1.5x as far from the sun as Kerbin, so when they are nearly on opposite sides, its about 2.5 "KU" away. Putting a relay to be just outside Mohos orbit at PE will make the relay to Duna distance be about 1.1 KU.

Problem solved? I'm still not sure what you're using to link to your probe, you mentioned a communitron, but also RA-2s... If you have RA-2s around Duna, then a threesome of RA-100's sent on a low PE trajectory should get you what you need - its not that much dV, most of the DV going to Moho is spent on the capture burn, not th intercept burn. You dont need to capture, just for it to be around PE at the right time (although a capture could be convenient for the future).

Also, even if your pictures are in a spoiler box, I suggest splitting them into groups of multiple spoiler boxes, with a little description of what is in each box... I was quite surprised by the flood of pictures when I clicked that one box

Link to comment
Share on other sites

Quote

perhaps this is all the result of over reliance on MJ.

Kerbin has an Orbital period of 426 days- that is to go all the way around the sun. In 213 days, it gets halfway around the sun.

If you drop your PE outside Moho's orbit, your SMA is now ~0.7x what is was before.

https://en.wikipedia.org/wiki/Orbital_period#Small_body_orbiting_a_central_body

as SMA is the only thing that changes here, the relative orbital period is thus sqrt (0.7^3) = 0.586 -> so it takes 249 days to do a complete orbit, and 124.5 days to reach perapsis.

You don't care if moho is there or not when you reach perapsis, so don't bother having mechjeb give you a moho intercept.

Duna is about 1.5x as far from the sun as Kerbin, so when they are nearly on opposite sides, its about 2.5 "KU" away. Putting a relay to be just outside Mohos orbit at PE will make the relay to Duna distance be about 1.1 KU.

Problem solved? I'm still not sure what you're using to link to your probe, you mentioned a communitron, but also RA-2s... If you have RA-2s around Duna, then a threesome of RA-100's sent on a low PE trajectory should get you what you need - its not that much dV, most of the DV going to Moho is spent on the capture burn, not th intercept burn. You dont need to capture, just for it to be around PE at the right time (although a capture could be convenient for the future).

Also, even if your pictures are in a spoiler box, I suggest splitting them into groups of multiple spoiler boxes, with a little description of what is in each box... I was quite surprised by the flood of pictures when I clicked that one box

Nope the Relay had to go to duna, because thats where the probes going.

 

I completely didn't understand communications when i built the original probe, and because the eve trip had no issues, green comms the whole way outside of occlusion i assumed i could do duna ok. The probe had two 2 2GJ communications on it. Aside from a  sunprobe with i'm not sure what on it, (i thought it has a single 2Gj communtron, but since it's solid green i must have symmetryed more on than i remembered putting on), i have no other comms stuff outside kerbin orbit, and nothing outside Kerbin orbit with relay. I had to get a relay that could handle the Duna-Kerbin leg on it;s own all the way to duna, (or so close as makes no never mind).

 

You can see here as i decelerated at duna the red comm line to the probe, that popped up just a few days prior and look how close the probe is to duna and the relay:

 

dxGD59h.png

 

Give me a second and i'll upload the save to my google drive and you can mess, see if you can get there easier with a suitable relay.

 

Save file, sfs only, don;t know if you need metadata, if you do let me know.

 

 

Edited by Carl
Link to comment
Share on other sites

While it might have been a bit excessive ( would have spammed relays on the ground until I get contact) I think that is the most kerbal mission that has been launched this year. *Tips hat in appreciation*

Link to comment
Share on other sites

17 minutes ago, JacobJHC said:

While it might have been a bit excessive ( would have spammed relays on the ground until I get contact) I think that is the most kerbal mission that has been launched this year. *Tips hat in appreciation*

That was more or less my thought at the time. Though as noted, i totally didn't realise i could just spam relays at the time. Still it would have taken a lot of them. Not sure what that would have done to my framerate. Thanks for the compliment though ;).

Edited by Carl
Link to comment
Share on other sites

On 1/19/2017 at 11:49 AM, omelaw said:

you need 23 RA100 to get 1000G range 10RA100 will only give you 560G

i had a look at the wiki, and the part.cfg files.

I think the wiki is out of date, and based upon one of the many iterations from the 1.2 pre-release, where stacking penalties were added/removed, combinabiltiy was added/removed, etc.

There is no combinability exponent of 0.75 specified in the relevant .cfg files, despite what the wiki says

Link to comment
Share on other sites

1 hour ago, KerikBalm said:

i had a look at the wiki, and the part.cfg files.

I think the wiki is out of date, and based upon one of the many iterations from the 1.2 pre-release, where stacking penalties were added/removed, combinabiltiy was added/removed, etc.

There is no combinability exponent of 0.75 specified in the relevant .cfg files, despite what the wiki says

I tested it by myself, and craft with 10 RA-100s lose singal with DSN level 3(25GM) at 400Gm (keep connected at 350Gm, 2%) orbit around Sun, while legit 1000GM antanna (it can keep connection to DSN lv 3 up to 500Gm) shouldn't

Edited by omelaw
Link to comment
Share on other sites

From commsDish16/commsAntenna16cfg (It isn't dish antenna by the way)

Spoiler

    MODULE
    {
        name = ModuleDataTransmitter
        antennaType = DIRECT
        packetInterval = 0.6
        packetSize = 2
        packetResourceCost = 12.0
        requiredResource = ElectricCharge
        DeployFxModules = 0
        antennaPower = 500000
        antennaCombinable = True
        antennaCombinableExponent = 1
    }

 

It looks like it can be changed in part cfg, while defaulted to 0.75(or something) if not found in it.

Edited by omelaw
Link to comment
Share on other sites

This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...