Rockomax mainsail and poodle

[Ziff]

No I meant it's moot because if the argument is "I'm going to burn a huge amount of fuel in orbit where the 20s Isp actually adds up to make a difference" then I would point out that there's an engine with 410s better Isp that you should be using instead for that ship.

Allow me to quote myself.

You are making a direct comparison between engines, which is useless because the engines are used to move mass. Let's compare it using ship weights.

So, you are still missing the point. The argument is not 'I am going to burn a huge amount of fuel in orbit.' The argument is you can't compare engines like that, and this is an example of why. Now, let me reiterate, I already told you I only used fuel tanks because they're heavy and easy to add. I simply created a situation, landing a 15tn vs 85tn craft on the Mun with a single engine only. It has nothing to do with how much fuel is currently in the craft. It has to do with the weight of the craft. The argument is actually this: In a 15tn craft the T30/45 are better choices because the added extra weight of the Poodle is worse than the gain from the 20ISP. In an 85tn craft, the weight savings of the T30/45 engines aren't as effective as the gain you get from the 20 ISP on the Poodle. It doesn't matter how much fuel is in the craft, it matters how heavy the craft is vs the weight of the engine and Isp. Get it now? I am trying to show you why you cannot compare engines directly to each other without using actual test craft with specific mass. I hope that you at least see why that is now.

The craft doesn't have to be 85tns. Here, I built you a 3man Mun lander. It's pretty typical as far as landers go, RCS and SAS and some legs. The 3 external tanks are so it has enough fuel to go from Kerbin orbit to the Mun, Land, and Return. It's going to drop the external tanks when they're empty too!

The first has slightly more Delta-V and Thrust on the Mun. Notice that the second craft even weighs less then the one with the poodle. Which engine would you choose? Now, the NERVA won't work here because it's too tall for the landing legs. The Aerospike and T30 aren't gimbaled and thus, probably wouldn't make a wise choice for a landing engine. Also, the Aerospike wouldn't work anyway because you can't put decouplers on them which would prevent it from having a launch stage below it. I am sure your next argument is going to be 'Well, the differences are negligible.' Sure, maybe it's close enough to not really matter. It just serves to clarify my point that you cannot compare engines the way you were.

Edited October 4, 2012 by Ziff

[doggie015]

For launching heavy (Interplanetry) rockets I have found that the lifting capacity of the mainsail off the pad is really quite staggering! If you want to get a really heavy rocket off the pad then mainsails are your best bet... But after the first stage is jettisoned you will want to use something a bit more efficient like a T-30 cluster (For maneuverability just add a large RCS tank and thrusters (preferably 4 symmetry) as close to the bottom of the stage as you can get in order to increase the leverage around the centre of mass). If you need a 3rd stage to deliver the final push to orbit (A good interplanetary rocket has 2 stages to orbit and also uses the second stage to perform the injection burn or at least most of it!) then just use aerospikes or poodles or T-45's (Depending on your stage width, weight and delta-V requirements)

[Temstar]

The argument is you can't compare engines like that, and this is an example of why.

So why do we calculate Isp and TWR for engine then? Isn't the whole point of calculating these values for different engines is so we can compare them?

Now let's have a look at your lander. 6400L of fuel is a lot of fuel for a lander and I can only assume it's intended for both TMI and TKI as well as the actual landing. That's exactly the point I said earlier - if you're going to burn a lot of fuel with these engines then the 20s higher Isp will catch up. But if you're using Poodle then it means you're firing them out of atmosphere. If you're going to fire an engine in space and burn through that much fuel with it then why not use atomic rocket? Take your poodle lander for example, the service module is 48.17kg fuelled and 32kg of that is main engine fuel. I ran it through my own Tsiolkovsky rocket equation and 4177m/s is about right, not withstanding dumping fuel tanks and RCS fuel usage:

ln(48.17 / 16.17) * 390 * 9.81 = 4176.27m/s

TWR on Kerbin is (220 / 9.81) / 48.17 = 0.47

But why would you build a ship like that? Instead of a Poodle, let's swap it out for four LV-N atomic rocket motor, the ship is now 54.67kg fuelled and still have 32kg of fuel. If we plug that into Tsiolkovsky rocket equation:

ln(54.67 / 22.67) * 800 * 9.81 = 6908.38m/s

TWR on Kerbin is (240 / 9.81) / 54.67 = 0.45

But Tem they say, I can't live without at least 0.47 Kerbin TWR. Well okay, let's be inefficient and put 5 LV-N on:

ln(56.92 / 24.92) * 800 * 9.81 = 6482.26m/s

TWR on Kerbin is (300 / 9.81) / 56.92 = 0.54

So there you go, instead of a Poodle, if we instead put on 5 LV-N engines on you end up with the same ship but with 55% better delta-V performance AND 15% better thrust to weight ratio despite LV-N's terrible TWR. So why would you use Poodle?

[EndlessWaves]

The first has slightly more Delta-V and Thrust on the Mun. Notice that the second craft even weighs less then the one with the poodle. Which engine would you choose?

Kerbal engineer can't calculate the extra DeltaV from dropping tanks so given that vehicle will be burning a quarter of it's fuel when substantially lighter I'd think the poodle is actually less efficient.

It's certainly possible to come up with a design that's too unbalanced for an aerospike to handle, too heavy for a 909 and the wrong geometry for LV-Ns. However, it does seem far too rare for such situations to actually come up for the current stats of the poodle.

Having a Mainsail under a stack of 2m fuel tanks is better for a launch stage because the 2m core allows you to attach loads of extras around it, something the one meter models cannot do as well.

Fuel tanks are independent from the engine below though, you can mount 1m engines under 2m tanks.

those adaptors and struts have their own mass, and drag - more parts will inevitably mean more drag losses than with a simpler design...

Drag is based on mass, not part count. Fuel tanks containing the same fuel have the same amount of drag whichever combination of them you use.

well, one must account for the added "complexity mass" from a multi-engine setup...

while if having one large beast of a burner doesn't seem top add up against many itty bitty ones, remember that you're gonna need adaptors and other structural bits to get the cluster mounted, and more parts equal more wobble - which you'll have to beef up the airframe with struts to eliminate

Except that the fuel tanks in the game are currently relatively small so you need more fuel tanks for the inefficient larger engines than you do for efficient smaller ones if you're talking about long burns like getting into orbit around Kerbin, and long burns generally go with a high thrust to weight ratio requirement.

Perhaps you could provide an example craft file of a design where a mainsail-based setup is the fewest number of parts needed to achieve an objective?

Edited October 4, 2012 by EndlessWaves

[G Addict]

Am I the only one who uses the Poodle because it looks nice on a 2m command pod along with a 2m halftank? As has been said a few times already, the Mainsail is wonderful for getting those big payloads off the pad without building a stupidly huge cluster of smaller engines and inviting lag. Sure it's inefficient but it's easy. Most of my 3 kerbal rockets use a trio of 3x2m tank+mainsail stacks to get them to a reasonable height. One even uses mainsails all the way up to orbit! Efficient? Nah. Hassle free and ridiculously fast? You bet.

[sal_vager]

I use it for that reason as well G Addict, it just suits the large tanks better and I'm not worried so much be performance.

[EndlessWaves]

As has been said a few times already, the Mainsail is wonderful for getting those big payloads off the pad without building a stupidly huge cluster of smaller engines and inviting lag.

Show me your design that gets a mainsail powered vehicle to orbit using a minimum of parts because every time I've tried it I've ended up using no fewer parts than I would doing the same thing with aerospikes.

[Ziff]

So why do we calculate Isp and TWR for engine then? Isn't the whole point of calculating these values for different engines is so we can compare them?

You have to calculate the T:W of the craft with the engine, not the engine by itself. The T:W ratio of an engine by itself is meaningless as I have proven to you twice now.

Now let's have a look at your lander. 6400L of fuel is a lot of fuel for a lander and I can only assume it's intended for both TMI and TKI as well as the actual landing. That's exactly the point I said earlier - if you're going to burn a lot of fuel with these engines then the 20s higher Isp will catch up. But if you're using Poodle then it means you're firing them out of atmosphere. If you're going to fire an engine in space and burn through that much fuel with it then why not use atomic rocket? Take your poodle lander for example, the service module is 48.17kg fuelled and 32kg of that is main engine fuel. I ran it through my own Tsiolkovsky rocket equation and 4177m/s is about right, not withstanding dumping fuel tanks and RCS fuel usage:

Did you read my post, at all? It's clear you didn't. I told you the extra fuel is for the TMI from Kerbin to Mun, and landing, then you assume that is what it's for. I can't have a conversation with you if you totally fail to even read my post. That's twice now you have failed to read what I have written. Sigh. I gave you all the reasons why we can't do what you said, and you ignore all that and ask them anyway. You can make up excuses not to use the Poodle all you want, that doesn't mean it does not have a use. Either way, that's still not my point. My point is you cannot compare engines based on their own T:W ratio. You have to calculate the T:W ratio of the entire craft.

You want another reason? 5x NERVA's (Atomic Rocket) cost 8500 vs the 1 Poodle's 600. This doesn't matter right now, but it will matter when Campaign mode arrives.

Kerbal engineer can't calculate the extra DeltaV from dropping tanks so given that vehicle will be burning a quarter of it's fuel when substantially lighter I'd think the poodle is actually less efficient.

Actually, you can. Put the fuel tanks and engine for each radial/asparagus stage in a separate stage. That's exactly how you would figure it out anyway. Indeed, it's the exact same way you would calculate a vertical stage as well. I used to do that with pencil and paper long before the Kerbal Engineer mod came along.

[nhnifong]

I know the aerospike has superior performance, but I think it's ugly, and you can't attach anything to the bottom, and it has a set-back fiery effect instead of a close-in blue jet. So I never use it. I'd rather sacrifice efficiency for coolness.

[AmpsterMan]

Show me your design that gets a mainsail powered vehicle to orbit using a minimum of parts because every time I've tried it I've ended up using no fewer parts than I would doing the same thing with aerospikes.

Is this something along the lines of what you were looking for?

It can get 25 tons into LKO with some fuel to spare. It is my most refined booster rocket yet.

[EndlessWaves]

Actually, you can. Put the fuel tanks and engine for each radial/asparagus stage in a separate stage.

Yes, you're correct. What I should have said is that you can't show it in a single screenshot.

Is this something along the lines of what you were looking for?

It can get 25 tons into LKO with some fuel to spare. It is my most refined booster rocket yet.

While that's certainly a fuel efficient rocket the argument was that it used fewer parts and that one uses quite a lot, I count 27 in the picture and I assume there are six more fuel lines.

To lift 25 tons to a 75km orbit using aerospikes in the fewest number of parts you can use is 8 3200L fuel tanks, each with an aerospike on the bottom (16 parts), plus one or two decouplers for a total of 17-18 parts.

Is it possible to lift 25 tons to a low orbit using at least one mainsail engine in fewer than 17-18 parts?

I did a quick experiment myself and the best I managed was to equal it at 18 parts.

For reference I used 1x 4ton command pod, 1x large RCS tank and 1x 3200L fuel tank as my 25 ton payload.

[Ziff]

Yes, you're correct. What I should have said is that you can't show it in a single screenshot.

I debated about taking multiple pictures in order to show that, but I thought it might confuse people and steer them away from the general idea of my post.

[AmpsterMan]

While that's certainly a fuel efficient rocket the argument was that it used fewer parts and that one uses quite a lot, I count 27 in the picture and I assume there are six more fuel lines.

The total ship uses 86 parts I believe, (don't really know for the Lifter stage). Here I thought this was a part-efficient build (with some obvious aesthetics, I like building for the looks). Yet what you said is a lot more part-efficient than what I built. Maybe there is a payload-to-LKO mass number where it becomes more part-efficient to use the larger engines and 25 tons is below that number.