Engine ISP and Thrust

[NoMrBond]

So I'm trying to work out a good model for how ISP and thrust are related with the various other factors (such as gimballing and slISP/vISP), and while there isn't a lot of stock reference points I've come up with this

The CFit line (for 1% gimbaled engines); T/t = 150.66+ (150.5e^{-0.035(ISP-330)})-(0.155 x ISP)

For example, if you plug in 800 ISP, you get 26.66 Thrust per ton (T/t), the NV has 60 thrust so 60/26.66 = 2.25 tons (engine weight) for 60 thrust at 800 ISP

This isn't quite right though because the gap between slISP/vISP is larger than 50 for the other 1% TVC engines (the ISP gap figures somewhere as well)

Similarly, if you plug in 330 ISP, you get 250 T/t, the mainsail has 1500 thrust so 1500/250 = 6 tons

Other stuff is close, but not perfect, but there just aren't many examples, and some engines (I'm looking at you Rockomax 48-7S) don't seem to fit at all.

• high slISP (against vISP) decreases T/t by ~13%
  
• gimbal/TVC seems to decrease T/t by 13% multiplicatively for each 0.5% (so non-gimbal engines have ~30% more T/t (but only for 0 -> 0.5% and 0.5 to 1%)
  

Other; I ended up using T/t instead of TWR because I didn't think dividing all results by 9.81 was really necessary

Edited August 15, 2013 by NoMrBond

[NoMrBond]

Maybe more helpful numbers regarding conversion

vISP = How much thrust you get per ton of engine weight (@ 1% TVC)

800 = 26.66

780 = 29.76

760 = 32.86

740 = 35.96

720 = 39.06

700 = 42.16

680 = 45.26

660 = 48.36

640 = 51.46

620 = 54.57

600 = 57.67

580 = 60.78

560 = 63.91

540 = 67.06

520 = 70.25

500 = 73.55

480 = 77.05

460 = 80.95

440 = 85.66

420 = 92.01

410 = 96.26

400 = 101.65

390 = 108.64

380 = 117.91

370 = 130.42

360 = 147.53

350 = 171.15

345 = 186.21

340 = 204.02

335 = 225.07

330 = 250.01

325 = 279.57

320 = 314.63

Using the numbers

If your engine had 320/270 ISP with 1% gimbal, a 1 ton engine would make 315 thrust

If your engine had 320/270 ISP with 0.5% gimbal, a 1 ton engine would make 362 thrust

If your engine had 410/360 ISP with 1% gimbal, a 1 ton engine would make 96 thrust

or

I want an engine with 380/330 ISP with 1% gimbal and 1000 thrust, 1000/117.91 = 8.48, so the engine would have mass of 8.48 tons

Just want to make balanced engines really, and there isn't much to model after

Edited August 15, 2013 by NoMrBond

[Eric S]

That's pretty useful info for people making mod engines, thank you :-)

[Tiberion]

I tried to balance along this line of thinking once upon a time with Novapunch, back when there were even fewer stock engines to base numbers off of. It was pretty frustrating.

I would be interested to see how it works out, for a while now I have sort of manually set in ISP values after working out mass and thrusts using a TWR curve.

Not really sure what the stock numbers are based on, if anything, so it may be a bit like reading tea leaves.

[NoMrBond]

Thanks Eric and Tiberion, I guess this is a bit of an obscure game facet to be interested in

Hopefully there will be some official stock-alike guidance at some point, I guess I'll just keep fiddling with it until then

[rkman]

I want to use this to adjust some engines, so i put the formula in a spreadsheet. But i can't get a match with the OP's numbers (the graph and post #2) without adjusting a value in the formula.

The formula in spreadsheet form:

150,66+(150,5*(POWER(10; -0,035*(B3-330))))-(0,155*B3)

where "B3"=ISP

With that there is a good match for high ISP values and a perfect match at ISP=330, but other values diverge quite a bit from the OP's numbers.

More or less on a whim i started tinkering with the value "-0.035", and i get a perfect match over the entire range when i change it to -0.0152.

Although it's entirely likely that i did something wrong, it seems unlikely that i did something wrong that could then be corrected by adjusting that particular number.

[NoMrBond]

Hrm, I'll check the opening formula

In the meantime you can view the sheet itself here

[rkman]

Looks like i made a mistake after all. I wasn't sure whether "e" means "exponent" or "natural log". I assumed the former, and apparently that mistake can in fact be corrected by reducing the decay value (-0.035). On closer inspection it turns out my interpretation does give a small mismatch for ISP values <310.

Sorry about this, i'm not at all good at math.

Edited September 4, 2013 by rkman