Darkstar

Bearhedge · June 3, 2022, 4:56am

Hi team,

I had a slow day at work, first one in months…and I’m isolating at home with COVID (but feeling okay) so I thought I’d take the Darkstar through its paces a bit so I can use it in future Xmas treks if I’m threatening to run out of time.

I know it’s not a real aircraft and they’ve made the thing pretty darn forgiving…but it’s still kind of nice to know what it can do…and honestly, it is quite fun.

The thing I’m most interested in, is the range. The game states the range as 2,500 NM and endurance one hour:

Let’s have a look at that in a bit more detail. The Darkstar has four fuel tanks for the turbine engines and two fuel tanks for the scramjets.

The four conventional tanks total 17,849 lbs of fuel.

The scramjet tanks total 21,815 lbs of fuel.

As far as I can tell, the only fuel gauge in the Darkstar is the visualization of the aircraft with the tanks showing in green. The middle tanks (marked red below) are the scramjet tanks and the outer tanks (marked blue) are the conventional jet fuel tanks.

It appears that the fuel flow gauge shows turbine FF when the scramjets are off and scramjet FF when the scramjets are on and the turbines off.

The turbine fuel flow varies quite a bit, but full afterburner at low altitude tops out somewhere around 120-150 x 1000 PPH, less on lower speeds (perhaps 80-100 PPH). So somewhere between 7 and 12 minutes of full burner at low altitude before flameout. Not that we would do that, but it’s good to know.

A more useful statistic is that 90% RPM maintains a benign 400 kts at the kind of altitudes you may find yourself if looking for an airport after a speed run, and results in a 14 x 1000 PPH fuel flow, i.e. 233 lbs/min.

This was a useful thing to know - because I found that a typical ascent to the scramjet transition takes a bit less than 50% of turbine fuel, so the other 9,000 lbs actually gives over 30 minutes of powered flight at the other end.

I decided to try an Auckland - Darwin speed run to see if my theory held true.

I chose an IFR MSFS flight plan but deleted everything except the approach waypoints so the computer would show me a useful distance through the flight. The shaded area shows the approximate location of my projected crash site. 2777 NM distance to cover.

Acknowledge last transmission? So you can moan about how I need to descent back to FL450 or whatever? No thanks…farewell!

According to the FF gauge, the scramjet cruise FF is around 45-68 x 1000 PPH, which results in around 20-30 minutes of scramjet time. Depending on the speed, the Darkstar might be covering around 90-95 NM / minute at this stage of the flight, so somewhere between 1,800-2,800 NM range (that’s a bit of a large window, sorry, but I don’t have more granular info right now! I did get my highest speed towards the end of the cruise when the tanks were almost empty - but haven’t tested more). I topped out at M9.8.

I was running some rough numbers halfway through and estimated that I’d run out of scramjet fuel some 400NM before the approach waypoint, which is what happened:

Still going (positive FF), 414NM to waypoint:

Out of juice (no FF, rapid deceleration), 378 NM to waypoint:

I figured I’d gently coast down and see how things go. I coasted down to 80,000 ft, at which point the turbine engines automatically spooled back up to idle. I brought them back to full RPM but no afterburner, which maintained around M3.5 with 15 x 1000 PPH. Not bad at all: at 250 lbs / min I had half an hour of powered flight left and just 189 NM to go. I probably should have started descending earlier, but altitude was money in the bank at this point.

The descent at idle took very little fuel, so at the first approach waypoint I still had my wing tanks full.

On final.

Coming in to land at 160 kts. A bit slow, perhaps - at 150 kts she falls off the sky, I think.

So…2777 NM flight wasn’t a problem. I think I could do a lot longer, just need to run some numbers.

Aginor · June 3, 2022, 6:39am

You could save a bit of fuel by ascending without afterburners. I tested that you can change the profile a bit and take off without AB and IIRC even climb to FL300+ without AB.

Bearhedge · June 3, 2022, 6:59am

That’s a good finding…the more fuel left for the post-scramjet supersonic cruise, the better.

There’s also the question on if 120,000 ft is the most efficient altitude for the scramjet distance-wise. It appears MSFS altitude gauges stop working at 270,000 ft…

Aginor · June 3, 2022, 7:15am

I got it to cruise rather nicely 150,000 IIRC but I haven’t done any strict tests yet either.

Endurance might be better at the cost of some speed, or the speed might even be better.

I also did a steady 2.8G pull-up at 120k and 40° climb and got it way past 200k ft and speed stuck at Mach 10 (which I actually expected to drop at some point but it didn’t).

In fact I might have gone into space there. The speed and altitude were climbing rapidly until they just got stuck.

I now wish that MSFS supported spaceflight, at least in lower Earth orbit. A space plane like the default one from Orbiter would be awesome.

Bearhedge · June 3, 2022, 7:47am

It wouldn’t be great, wouldn’t it! X-15 in MSFS would be fun…and ideally the return from beyond the “edge of space” would be a hairy affair, like it was back then.

Hmm…so I ran out of scramjet at about 2300NM from the departure field, I think. I reckon I might get another 1,000NM of supersonic flight at 80k out of the remaining jet fuel, especially with the mil power initial ascent. Dead stick landing at the end like @PaulRix already demonstrated.

I reckon 3,500 NM is possible, especially with favourable winds: I had hefty headwinds on the Darwin run.

Aginor · June 3, 2022, 8:17am

@PaulRix is one hell of a pilot. I barely managed to land the thing under power, it handles quite poorly when flying slow (which means below 200 knots for that plane) and, well, you cannot see the runway out of the window.

But yeah, carefully planned a Space Shuttle style gliding landing should work fine and would save a lot of fuel.

Bearhedge · June 3, 2022, 9:02am

Yes, he is - but I hear he is blatantly doing things that keep his flight simming chops sharp on work time!

I can’t do any more testing today (the toddler is rudely using my simming space as her bedroom), however I checked out some YT videos of people toying around with the Darkstar:

This guy did what you described - go above 275k ft at Mach 10. It looks like he is getting 27,000 PPH and about 5300 kts. A bit slower… but at that fuel flow rate you might get almost 50 minutes out of the scramjets. Potentially above 4,000NM out of the hypersonic phase alone.

NEVO · June 3, 2022, 10:42am

great to see you Bearhedge logging some sim time again

looking at time estimate for your profile. 0:33 thats around the same time I am doing in Boeing 247 on my hops… just not the same distance…

…but I pushed it today also, was doing around 180kt IAS

PaulRix · June 3, 2022, 11:32am

You are too kind…I am still waiting for the day when I’m told it was all a big mistake and that I shouldn’t be flying real planes at all. .

IIRC I kept the aircraft at 200+ knots all the way down and crossed the threshold at about 170 Knots. You need a long runway to get it stopped when the engines are shut down. Also remember you have to use the emergency gear extension handle. I guess there is some basic modeling of a hydraulic system going on in the background although the flight controls seem unaffected… I didn’t see a RAT when I was gliding, but maybe that is what they are simulating.

Cib · June 3, 2022, 11:44am

Listen to this during the flight
https://youtu.be/-Xic-CHInek

Deacon211 · June 3, 2022, 1:21pm

Awesome stuff!

I tried this thing only a bit, but it seems like there is a lot of room for me to do some, “Stupometry” (using my Stupimeter as a reference of course) in it.

Bearhedge · June 3, 2022, 11:12pm

I’ll put together an update later but in short, you can get 4700+ NM range - I just flew from Narita, Japan to LAX. Only slightly game-breaking

Aginor · June 4, 2022, 2:22pm

Good job! Can’t wait for the AAR.

Imagine having it in neofly. If someone has that really urgent parcel.

Bearhedge · June 6, 2022, 3:13am

It’s hard to get a good session in at the moment, but I got to run a few ascent profiles today.

It’s quite fun - reminds me a lot of doing spaceplane things in Kerbal Space Program! About as realistic, also - but who cares! It’s pretty neat to get to do this in the beautiful MSFS world. Some of you might recall how in KSP the “space” music starts to play when you escape the atmosphere (70k altitude, from memory)…I was half expecting that music to start playing when I reached 275k ft and chuckled at myself

I haven’t ran the numbers yet, but there are a few things I’ll type up. Might be stating the obvious, but it helps me with my ad-hoc analysis.

Profile-wise, the flight can be divided as follows:

Takeoff
Initial climb (airfield elevation to 30-35k ft)
Supersonic transition
Supersonic climb
Acceleration to minimum scramjet speed (M3.0?)
Scramjet ignition
Scramjet climb and hypersonic cruise
Scramjet flameout
Hypersonic deceleration and descent
Turbine engine restart and throttle-up, arrest descent for supersonic cruise phase
Supersonic cruise
Final descent and landing (powered or glide)

Basically the main consideration from here is fuel flow on each of the phases - and what you do seems to make a big difference.

The first observation is that @aginor is correct - it makes little sense to light the afterburners at takeoff - the Darkstar has plenty of power for the takeoff roll and initial climb on dry thrust.

I need to do more testing, but it feels more efficient to go on almost a zoom climb after reaching 350kts or so, because the fuel flow at military power drops off rapidly above 10k feet. I’m not sure what the optimum climb speed is, yet - somewhere in the 300 kt - 400 kt range, possibly.

After getting out of the thick sea level air, the climb from 15k ft to the supersonic transition altitude is next. Getting to M0.9 in the climb seems to be hard without A/B, so the optimum speed here might be somewhere in the M0.6 - M0.8M range. Once levelled out at 30k-35k ft, the dry thrust can accelerate the Darkstar to M0.9, just.

Then it’s time to tip the nose down, light the cans and break the sound barrier.

This bit I’m not quite sure about which is the most fuel efficient way. Basically, the fast way to get up to speed is to get to a relatively steep, 20 degree nose down and then start the 1.2G pull up - doing it this way will see the altitude drop to 20k ft before coming back up, but you’ll get past M2.0 which seems to be a bit of a hurdle speed: if you’re doing M2.0+, then the acceleration to M3.0 and the climb up to 80k+ ft takes very little time…however, doing it that way results in very high fuel flow down low.

The alternative way I’ve been doing it is a gentler, slower climb. You still nudge the nose down and light the afterburners at 35k, but only tip to 10 degrees down, only descend back down to 25-30k ft and climb at M1.5 or so on a very gentle climb up to 80k-100k before accelerating to M3.0 and lighting the scramjets. The fuel flow is much less but it takes longer…so not quite sure which one is better. I think there’s a sweet spot between “too agressive” and “too gentle”, which can get you up to 80k ft and M3.0 with almost 3/4 fuel left.

The scramjet operation I’ve pretty much figured out, I think. It’s a bit game-breaking, but at the moment it seems the most range you can get is at the max altitude, which is 275k ft. You get circa M10.0 speed and the fuel flow is way lower - I’ve had 26-27,000 lbs/hr a couple times, sometimes seems to be higher, although I don’t know why.

The faster you get up there, the better, as the scramjet accelerates fast but the faster you go, the more fuel it gobbles up at lower (sub-250k ft) altitudes…the fuel flow only starts to taper off close to max altitude. The climb doesn’t take very long, so you end up with around 45-48 minutes of scramjet fuel, which gives you around 4,000 NM of range from the scramjet phase.

I put my estimates to the test by flying from Narita, Japan (RJAA) to Los Angeles (KLAX), a 4736.9 NM flight.

Taking off fron Narita - the distance showing is to the top of the approach.

Accelerating to M0.90 at 33k ft on dry thrust before lighting the A/B. FF 13,000 lbs/hr, pretty sippy.

I did the afterburner climb at M1.50 - M2.0 on this flight. It was slow. I had a little bit of jet fuel in the center tanks left at cut off, but not a lot - the climb took too long.

I lit the scramjets at M3.0 and 100k ft and climbed fairly steeply, maybe 20-30 degrees, all the way to 275k.

At 275k ft, the game shows a positive climb rate, but it isn’t correct - actually you’ve hit a ceiling and as soon as your vertical speed is negative, you start seeing the altitude to reduce (so rather than continuing to climb and the game not showing it on the altimeter, you actually do hit a hard ceiling). I just keep a modest positive climb rate here which keeps the AOA close to zero but doesn’t require accurate vertical speed management.

Not a heck of a lot happens during the hypersonic cruise, to be fair…keeping the magenta line centered does take a bit of continuous tweaking and at these speeds and air thickness you need quite a bit of bank angle for even modest changes in your heading.

North American coastline in the horizon.

Arriving to the approach waypoints. As you can see, I still had bit of scramjet fuel left at this point. I believe I could have reached Texas if I continued on, but I’m strapped for simming time so wanted to finish the flight properly rather than getting to Texas and having to turn the PC off before being able to land.

I kept the scramjets going for a wee while longer, though, whilst starting to drop down slowly.

Somewhat unsurprisingly, it took fairly robust bank angles to stay on the track at high Mach numbers.

I was aiming to decelerate to supersonic speeds by 80k ft and mostly got there. I think we decelerated through M3.0 at 75k ft or so.

Subsonic at 30k-ish ft…still “a bit on the high side” for the approach, but at least I wasn’t going to be torn apart by the speed anymore, and I assume I’m wearing a full pressure suit so don’t have to worry about popping ears.

Arrived to KLAX.

Engines spooling down. Wing tanks still full. I could have ran the scramjets to flameout, descended to 120k ft and flown at M3.5 for perhaps another 1,000 NM or so.

Some other notes:

The navigation suite is a bit limited at the moment, but the coordinates on the camera screen are handy and can be used in conjunction with Skyvector on the go. The Nav radio works, too, and gives you a green arrow on the bottom right-hand navigation screen.

So, while I don’t believe it is possible at this time to punch in a new magenta line destination if you choose to re-plan your journey mid-flight, the camera screen coordinates, Skyvector.com, a bit of pen and paper / whizz-wheeling for remaining range calculations and the navigation radio can get you to new destinations mid-flight easily enough.

PaulRix · June 6, 2022, 5:17am

Nicely done! I have to admit that the Darkstar is a lot of fun. There really isn’t anything else quite like it in the sim…the closest would be the Sim Skunkworks Starfighters, which also have to be flown on a similar profile to get up to speed (but much slower and with far less range)…

Bearhedge · June 6, 2022, 6:19am

Agreed! I like how different it is - basically an unexpected Kerbal almost-spaceplane experience in MSFS

I’m stoked because it gives me a “legitimate” way within the vanilla game to get a bit closer to the rest of the participants on the Xmas trek without breaking my self-imposed rules and spend more time flying low and slow in interesting locations. I wanted to know if it could get me to Hawaii and it can - that’s awesome!

NEVO · June 6, 2022, 2:55pm

absorbed by your Darkstar story I somehow forgot to say speedy recovery !