QED-ARC Engine ISP figures

If polywell fusion is developed, in what ways will the world change for better or worse? Discuss.

Moderators: tonybarry, MSimon

Mindblast
Posts: 13
Joined: Tue Sep 15, 2009 4:39 pm
Location: Berlin / Germany
Contact:

QED-ARC Engine ISP figures

Hello all,

i'm in the process of developing an addon for Orbiter, the free spaceflight-simulator. (http://www.orbitersim.com)
What i want to do is a SSTO spaceplane using Polywell reactors, inspired by the design proposed by Dr. Bussard for example in this paper:

I have blogged about my current concept state here: http://www.orbiter-forum.com/blog.php?b=233 I would like to invite you all to comment on it.

Now my question concerns Dr. Bussards calculations for the attainable ISPs of a QED-ARC engine. In the paper above he mentions an ISP range of 1500-5500s for the ARC engine. I have run some numbers and can't quite find out how this would be possible:

Assuming a 6GW Reactor that has an electrical efficiency of 80% (i think that was about the efficiency that was assumed in the papers), that would mean 1.2GW of energy would have do be cooled regeneratively. Assuming also (like in the paper) that the reactor could still work at a temperature of 2087K and assuming LH2 as propellant:

H2 has a heat capacity of ~14300 J/(kg*K)
so the massflow required to keep the reactor at 2087K would be
mdot = 1.2e9 / (14300 * 2087) =~ 40.2 (units W * K * kg / (J * K) = kg/s)

Now when i assume the engine can convert all 6GW of reactor power to kinetic energy of the propellant i get:

eKin = 0.5 * mdot * vexhaust^2

so

vexhaust = sqrt(eKin * 2 / mdot)

vexhaust = sqrt(6e9 * 2 / 40.2) =~ 17277 (units sqrt(W * s / kg) = m/s)

ISP = vexhaust / 9.81 = 1761s (units m * s^2 / m * s = s)

So i get an absolute maximum of 1761s ISP at a reactor temperature of 2087K. So how did Bussard figure out a maximum of 5500s ? If i take that value and do the calculation in reverse i get a reactor temperature of about 20000K (!!!).

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am
Mindblast, if you've not yet seen it, look at :

On page 3 SSTO is discussed.

Dan Tibbets
To error is human... and I'm very human.

Mindblast
Posts: 13
Joined: Tue Sep 15, 2009 4:39 pm
Location: Berlin / Germany
Contact:
Yes i think i've read that paper before.

I think my problem with the ISP calculations is that i didn't take into account ionization energy. That one is huge.. i just calculated about 650MJ/kg for H2. So depending on the degree of ionization of the H2 at ~2000K the ISP in the calculations of Dr. Bussard are probably right. (After all the man should know his business, having worked on actual rocket programs.. )

For water as propellant the chemical dissociation into H2 and O2 would come into play as well (16MJ/kg of water).

Edit: Now i digged some more and from what i found ionization of H2 wouldn't take place in any significant amount below 6000-7000K... is this correct ?
Maybe its the dissociation of H2 into atoms rather.. i found a value of 435kJ/mol or ~217MJ/kg but i have no idea at what temperatures this would start to become significant.

Nik
Posts: 181
Joined: Thu Mar 12, 2009 8:14 pm
Location: UK

Lateral thinking...

Would ther be any advantage in grafting this approach onto something like the SABRE core from Reaction Engines ??

http://www.reactionengines.co.uk/

Solves the problem of trying to get additional energy into a hot-shocked hypersonic flow...

Mindblast
Posts: 13
Joined: Tue Sep 15, 2009 4:39 pm
Location: Berlin / Germany
Contact:
I got some inspiration from the SABRE research.. especially the possible density of the heat exchanger (3000m^2 / m^3, whoohooo). The thing is, Skylon is supposed to fly atmospheric up to around Mach5 i think and it uses LH2 for cooling all the time. What i'm trying to do with this Addon is fly without any regenerative cooling up to Mach4-5, just dumping the heat into the airflow. So you could cruise with Mach4-5 without using any fuel apart from a few grams of Hydrogen and Boron for the reactors. Only beyond that, when you accelerate to orbital speeds i cool regeneratively because theres probably no viable way of dumping heat into the airflow above Mach5.

MSimon
Posts: 14332
Joined: Mon Jul 16, 2007 7:37 pm
Location: Rockford, Illinois
Contact:
If you are using LH2 I would suggest a maglev slingshot to get the device up to 500 mph or so. This distributes the large amounts of hydrogen exhaust over a large volume reducing the risk of accidents. i.e you don't turn on the rocket until you are going over 450 mph.

Of course you have the problem of going supersonic at low altitude.
Engineering is the art of making what you want from what you can get at a profit.

KitemanSA
Posts: 6114
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA
With a large subsonic tow plane and a Kinetics Interchange TEther (KITE) Launcher, you can get to several mach before you turn on your motor.

MSimon
Posts: 14332
Joined: Mon Jul 16, 2007 7:37 pm
Location: Rockford, Illinois
Contact:
KitemanSA wrote:With a large subsonic tow plane and a Kinetics Interchange TEther (KITE) Launcher, you can get to several mach before you turn on your motor.
That may do for bulk cargo. I think it would be harder to get such a device man rated.
Engineering is the art of making what you want from what you can get at a profit.

djolds1
Posts: 1296
Joined: Fri Jul 13, 2007 8:03 am
MSimon wrote:If you are using LH2 I would suggest a maglev slingshot to get the device up to 500 mph or so. This distributes the large amounts of hydrogen exhaust over a large volume reducing the risk of accidents. i.e you don't turn on the rocket until you are going over 450 mph.

Of course you have the problem of going supersonic at low altitude.
IIRC one of Bussard's early IEC ideas was to raster-scan an REB across an airflow moving at ram speeds. Use the QED/ARC rocket to boost to the first 350m/s, switch over to raster scanning air flow in a thrust tunnel up to M5-M8, and then switch back to QED/ARC. Call it a QED-RBCC cycle.
Vae Victis

KitemanSA
Posts: 6114
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA
MSimon wrote:
KitemanSA wrote:With a large subsonic tow plane and a Kinetics Interchange TEther (KITE) Launcher, you can get to several mach before you turn on your motor.
That may do for bulk cargo. I think it would be harder to get such a device man rated.
Tow systems are man rated now. this just adds some aerodynamics.

MSimon
Posts: 14332
Joined: Mon Jul 16, 2007 7:37 pm
Location: Rockford, Illinois
Contact:
KitemanSA wrote:
MSimon wrote:
KitemanSA wrote:With a large subsonic tow plane and a Kinetics Interchange TEther (KITE) Launcher, you can get to several mach before you turn on your motor.
That may do for bulk cargo. I think it would be harder to get such a device man rated.
Tow systems are man rated now. this just adds some aerodynamics.
A sling shot vs a straight tow is going to have some serious questions asked. Like how do you "guarantee" the towed vehicle NEVER hits the towing vehicle? Or the tow line? Or that the tow line never hits the towing vehicle.

The problem is not straight forward because there are way more possible failure modes than in a straight tow.
Engineering is the art of making what you want from what you can get at a profit.

Mindblast
Posts: 13
Joined: Tue Sep 15, 2009 4:39 pm
Location: Berlin / Germany
Contact:
Well my current idea here is to just use some ducted fans driven by turbines in the reactors cooling loop for taxiing/takeoff/climbout and acceleration to M0.9 or so. This would give the thing the capability to start and land on a normal airport and also gain some altitude before switching to airbreathing QED Mode to avoid the ozone production thing Bussard mentioned close the ground.
Also my concept wouldn't use the "rasterscanned REB into the airstream on the underside" concept but rather normal QED engines that can use any mix of air and hydrogen as propellant mass. Don't know if there is any problem with that.

KitemanSA
Posts: 6114
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA
MSimon wrote: A sling shot vs a straight tow is going to have some serious questions asked. Like how do you "guarantee" the [1:] towed vehicle NEVER hits the towing vehicle? [2:] Or the tow line? [3:] Or that the tow line never hits the towing vehicle.

The problem is not straight forward because there are way more possible failure modes than in a straight tow.
1: Towed vehicle doesn't hit tow plane, unh .. its MANY kilometers away and not flying in the same direction, except while being straight towed to the staging area.

2: Upon release, the tow line and vehicle seperate at several G.

3: The towline was, is and (except for a very short time just at release when using a two stage unit, always behind the tow plane. And while it is "ahead" it is many many kilometers away and decelerating rapidly.

djolds1
Posts: 1296
Joined: Fri Jul 13, 2007 8:03 am
Mindblast wrote:Well my current idea here is to just use some ducted fans driven by turbines in the reactors cooling loop for taxiing/takeoff/climbout and acceleration to M0.9 or so. This would give the thing the capability to start and land on a normal airport and also gain some altitude before switching to airbreathing QED Mode to avoid the ozone production thing Bussard mentioned close the ground.
Its just those first few hundred meters per second dv before the ram-air effect is possible. Twelve seconds at 3gees to boost upto 350m/s. Early designs such as the Saenger Silverbird and Soviet Keldysh used sleds, but the efficiency of QED/ARC makes a sled superfluous. Twelve seconds of low altitude QED/ARC burn per flight doesn't seem very significant. And using one type of engine (QED/ARC-RBCC) eliminates additional complexity.
Mindblast wrote:Also my concept wouldn't use the "rasterscanned REB into the airstream on the underside" concept but rather normal QED engines that can use any mix of air and hydrogen as propellant mass. Don't know if there is any problem with that.
Oh, IMO forget "Rasterscan the underbody." Instead, scan the REB inside a "standard" (sc)ramjet engine tube. The forebody/diffuser locks forward for "pure" ARC mode, but retracts to allow airflow. Place the REB generator in the engine tube like the rocket engine in a "standard" RBCC. Any unconditioned external mass augmentation should improve thrust and Isp. The problem with SABRE and the RB545 was that the mass of the precoolers defeated the theoretical performance advantage.
Vae Victis

93143
Posts: 1131
Joined: Fri Oct 19, 2007 7:51 pm
djolds1 wrote:The problem with SABRE and the RB545 was that the mass of the precoolers defeated the theoretical performance advantage.
As far as I know, the SABRE precooler is still expected to work. A piece has been manufactured and demonstrated, and they've solved the icing problem. They now appear to be working on manufacturing a complete precooler with proper quality control, so that they can build an actual engine. They've got an interesting side project going to try to come up with an advanced altitude-compensating nozzle for the thing... anyway, they haven't given up on it.

IIRC the big problem with the RB545 was that it was at the back, which made HOTOL very unstable during reentry. Skylon solves this.

Also, those few hundred metres per second could benefit from an injector-jet arrangement, couldn't they? And I really think some effort should be put into trying to burn the hot hydrogen exhaust productively, while it's still in the engine and can do some good. At least it should be able to effectively scavenge ozone from the REB-heated air coming down the ram duct...