Ramjet Using Dynajet Pipe?

[larry cottrill]

Back a few years ago I posted the graph shown below. It is a sort of weird idea to use a Dynajet tailpipe (which includes the chamber, of course) as the chamber and nozzle section of a ramjet. As you'll see on the graph, my idea was that such a ramjet should be able to give us TWICE the thrust that the same pipe does as a pulsejet (silly me). So, I ran the graph shown using a Java applet I coded based on the gas formulas in my old jet engine book. There is an obvious flaw, in that the 'critical' values for pressure, etc. are not right, but those would only hurt us if we got up to supersonic flow, which we certainly don't in this case. I think the numbers shown in the graph are right, though - if anybody can show them to be wrong, it would be appreciated.

The values of pressure, speed, temperature, etc. are calculated for each of 100 uniformly spaced stations throughout the nozzle region only, from largest to smallest diameter. 100 points were chosen, so the resultant graphs would be reasonably smooth. The heavy black lines attempt to represent the contour of the physical nozzle.

This is not a very serious idea for a project anyone would really build. However, there are some points of interest in the numbers and their graphical representation. One thing I pointed out before is the sharpness of most of the curves as they get to the throat of the nozzle. I originally found that quite surprising, though it makes sense when you see the equations, especially the exponents that have to be used. This shows something about how critical the throat area is in shaping the exhaust gas condition at exit.

But looking at this again today, what I find really surprising is how little the engine really has to do to get this level of theoretical performance. We assume 8.5 lb continuous net thrust and .28 lb/sec massflow, both about twice Dynajet values. The standard air being aspirated somewhere to the left of the chamber is assumed to be 520 degFabs and 14.7 PSIa air, which would have a specifiic volume of 13.33 ft^3/lb. Now, note what the chamber (and diffuser) of this engine have to do: Heat the air a little over 2000 degF; raise its pressure less than TWO PSI (16.262 - 14.697 = 1.565 PSI increase!) and deliver the heated air to the large end of the nozzle zone at 229 ft/sec! The nozzle does its magic and delivers the exhaust stream at atmospheric pressure and 977 ft/sec speed, with less than 100 degF temperature drop! The specific volume goes from 27.6 cu.ft/lb up to 29.7 cu.ft/lb due to the pressure drop experienced.

If these numbers are right, I find that pretty amazing. Less than 2 PSI to drive that nozzle up to a decent exhaust speed and thrust value! This seems like something so doable - you just need a good, efficient front-end design to thread onto the front of the pipe, with some kind of built-on flameholder. Of course, I'm pretty sure the forward airspeed would need to be pretty high just to achieve that 1.565 PSI pressure rise, even with a really well-designed diffuser. Oh, well ...

The legend for the graph is: Area A = magenta; static pressure p = blue; specific volume v = green; temperature T = red; speed u = white; thrust force from gas momentum f (mass flow x speed) = yellow.

The total thrust force forward at the left end would be the thrust from momentum PLUS the static pressure x area = 2.0 + (0.034 x 144.0 x 1.565) = 9.66 lb, so the nozzle itself is taking up only 9.66 - 8.5 = 1.16 lb, which seems very low - note that the exit area is just 1/4 of the entrance area. [Note: The 144.0 factor is to get the area into square inches, making it compatible with pressure in PSI.]

Comments / criticisms appreciated. I would like to verify that the output values delivered by this program are reasonable theoretical values.

L Cottrill

[Eric]

Everything sounds good in theory, until you actually make one and find out that it consumes 5 times as much fuel as a pulsejet at pulsejet speeds. The very thought is scary.

It is still summer and I dont feel like doing any of that fancy math stuff until absolutlely necessary. Its an interesting idea, but now adays if it doesnt go mach 15 it just aint good enough ;)

Eric

[larry cottrill]

Eric -

Yes, I know what you mean. But, this isn't a hot idea to try to compete with anything, on any basis. The only reason I thought about it a while back is that the Dynajet pipe gives you such a nice chamber and nozzle, with a really good area ratio I think, and it would be so easy to use - just make a different front end and thread it on with the lock ring! If I had a decent size lathe sitting around, I'd have actually tried this, long ago.

Of course it would be Gas Hog City - all "pure jets" are, until you can get them moving a few hundred MPH. That would be the least of my problems, in the early stages anyway. It would just be something really fun to try, and to get "quasi-perfected".

I wonder if using the long pipe would give it too great a tendency to pulse. My intuitive answer would be no, since I've seen so many valveless engines driven in to smooth-burning "blowtorch" mode with just a bit too much starting airflow. But, I can see the potential for a problem there. Or maybe for a coup: the valveless that turns into a ramjet at [insert reasonable value here] MPH ...

L Cottrill

[Eric]

Well I think the pipe is too long for a number of reasons. The longer the pipe the farther away you get from an ideal ramjet design. The intake velocity may be subsonic but the exhaust could/should be > m1 so an expanding nozzle would be best to accelerate the flow and it would need to be fairly short. Most of the engine could be used though, and you could probably even use a stock intake minus the valves and have a fuel injecting "retainer" for a flame holder.

If you trim down a stock dynajet head you can get it down to 1/4 pound pretty easy, and just have a aerodynamic cowl over it. Total engine weight could be around half a pound with considerably excessive thrust. Something that size could be launced as a second stage of a rocket rockets pretty easy. Of course you have to go to a country that doesnt care about launching things above the speed of sound and all.

Now if you made one out of a tigerjet sized engine you could definately launch it with model rockets, or even run it off of some high volume air source.

Maybe a vacuum cleaner from hell with a nozzle which constricts to the intake diameter. A gas powered leaf blower should be able to provide considerable MPH at such a small area. Infact I might try that tomorrow. Maybe something with a max diameter of 1.5".

Or my favorite concept of a ramjet augmenter...... hmmmmmm too many thoughts for this time of morning

Eric

[larry cottrill]

Maybe a vacuum cleaner from hell with a nozzle which constricts to the intake diameter. A gas powered leaf blower should be able to provide considerable MPH at such a small area. Infact I might try that tomorrow. Maybe something with a max diameter of 1.5".

Eric -

My B&D Leaf Hog was the strongest electric I could find without a lot of research, and they only claim 125 MPH maximum. I'm sure there are big gas powered ones that do better. However, you really would need almost sonic speed to get the Dynajet head to perform well. The reason is that the diffuser ratio is really low. It only has a slight diffuser effect (remember - you must base the ratio on the entrance area, which is much larger than the throat area). Of course, if you didn't care about carburetion, you could modify the head to a much smaller entrance area, just a fraction of an inch in front of the throat!

I hate to throw on a wet blanket, but it might be harder to get up to supersonic speed in a 1.25 inch outlet than you think ;-) The critical pressure ratio for air is .528, which means the chamber pressure would need to be almost twice what you have in the throat - compare p1 and p2 in the graph to see how close we are in the example I gave - and that's for 8.5 lb thrust! And unless you can achieve that, you won't have the sonic speed in the throat that will make a deLaval nozzle work for you. So, I think I'd try it first with the unmodified Dynajet pipe, and see what you can get!

Good line of thinking, though. I basically just think we'd need to do a lot better on the diffuser design than what the original head would ever give us, no matter how we modified it. There's nothing wrong with the entrance area per se - it's about the same as the exhaust pipe area, which I think would be OK.

L Cottrill

[Eric]

ROFLMAO .....

Well I wasnt talking about getting supersonic flow from a leaf blower.... but that would be sooo much fun in itself wouldnt it. Probably would turn those pesky leaves into mulch and rip up the lawn.

Even mach .8 or so would be very significant. I have seen some gas leaf blowers that claim 200mph with a low restriction nozzle. That could easily be boosted to 400 mph which is already above .5 and into low ramjet speeds. The engine wouldnt have to necessarily go supersonic for the nozzle to work, so long as the exhaust is.

[larry cottrill]

Even mach .8 or so would be very significant. I have seen some gas leaf blowers that claim 200mph with a low restriction nozzle. That could easily be boosted to 400 mph which is already above .5 and into low ramjet speeds. The engine wouldnt have to necessarily go supersonic for the nozzle to work, so long as the exhaust is.

You need to be careful doing that. With a centifugal blower / impeller, I know you can't stall it by blocking the intake - but I don't know what happens when you increase the load on the exit side! If the blower is designed so that the engine is giving all it has to get that 200 mph, it may just bog down from the increased "back pressure" of a smaller nozzle. Of course, it would't hurt to try it - you could tell right away whether it was giving you what you wanted without undue strain.

When I was trying to get my NASA ISBR grant for Cyclodyne(TM) development, I tried to buy a big fan from a grain drying bin that was being dismantled, but it turned out the contractor had dibs on everything he took out. I couldn't really have used it anyway - it was like 30 or 40 HP, three-phase.

L Cottrill

[larry cottrill]

Even mach .8 or so would be very significant. I have seen some gas leaf blowers that claim 200mph with a low restriction nozzle. That could easily be boosted to 400 mph which is already above .5 and into low ramjet speeds. The engine wouldnt have to necessarily go supersonic for the nozzle to work, so long as the exhaust is.

Well, Eric, you really got me thinking about this again! And here's what I think: What we need is the air end (not the turbine end) of a turbocharger. I know the turbojet boys could tell us what a good one would be and what speed we should spin her up, but I'll bet they CAN'T tell us (without some head-scratching, anyway) what size engine or motor would be needed to do it! Ha!

Man, one of those things with a little Honda engine and a heck of a gear train would pump out a nice breeze, eh? Probably, forced lubrication required.

L Cottrill

[pezman]

I did a little playing around with designing nozzles, supersonic wind tunnels etc. using UFLOW.

The basic idea is just to increase the ambient pressure on the intake side until you get the steady state flow that you want at a particular point (e.g. say 200 mph at the entrance). The simulation starts out with a bunch of transient artifacts, but quickly settles into a steady-state solution.

It works pretty well. I was mostly using it to see how intense a vacuum you could pull with a venturi that is fed propane at 4 atm. It's cool to see how a choked nozzle opening into a wider area results in a super-sonic flow with remarkably low presure.

At any rate, it should be relatively easy to adapt the technique to designing ramjet chambers.

[larry cottrill]

pezman -

I tried UFLOW also, using the contours of the Maggie Muggs design. It seemed to give very good results on the diffuser. I used a small necked-down area to represent the draggy flameholder with all its little holes. Pretty crude, but the results looked promising (except when the 'velocity head' pressure difference was pretty low).

Note that you should run the graphic display long enough for the Massflow to become a flat line (constant throughout the duct) - that's the actual steady flow condition. At that point, you'll see that there's no bouncing around of the pressure, velocity and density curves, so you can accept them. But, I'm probably telling you what you already know!

The combustion chamber needs to be done separately, and I have yet to perfect the technique. It doesn't do you any good to set up 'Initial Temperature' and 'Initial Pressure' values in the hot stations of the pipe - they immediately get "blown away". The thing to do, I guess, is to set up the chamber pressure and temperature externally at the left end. But I haven't re-entered the data to try it yet, although I do have the chamber contours entered.

You may get a laugh out of this: What I originally tried to do is model the whole engine. Of course, I quickly discovered that without a way to maintain constant pressure and temperature in the combustion zone, all I had was a fancy-contoured air nozzle! The one thing that was interesting to find was the sharp pressure drop through the draggy diffuser. Of course, I knew that would be there, but it was fun to see it verified in the plotted curves.

L Cottrill