Pneumatic / Combustion Performance Disconnect


Postby jimmy » Thu Sep 28, 2006 11:49 am

Another problem with the "jet accelerator" approach is that you will lose a fair amount of heat to the walls of the smaller jet chamber as well as to the flow restrictor.

Of course, the other problem is that no one has gotten it to work in a spud gun. IIRC, DR gave it a pretty good shot and found that with nothing but a sabot in the barrel it seems to work. With an actual projectile it lowered the muzzle velocity.

I wonder if the problem is more than just slow combustion (a burst disk should fix that problem), but instead, if a combustion gun just loses too much energy to the walls of the chamber (which a burst disk might make even worse).

Wild ass idea, insulate the inside walls of the chamber?

What is the heat capacity of PVC? What is the mass of the chamber walls? How much combustion energy is lost if the chamber walls heat up just 10F (another number from my ass)? Is this a significant amount of the combustion energy?

Very quick calculation suggest that if the chamber walls are heated 10F then perhaps 15% of the combustion energy is lost doing that.

Heat capacity PVC = 1.047 J/gC
Density PVC = 1.4g/ml
Mass of a 4"x12" chamber ~25g?
4% propane in this size chamber ~0.0005 mole propane
heat of combustion of propane 2220 KJ/mol

(I had to do a fair number of unit conversions so there is a good chance I F-d something up along the way.)

Another possibility is that too much heat (energy) remains in the combustion gases after the shell has been fired. In a "perfect" heat engine, (a spud gun is a heat engine), the working fluid is at zero Kelvin at the end of a cycle. In a "good" engine, the working fluid is at ambient temperature at the end of the cycle. In a spud gun, the gases are probably pretty hot after the spud leaves the barrel. How much energy is lost in the hot gases?
jimmy
 
Posts: 790
Joined: Wed Jul 30, 2003 11:02 am
Location: USA
 

Postby D_Hall » Thu Sep 28, 2006 5:09 pm

OK, I'm lost. What are you talking about? I get the distinct impression that what I call "jet ignition" and what you call jet ignition are two distinctly different things. For starters, in what I call jet ignition the jet is not intended to add significant energy to the system. As such losing a fair amount of heat is no big deal provided it is still up to the task of igniting the main chamber. Also you mention a flow restrictor... To my knowledge such are not required design features (most literature makes reference to them being straight tubes but I acknowledge the one I've built in the past had a very mild throat to it(*)).

I'm left wondering if we're not on different pages.


As for the thermal transfer issues... No experience here. Is 10 F a good guesstimate for the temp increase after *A* shot? In any event, for what I'm trying to do the square/cube law is definately in my favor on this topic.



(*) Since it was expected to be used in high wind conditions the restriction helped maintain a combustible mixture.
D_Hall
 
Posts: 1496
Joined: Sat Aug 09, 2003 6:06 pm
Location: USA
 

Postby BigBang » Fri Sep 29, 2006 9:34 am

I fear that I have virtually nothing intelligent to add to this discusson. It's getting way over my head. Nevertheless, I though I'd share something I did a while back and see where it goes. Put this in the "for what it's worth" category.

D_Hall and I corresponded off line maybe a year ago about calculating first order burst disc efficiency. My goal was to attempt using GGDT to model combustion guns. My thought was to first model Latke's MiniL1 since at the time it was probably the most extensively tested gun known.

I've uploaded the GGDT file <a href="http://www.spudfiles.com/uploader/uploadFiles/MiniL1.gdt">HERE</a>

The chamber size was adjusted to yield a volume of 373ml, and the length of the barrel was adjusted for C:B = 0.8. I used pipe diameters I found from Harvel Plastics, and I used the propane combustion temperature I found at Bernzomatic's web site. The combustion pressure I obtained through my own calculations (which can be probably be found somewhere in the archives), and I used a burst disc efficieny of 50% - corresponding to D_Hall's first order calculations.

The only thing I didn't really have an estimate of was the mass of the projectile. A recent discussion in the combustion section suggests that Latke's 0.75" gasket slugs weighed on the order of 6.5gm. That seems low to me. I can't justify that statement, but it just seems low. I used a mass of 28.25gm (i.e., 1 oz.).

To my knowledge no one has ever done a direct head-to-head comparison of otherwise identical combustion and pneumatic guns. Since the performance of GGDT in modeling pneumatic guns has been verified by many, my thought was that what I was doing was as close as I was likely to get to answering the question of whether combustion or pneumatic guns are more "powerful."

The results? Well GGDT predicted a muzzle velocity of 360 fps, whereas Latke measured 506 fps. GGDT predicted good efficieny, but using the gun optimizer tool, GGDT showed that peak efficiency would occur with a C:B of 0.21. I've fiddled with as many of the variables as I can think of and have never been able to simultaneously obtain Latke's muzzle velocity and peak efficiency at a C:B of 0.8.

These results, of course, go directly against conventional wisdom concerning these things. Out of fear of being laughed off the forums I decided to keep my silly "experiment" to myself.

Of course, something may be wrong with my GGDT model of the MiniL1, but ultimately I concluded that combustion and pneumatic guns were just simply different beasts. Nothing new there, huh!

Not having the skills to pursue this further, I just set the project aside to collect dust. Now that you (i.e., D_Hall) have taken an interest in combustion/hybrid guns, might we have a hope of seeing a later edition of GGDT that addresses combustion guns? Pretty Please??

Have a great weekend all,
BigBang
BigBang
 
Posts: 155
Joined: Thu Jul 31, 2003 11:45 am
Location: USA
 

Postby jimmy » Fri Sep 29, 2006 5:15 pm

Sounds like or definitions of "jet ignition" are different. By "jet ignition" I meant essentially a small chamber that is ignited and then "jets" into the main chamber igniting it. The "jet" is hoped to pressurize the main chamber prior to its ignition.

IIRC correctly, in DRs test the "jet" was perhaps 10-20% of the main chamber and the "nozzle" was simply a hole in a plate. The "jet" chamber was a significant fraction, if you consider 10% significant. The "nozzle" would rob a fair amount of the energy from the jet just by thermal transfer.

I think it would be useful if we knew what the actual efficiency of a typical 1x combustion spud gun is. You can't use the calculated combustion energy as the energy reference point since too much energy is lost to processes other than launching the spud. Some of those processes are well known, such as the residual heat in the combustion gases, but we don't have enough data to determine the exact quantity of energy lost.

IIRC, a 70% efficient heat engine is considered to be <u>very</u> good. Most heat engines are less than 50% efficient. A typical gasoline powered car engine is only about ~30% efficient at converting the energy in the fuel to mechanical energy at the crank shaft.

How efficient is a 1x combustion spud gun? If it is 20~30% efficient then I would think it is unlikely that the efficiency can be improved much. If it is only 5% efficient then there is a possibility of getting a lot more power out of the gun.
jimmy
 
Posts: 790
Joined: Wed Jul 30, 2003 11:02 am
Location: USA
 

Sponsored

Sponsor
 
 

Postby D_Hall » Sat Sep 30, 2006 12:12 am

<blockquote id="quote"><font size="1" face="tahoma,verdana,arial" id="quote">quote:<hr height="1" noshade id="quote">Originally posted by jimmy
[br]Sounds like or definitions of "jet ignition" are different. By "jet ignition" I meant essentially a small chamber that is ignited and then "jets" into the main chamber igniting it. The "jet" is hoped to pressurize the main chamber prior to its ignition.

IIRC correctly, in DRs test the "jet" was perhaps 10-20% of the main chamber and the "nozzle" was simply a hole in a plate. The "jet" chamber was a significant fraction, if you consider 10% significant. The "nozzle" would rob a fair amount of the energy from the jet just by thermal transfer.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">

At least in my understanding and past usage of jet igniters... It sounds like you've seen a picture of one but not really understood what they are supposed to *do*.

First off, a jet igniter isn't going to do much precompression. The very fact that you've a nozzle restricting the flow (creating the jet) means that you're limiting the mass flow. Once you've done that, you've minimized any precompression effects you may have hoped for.

What a jet igniter is supposed to do is provide a means to (more or less) simultaneously ignite a LARGE portion of the fuel. Put it this way... Suppose that you have a 1' diameter chamber 10' long. Further suppose that your flame front will propogate at a rate of 30 fps.

Light the mixture with a simple spark at one end. The flame takes about 0.333 of a second to consume all the fuel as it moves from one end of the tank to the other.

Now assume that a (really really good) jet igniter can shoot a pencil width flame at Mach 3 down the center of the tank. For all intents and purposes, the flame front is instantaneously along the entire length of the tank. Now it need only travel radially. How far does it need to go? 6 inches. It now takes about 0.017 seconds to consume the fuel. While your primary flame front's velocity is unchanged, your fuel consumption rate (and pressure ramp up) has increased by a factor of 20!

THAT is what a jet igniter is supposed to do for you.


In DR's case, no offense to DR, but I don't imagine his jet was very effective if it was simply a hole in a plate. For a jet igniter to be effective it needs to produce a reasonably laminar jet that will maintain it's integrity for some distance. A sharp edged plate like that will likely produce a very turbulent flow that won't hold together.
D_Hall
 
Posts: 1496
Joined: Sat Aug 09, 2003 6:06 pm
Location: USA
 

Postby Rex_R » Sat Sep 30, 2006 1:56 am

IIRC we've calculated that standard combustion launchers use 10% of the energy availible into accelerating the projectile. somewhere(archived prolly) there is a topic labled (iirc)'thought experiment' where we hashed it out.
edit
found it
http://forums.spudtech.com/topic.asp?AR ... experiment
under combustion :)
Rex_R
 
Posts: 1064
Joined: Tue Jul 29, 2003 1:25 am
Location: USA
 

Postby aturner » Sat Sep 30, 2006 8:05 pm

<blockquote id="quote"><font size="1" face="tahoma,verdana,arial" id="quote">quote:<hr height="1" noshade id="quote">Originally posted by D_Hall
What a jet igniter is supposed to do is provide a means to (more or less) simultaneously ignite a LARGE portion of the fuel.
<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">
Is there a reasonable way for the hobby spudgunner to improve upon DR's homemade "hole in plate" jet ignition? It sounds like the key is finding a nozzle that is designed to give roughly laminar flow? But where to find that?

Or is this only witin the reach of military research labs?
aturner
 
Posts: 1470
Joined: Thu May 27, 2004 1:50 pm
Location: USA
 

Postby D_Hall » Sat Sep 30, 2006 9:02 pm

<blockquote id="quote"><font size="1" face="tahoma,verdana,arial" id="quote">quote:<hr height="1" noshade id="quote">Originally posted by aturner
Is there a reasonable way for the hobby spudgunner to improve upon DR's homemade "hole in plate" jet ignition? It sounds like the key is finding a nozzle that is designed to give roughly laminar flow? But where to find that?<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">
Absolutely. A jet igniter is well within the realm of the hobbyist.

Heck, the classic jet igniter is nothing more than a thin straight tube filled with a combustible mixture and ignited at one end. That's it! The trick is that the tube must be long enough to allow the flame front to accelerate to [some good speed]. Admittedly, getting a combustible mixture in there can be a bit of a pain but that need not be THAT hard.

What I did in my igniter(*) was have a small chamber (roughly .75" diameter x .75" long) in which I fed oxygen and acetylene in roughly stoichiometric quantities. At the end of the chamber was a tube maybe 0.375" in diameter and 6" long. I would start the gases flowing. The mixture would fill the chamber and tube in a short period of time (<1 second). At that point I simultaneously severed the gas flow and initiated a spark (via an modified automotive ignition system) in the chamber. The resultant jet was about 0.5" in diameter and about 24" long.

Admittedly, one would not want to use acetylene for a hybrid, but the basic concept should work well enough using propane. Also, while my system was automated, the timelines were not so critical that it HAD to be automated (my proof of concept was indeed manually controled)... especially since we're not dealing with wind like I was(*).




(*) It was not for spud guns. It was literally designed to start fires in aircraft "in flight" (read: sitting in a wind tunnel, engines running full bore) for onboard fire suppression system tests.
D_Hall
 
Posts: 1496
Joined: Sat Aug 09, 2003 6:06 pm
Location: USA
 

Postby aturner » Sat Sep 30, 2006 11:26 pm

Just a long thin tube with NO nozzle!? Sounds like a section of brakeline might do the trick.
aturner
 
Posts: 1470
Joined: Thu May 27, 2004 1:50 pm
Location: USA
 

Postby D_Hall » Sun Oct 01, 2006 12:47 am

A brake line may indeed do it.

I don't know how the change to propane would affect things, but if you have problems forming the jet (read: accelerating the flame front) you could always run some O2 with it. Mind you, I'm not talking about filling the entire combustion chamber with propane and O2. Just run 'em long enough to fill the igniter with 'em.
D_Hall
 
Posts: 1496
Joined: Sat Aug 09, 2003 6:06 pm
Location: USA
 
 
PreviousNext

Return to Hybrid Launchers

cron