SpudFiles

In another thread Rmich732 kicked up perhaps the most contentious question in combustion guns.

1. Everyone agrees that there is an optimal C:B for a particular gun.

2. Usually the optimal C:B is somewhere in the range of say 0.7 ~ 1.0 for a basic combustion gun. Usually, folks assume that the number is actually 0.8. (Sometimes folks assume the number is 1.0 or 1.5.) It is assumed that changing things like the number of sparks or the fuel has an affect on the optimal C:B but the magnitude of these affect is not well understood.

3. For a particular chamber volume there is an optimal barrel volume (see points 1 and 2 above and Latke's data).

and the million dollar question....

4. For a particular barrel what is the optimum chamber volume? Is it the "Latke" C:B of 0.8?

We will define "optimal" and "maximum performance" as maximizing muzzle velocity, which is not the same as maximizing the efficiency of the gun.

If we make two guns with the same barrel;
a.) One of the guns has a chamber sized using the measured optimal C:B for the gun (for example you based the gun on Latke's)
b.) The other gun has a marginally larger chamber.

Which of these two guns will give higher muzzle velocities?

Assuming that by "marginally" larger we are talking about the C:B going from say 0.8 to 1.0. We are not talking about "mine" sized chambers or the C:B going to 10:1.

Bolingleadbath made an interesting point in Rmish's thread. There are perhaps two conflicting models of a basic combustion spud gun.

We will call the first model the "Fast Combustion" model. In the fast model combustion is complete (or is essentially complete) before the spud has moved significantly.

The second model is the "Slow Combustion" model. In this model, the spud starts to move long before combustion is complete.
(I tried to paraphrase BoilingLB, hopeful I'm pretty close to what he meant.)

These two models give very different predictions of the performance of a combustion gun.

One thing of interest is that in the fast model the combustion process is completely independent of the spud. The combustion process is essentially a "closed chamber" process.

In the slow model the combustion process (flame speed, pressure, burn time ...) responds to the movement of the spud.

The fast model would have many characteristics in common with a compressed air gun. In particular the pressure versus time curve would be qualitatively similar for the two types of guns. (Though the gas density is much lower in a combustion gun and heat loss affects a combustion gun much more than a compressed air gun.)

The slow model has little in common with a compressed air gun. In particular, there is no similarity between the pressure versus time curves of a combustion and a compressed gas gun.

The fast model predicts that an oversized chamber will increase performance. The fast model predicts that even a very large chamber will increase performance. ("Very large" is perhaps a C:B of 10:1)

The slow model predicts that ... not really sure what it predicts ... Depending on the details of the model and the details of what is really happening in the gun, and the characteristics of the gun, the slow model might predict a slightly larger chamber is better or it might predict that the optimal is the C:B predicted based on a variable barrel.

Anyone have any data for fast versus slow combustion?

I've got the "pressure" versus time data I recorded with a <a href="http://home.earthlink.net/~jimsluka/Pie ... tml">piezo and PC</a>. Clearly the P vs T curve is of a slow combustion process.

BoilingLB thinks Latke's data with multiple spark gaps supports the fast model. (Or the fast model is adequate to explain the experimental data.)

Any other thoughts?

Any thoughts, or better yet data, on the important question of a "reasonably" oversize chamber increasing performance for a particular barrel?

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Cras nec placerat erat. Vivamus dapibus egestas nunc, at eleifend neque. Suspendisse potenti. Sed dictum lacus eu nisl pretium vehicula. Ut faucibus hendrerit nisi. Integer ultricies orci eu ultrices malesuada. Fusce id mauris risus. Suspendisse finibus ligula et nisl rutrum efficitur. Vestibulum posuere erat pellentesque ornare venenatis. Integer commodo fermentum tortor in pharetra. Proin scelerisque consectetur posuere. Vestibulum molestie augue ac nibh feugiat scelerisque. Sed aliquet a nunc in mattis.

I've dropped the latke spark-gap tests as strong support of the fast combustion scenario, on account of the geometric differences being complicated.

However, the fact that the <2" long barrel of the 3/4" latke tests gave us as much speed as it did indicates that pressures where in the vicinity of 70 psig - average - during those two inches.

As a random note, EVBEC predicts that velocity will be maximized for a 2"x48" barrel and 3 oz spud if the chamber is 4"x140" (+- 5").
I thought that was pretty interesting, although it's almost certainly numericaly worthless.

I always think "bigger have to be better"
So on my combustion i have a chamber that is 9 Liter (550 Cubic Inch), and a barrel that is 4 Liter (245 Cubic Inch), and that is realy powerfull, shots golfballs over 350 Meters. I have bought steel pipe the same size now, and i'm gonna make a 8-9 Liter hybrid out of it!

Spudbalster:

It sure seems as if multiple sparks have no effect on the ideal ratio (or burn rate), because even with 4 sparks, burst disks still improved the performance of a marginally large ratio cannon. But that is another debate.

What are the times on that page supposed to be? Hang times?

I think it is pretty tough to use the advancedspuds data to prove anything concerning a basic combustion gun. Isn't Quadzilla a 2x hybrid with four sparks? If I'm reading the table correctly the larger CB's always underperfom the 0.8 gun.

BoilingLB: You are right, the 2" long barrel sure suggests a fast combustion. Need to cogitate on that one for a while.

EVBEC predicts that velocity will be maximized for a 2"x48" barrel and 3 oz spud if the chamber is 4"x140" (+- 5").

Wait a minute, that is a barrel volume of 151in³ and a chamber volume of 1760in³ for a C:B of 0.08! Is EVBEC prdicting an asymtote? (Which migh be reasonable) Or is this prediction just nonsense because none of the guns EVBEC is based on is anywhere near this theroretical gun?

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Cras nec placerat erat. Vivamus dapibus egestas nunc, at eleifend neque. Suspendisse potenti. Sed dictum lacus eu nisl pretium vehicula. Ut faucibus hendrerit nisi. Integer ultricies orci eu ultrices malesuada. Fusce id mauris risus. Suspendisse finibus ligula et nisl rutrum efficitur. Vestibulum posuere erat pellentesque ornare venenatis. Integer commodo fermentum tortor in pharetra. Proin scelerisque consectetur posuere. Vestibulum molestie augue ac nibh feugiat scelerisque. Sed aliquet a nunc in mattis.

As I have mentioned before the latke tests should not be depended upon. although they were carried out perfectly, there is just not enough data.
Only a few cannons were tested, all in different tests.
Also, the projectile was always a spud which is not very uniform for projectiles. The latke tests use of potatoes therefor indicates a fast combustion, due to potatoes characteristic of fitting in the barrel with much friction. (same with tennis balls)

Benstern: Good points but Latke used spuds and a gasketed wood rounds. You really can't "depend" on his data, but it is just about the only data there is.

Spudblaster said:

The numbers are hang times, and the gun used was a metered propane fueled combustion with four sparks, not a hybrid.

Well, actually, with a burst disk isn't the gun more like a hybrid than a standard combustion? (Isn't that what is meant by "1 layer, 2 layer" ..., that is the number layers of ??? used as the burst disk?)

Spudblaster said:

The reason the larger ratios performed better than the smaller ratios is the fact that the test was based on variable barrel lengths, and so the shorter barrels performed worse (as expected).

I must not be reading the data the way you are. In every case the larger CB values underperformed the 0.8 gun. For example, the control gun;
C:B=0.8, hang time 11.87 sec, WAG velocity=698 FPS
C:B=1.5, hang time 11.18 sec, WAG velocity=569 FPS
C:B=3.0, hang time 10.10 sec, WAG velocity=414 FPS
C:B=6.0, hang time 7.59 sec, WAG velocity=197 FPS
(WAG velocity is based on a terminal velocity of 90 FPS and a mass of 100g)

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Cras nec placerat erat. Vivamus dapibus egestas nunc, at eleifend neque. Suspendisse potenti. Sed dictum lacus eu nisl pretium vehicula. Ut faucibus hendrerit nisi. Integer ultricies orci eu ultrices malesuada. Fusce id mauris risus. Suspendisse finibus ligula et nisl rutrum efficitur. Vestibulum posuere erat pellentesque ornare venenatis. Integer commodo fermentum tortor in pharetra. Proin scelerisque consectetur posuere. Vestibulum molestie augue ac nibh feugiat scelerisque. Sed aliquet a nunc in mattis.

Thinking some more about "slow" versus "fast" combustion, I've realized that what some people call slow might be what others call fast. In really depends on how "speed" is measured.

I suspect that some of the disagreement that this topic brings up is caused by people measuring "speed" differently. You might think that the meaning of slow and fast and time and speed would be pretty obvious but I've just realized that it really isn't. I'm not just talking about "Joe thinks 20mSec is fast but bob thinks it is slow". I'm talking about two fundamentally different ways of looking at, and measuring, slow and fast.

The graphs below were generated with my combustion model. The guns is; 3"Dx11"L chamber, 2"Dx30" barrel, 0.8 CB, 80g apple, metered propane, chamber fan (those two aren't modeled), single spark at the center of the gun. Measured muzzle velocity for this gun is 330 FPS (101 m/s), predicted velocity is 311 FPS.


The two graphs are from the same modeling run. Both graphs have pressure (absolute) on the Y-axis (vertical). The upper graph shows the pressure as a function of time. The lower graph shows the pressure as a function of the position of the spud in the barrel (similar to what Evbec does).

The model predicts that the optimal barrel for this gun is 39.2" (1m), about 9" longer than the gun's actual barrel. The optimal barrel length is marked on the graphs by a vertical dashed line with the triangle markers.

The upper graph also has a vertical line (with circle markers) indicating the time at which the spud started to move. So, for the upper graph, the movement of the spud through the barrel occurs between the two dashed lines.

Now looking at the graphs, is this a "slow" or a "fast" combustion process?

Looking at the lower graph we see that the peak pressure occurs when the spud has moved only ~20% of the length of the barrel. I would say that is a "fast" process. Combustion is largely complete before much of the barrel has been used. (Actually, you really can't judge if "combustion is largely complete" by this graph. All you can say is the peak pressure occurs early.)

Looking at the upper graph we see that the peak pressure occurs when about 1/2 of the transit time has passed. The upper graph suggests that the rise and fall of the chamber pressure occurs fairly symmetrically relative to the transit time of the projectile through the barrel. Could this be considered a "slow" process?

So, is the disagreement about "slow" versus "fast" (particularly as it relates to multiple sparks or oversized chambers) being confused by fundamentally different definitions of "slow" and "fast combustion"?

My brain hurts.
EDIT: Fixed dead link to the graph.

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Cras nec placerat erat. Vivamus dapibus egestas nunc, at eleifend neque. Suspendisse potenti. Sed dictum lacus eu nisl pretium vehicula. Ut faucibus hendrerit nisi. Integer ultricies orci eu ultrices malesuada. Fusce id mauris risus. Suspendisse finibus ligula et nisl rutrum efficitur. Vestibulum posuere erat pellentesque ornare venenatis. Integer commodo fermentum tortor in pharetra. Proin scelerisque consectetur posuere. Vestibulum molestie augue ac nibh feugiat scelerisque. Sed aliquet a nunc in mattis.