Ratio of marble cannon

[tamirchiki]

hello , i don't understand the all subject of ratio . and now i want to build a Marble size combustion, so i want to ask you what is the best ratio for this cannon (for the best performances ), and how i calculate him . and i saw that there are loud cannons and quiet cannons , what the differences between these two kinds of cannons. And i want a quiet cannon , and i want also to add a silencer so tell me or show me what the best silencer i can make (i mean one that really work ).

so thanks to the helpers !

[Pete Zaria]

Welcome to the forums.

Man, you need to search around. This question has been asked 10x before.

I'll spoonfeed you a bit:

Ratio refers to the volume of the combustion chamber to the volume of the barrel. This is the C:B ratio (combustion chamber:barrel ratio). Measurements are in VOLUME, not in length or width or anything else.
This site calculates the volume of a piece of pipe for you: http://www.online-calculators.co.uk/vol ... volume.php

"Proper" ratios, as generally accepted by the spudgun community:

For propane, .8:1 c:b is the most efficient.
For hairspray or other aerosols, anywhere from 1:1 to 1.5:1 is fine.
For pneumatics, anywhere from 2:1 - 4:1 is generally best, depending on the launcher.

If the chamber is too small, the projectile will loose speed in the barrel. If the chamber is too big, the cannon will be too loud, but there's no real negative effect. For this reason, I always tend to go chamber-heavy.

For a quiet-ish combustion marble gun, I'd just use a roughly 1:1 ratio. I wouldn't use anything bigger than 2" pipe for the chamber, either.

Peace,
Pete Zaria.

[tamirchiki]

thanks . i searched this subject before i ask ,but now i understand better thanks to you!

[Brooner115]

Dude than i have like a 4:1 ratio on my pneumatic SWEET

thanx a lot Pete u really helped me a bundle
and thank u tarmirchiki for asking the question.

No wonder I can penetrate 3/4 inch ply wood with a potato Thanx guys

[jimmy101]

If the chamber is too big, the cannon will be too loud, but there's no real negative effect. For this reason, I always tend to go chamber-heavy.

This is not true for combustion guns. A too-big chamber will decrease the muzzle velocity of the gun. The maximum muzzle velocity for a combustion gun occurs at the ideal C:B ratio for the particular gun. The ideal C:B ratio varies somewhat depending on the fuel, barrel diameter, projectile etc.

An oversize chamber on a combustion gun decreases the performance of the gun.

[fom]

If the chamber is too big, the cannon will be too loud, but there's no real negative effect. For this reason, I always tend to go chamber-heavy.

This is not true for combustion guns. A too-big chamber will decrease the muzzle velocity of the gun. The maximum muzzle velocity for a combustion gun occurs at the ideal C:B ratio for the particular gun. The ideal C:B ratio varies somewhat depending on the fuel, barrel diameter, projectile etc.

An oversize chamber on a combustion gun decreases the performance of the gun.

HA, the only thing that decreases performance is too short of a barrel or too long of a barrel for a combustion.
If the barrel is too short the projectile won't go far at all because all this pressure is wasted.
If its too long all the pressure will be used up and the potato will slow down because of friction.

[jimmy101]

If the chamber is too big, the cannon will be too loud, but there's no real negative effect. For this reason, I always tend to go chamber-heavy.

This is not true for combustion guns. A too-big chamber will decrease the muzzle velocity of the gun. The maximum muzzle velocity for a combustion gun occurs at the ideal C:B ratio for the particular gun. The ideal C:B ratio varies somewhat depending on the fuel, barrel diameter, projectile etc.

An oversize chamber on a combustion gun decreases the performance of the gun.

HA, the only thing that decreases performance is too short of a barrel or too long of a barrel for a combustion.
If the barrel is too short the projectile won't go far at all because all this pressure is wasted.
If its too long all the pressure will be used up and the potato will slow down because of friction.

Uh, that is exactly what I said. Optimum performance occurs at the ideal C:B. If the barrel is too long or too short then you are not at the optimal C:B.

And, "If its too long all the pressure will be used up and the potato will slow down because of friction". Actually, the pressure drop is probably due more to heat loss than it is to "us(ing) up all the pressure".

[Pete Zaria]

An oversize chamber on a combustion gun decreases the performance of the gun.

What's your logic here? Do you have any examples or math to back up your statement?
I can't comprehend why putting *more* hot, expanding gas behind a projectile could possibly make it go *slower*. It would be less efficient for sure, as lots of power would be wasted in the form of noise and muzzle flash - but I can't see how this could possibly decrease projectile velocity.
Explain, please?

Peace,
Pete Zaria.

[Scope]

An oversize chamber on a combustion gun decreases the performance of the gun.

What's your logic here? Do you have any examples or math to back up your statement?
I can't comprehend why putting *more* hot, expanding gas behind a projectile could possibly make it go *slower*. It would be less efficient for sure, as lots of power would be wasted in the form of noise and muzzle flash - but I can't see how this could possibly decrease projectile velocity.
Explain, please?

Peace,
Pete Zaria.

I agree all that having a bigger chamber means is that after the projectile leaves the barrel there is still more gas expansion happening.

[boilingleadbath]

Well, fom, there's always the salt mine hypothesis.
That is where the chamber is so big that the combustion takes too long, so the chamber pressure never gets very high.

Jimmy is a strong advocate of the slow-combustion data - in which case, the salt mine effect becomes important at marginally over sized chambers.

Personally, I think he's over stating the effect, going by two bits of latke data:
1) The pressure curve of a spudgun (as derived from the latke C:B test data) matches a 1/barrel length curve fairly well.
2) The latke spark gap test showed only modest improvement from more sparks. (not what I'd expect if the rate of combustion was a major limiting factor)

And anyways... the salt mine situation can be fixed by just using a heavier projectile.

[SpudBlaster15]

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.

[jimmy101]

Spudblaster put it very well.

The Latke spark strip / multiple spark data is of only limited use. Indeed, I looked at his comparison of two versus three sparks on the strip and there is no statistically significant difference in muzzle velocities.

As you can see in the graph above there really isn't any difference between two and three sparks on the strip. (The Student's T-test values for the two averages is 0.33.) Furthermore, there is an outlier value for the 3-spark data (marked with an arrow in the graph). Without that one data point the two averages are even closer to each other. (The data is from http://www.burntlatke.com/strip.html)

Nobody has ever measured and reported hard numbers on how much multiple sparks affect the performance of a gun other than Latke's strip data. We really don't know how much multiple sparks boost performance when the sparks are not on a strip. Like Spudblaster15 said, the positioning of the spark strip limits the area of the burn front since the strip is up against the wall of the chamber. Basic physics would suggest that two sparks on a strip are roughly the same as a single spark at the center of the chamber (using screws for electrodes for example).

As to why a too large chamber would decrease performance in a combustion gun...

Obviously, you want as much pressure behind the spud as possible as the spud moves through the barrel. But combustion of propane + air is a fairly slow process. It takes tens of milliseconds from the time the fuel is ignited until the spud exits the barrel. So the sequence of events is not;
- You hit the trigger
- The fuel burns and maximum pressure in the chamber is reached
- The spud starts to move.

If that was the course of events then a burst disk gun wouldn't be any better than a standard gun.

To understand why the chamber size is critical you have to understand the combustion process.

1. The final temperature and pressure obtained in a closed chamber is independent of the chamber size (assuming an adiabatic process). However, the burn rate of propane + air is a function of the chamber size. A large chamber takes longer to fully combust then a small chamber. Why? Because combustion of 1 unit of fuel in a large chamber heats the rest of the gases less than the combustion of 1 unit of fuel in the small chamber. To put it another way, with a small chamber the acceleration of the burning of the fuel is faster because less combustion is needed to heat the remaining fuel. In a large chamber everything happens slower.

2. As the pressure in the chamber rises the spud starts to move as soon as the pressure*area exceeds (atmospheric pressure*area)+static friction values. If we figure 30 pounds of static friction and a 2"ID barrel, the spud starts to move when the chamber pressure exceeds just 10 PSIG. Complete combustion of propane in air has a peak pressure of about 120 PSIG (again for an adiabatic system). So our spud starts to move when the pressure has risen just 8% of the way to the maximum theoretical pressure.

3. If the chamber is very large the spud starts to move when the rate of pressure rise is still rather small. If the chamber is sufficiently large then the actual movement of the spud (and the resulting increase in chamber volume) is a small factor. As a result, the combustion process continues at the slow burn rate characteristic of a large chamber. If the barrel is too short for the chamber then while the spud is moving the pressure in the chamber is never very high and you run out of barrel before combustion is anywhere near complete.

4. For a properly sized chamber, once the spud starts to move things get complicated. The gases in the chamber expand and cool. The rate of combustion slows down relative to what it would have been if the spud hadn't moved. Nonetheless, the smaller chamber size means the unburned gases are getting heated faster than they would in a large chamber.

So, for a grossly oversized chamber the pressure rises until the spud starts to move. Since the pressure is rising relatively slowly the pressure during the entire time the spud is moving is limited to that was present when the spud started to move, ~10 PSIG.

For a smaller chamber the pressure is rising faster and during at least some of the time the spud is moving it is possible to get the pressure behind the spud to be greater then 10 PSIG.

Looking at Latke's CB data we can get a pretty good estimate of just how critical the CB ratio actually is. The peak in the muzzle velocity around a CB of 0.8 is pretty broad. Statistically, there is really no difference between CB ratios in the range of 0.5 and 1.0 (I'm looking at his graphs at http://www.burntlatke.com/jpg600/15cb-graph.gif and http://www.burntlatke.com/jpg600/25cb-graph.gif).

But, take a look at the 15cb-graph for a spud at a CB of 1.4 (chamber too large). The muzzle velocity has dropped from ~350 FPS to ~230 FPS. The energy in the spud has dropped by about 57% even though the total potential energy in the chamber hasn't changed. This chamber is too large for the barrel... or is the barrel too short for the chamber?

So we are at the trickiest question. Take that last test gun of Latke, the 1.4 CB giving a ~230 FPS muzzle velocity. Chop the chamber down to a ~0.8 CB and use the exact same barrel. Will the muzzle velocity of the gun go up or down with the new chamber size?

Nobody really knows.

What we can predict though is that if the chamber is grossly too large the muzzle velocity will drop. But what about more reasonable CB values of say 1.5?

[boilingleadbath]

Eh, thanks for reminding me that the 3-spark strip was side lit.
It complicates the comparison a lot, though, doesn't it?
(and jimmy, regardless... the difference is <b>small</b> either way. <i>That</i> was my point... which is fuddled by the side-lighting of the spark strip, so let's ignore it for now.)

I like to distinguish between an "overly large chamber" and an "overly small barrel"; in the former, I assume that it's the chamber size that's changed, and that in the latter, we are changing the barrel size.
In my previous post, I was speaking of an overly large chamber.

Anyway, sorry to stray off topic, but I figure I'd build a combustion this summer.
I plan on using a rather... different velocity sensor in the barrel; I'd have an electrical contact every 3" or so, and the projectile would contact them, completing the circuit. The time of this would be recorded with a soundcard at, at least, 96000 samples/sec.

The obvious advantages to me are that:
1) I could gather data for a C:B ratio test in <i>minutes</i> instead of a day.
2) I could gather much more informative data on the effects of variables of construction and use, without having to preform a "C:B ratio test like" experiment.
3) I don't have to buy a chronograph.

I'd like some commentary though:
1) I'm planning on using wisker-type brushes (and tea-candle like 'armatures') - and am worried that the charged nature of the plasma might cause large voltage differentials across the brushes between contacts.
Do you think this is a potential problem, and do you think that connecting the rails (the contacts would be in parallel) with a resistor would be a satisfactory fix? (and if so, how small of a resistance could I use and still get a signal?)
2) I think it'd be cool to use this to simultaneously record the pressure in the chamber using a peizo element. I'm thinking of wiring it in series with the 'rails'. This should eliminate the effect of the DC uncoupling and give us a handy way to know where the projectile is at the time of the reading, no?

Now, I realize that this would require me to use custom ammunition instead of taters. A mixed blessing (accuracy to real life vs. weight consistency).
I think I'll do some tests on that too; a felt wipe on the rear end with some water would answer my ponderings about the effect of projectile moisture on performance, which is one of the differences between the 1.5" latke test and the others. (and, if you've looked at the EVBEC comparisons, the 1.5" data is quite different)

[jimmy101]

Boiling:

Sounds like a great idea. Of course, it is a method to get "Latke type C:B data" quickly and does not directly address the issue of "is the chamber too large or the barrel too small". I believe that can only be answered with a constant barrel and a variable chamber.

Your proposed approach would make it easy to answer a couple of important questions. Like what affect does projectile mass and friction have on the optimal CB ratio?

My thoughts on your approach.

1. Sampling at 96 KHz is really not necisary. The standard 44.1 KHz that a generic PC sound card will do is more than adequate. Of course, if you already have a data logger that'll do 96 KHz then great, but it really won't increase the accuracy of the results. (Hmm, I suspect you actually meant 9.6 KHz which would be fine also.)

2. I don't think the "plasma" is going to give a signal accross your wisker switches. The quality of the flame plasma in a spud gun sucks. If you stick the probes of a volt-ohm meter into the flame of a propane torch you can't measure any conduction in the flame. Conductivity testers are used to measure the flame front position in laboratory settings but it takes a pretty good instrument amplifier to get a measurable signal. (I looked into conductivity testers some months back to see if I could use them to measure the flame front position in the chamber during firing. Looks like it would require a pretty expensive instrument amplifier to get a measurable signal.)

3. I don't think you want to wire the piezo pressure sensor and the projectile position sensors together because the signal is going to be a mess. If you have a stereo input to your sound card (most PCs do, most laptops do not) then you could use one channel for the piezo and one for the position sensors.

Do you have a drawing of your proposed switch setup? The problem with many kinds of switches is that they are slow, bounce and do not pass low currents well if there is insuficient pressure holding them closed.

Trying to think of a way to keep everything simple...

It would be best if the switches didn't require any modifications to the projectile.

I wonder if you couldn't just imbed photodiodes or photoresistors in the barrel and use that to detect the passing of the projectile? Perhaps drill conter sinks into the barrel that fit a particular photodiode/phototransistor or photoresistor. The counter sinks would not go all the way through the barrel wall. They would be deep enough though that the light level in the barrel is measurable. As the projectile passes the light reaching the detectors is reduced. If you take a 2" piece of PVC out into the sun and look into it it is fairly bright inside just from the light that passes through the pipe walls. That might be enough light to detect. (You could of course build in light sources opposite the detectors but then the holes would have to go all the way through the barrel wall, you have to get the lights lined up correctly, there are more wires to deal with ...)

However you decide to do it it would be a great experiment.

[boilingleadbath]

1) I think my laptop can do 96 kh.
I'm not sure exactly how to test that, mind you, but from looking at what should be sharp-edged signals, I think it can.
Worst case scenario, it can't, and I have to use a lower sampling rate.

2) Well, that's good.

3) Well, we own a desktop, but my brother isn't going to let me take it anywhere - so if I want synchronized recordings, I'll have to take them on the same channel. Perhaps a potentiometer in series with my switches, moved in a fashion similar to the dial on an altimeter (flexible metal disk deflects)?

<a href="http://img.photobucket.com/albums/v611/ ... rawing?</a>
(I'm thinking the direction of travel is "up")

You may be right about using light sensors instead of mechanical switches.
I'm wary of the diffusive effect of the PVC, though. Perhaps one could flush-mount them to the inside of the tube, and secure them with some clear epoxy? (I'd have to figure out a way to keep the epoxy out of the barrel, though... or sand it off afterwards)