Registered users: Bing [Bot], D_Hall, Yahoo [Bot]
So, after measuring the approximate rate of propane diffusion in air, I figured the long, skinny and clear combustion chamber would make for some interesting videos.
Unfortunately, I don't have access to a high speed video camera so I had to make do with the video mode of a 6MP digital still camera.
The first video is well equilibrated stoichiometric propane in air. A QuickTime video is here (465KB). You can grab the QuickTime progess slider and single step through the video frames.
Below are the six frames of video that span the entire combustion event. Note that the distances and flame speeds are very rough approximations. The times are calculated based on a 30 frames per second video rate.
In the first frame nothing has happened yet, this is the zero time point.
In the next frame ignition has occurred and the spark-end plastic wrap plug has been blown off.
There is a faint orange smudge to the left of the tube's end which may be the plastic wrap blowing off or, given the orange color, a flame. The denser blue color is the tape that covered the vent hole, the fainter blue smudge is the flame. The flame appears rather diffuse. This is probably caused by the relatively slow shutter speed of the camera. It may also be caused by a turbulent, instead of laminar, flame front.
The next three frames show the flame propagating the length of the tube, in the final frame combustion has finished. The blue color of the flame suggests that the fuel was indeed well mixed.
The fifth frame has some very interesting features:
The white streaks are simply room light reflecting off the outside of the tube. The broad blue smudge is the flame "front". The somewhat sharper and bluer smudge near the end of the tube is the blue tape that covered the injection hole. The interesting features of this frame are the two orange and pink "plumes". The smaller plume near the end of the pipe appears to be coming from the injection hole (the tape on the outside of the pipe was not dislodged during firing). The larger plume near the center of the flame is caused by ...? Close inspection of the tube reveals a small scratch on the inside of the tube which appears to be the source of the plume. So it appears that any flaw in the tube, even something as minor as a covered 0.05"D hole or small scratch causes an orange plume. An orange flame color for a propane in air combustion suggests incomplete combustion caused by insufficient oxygen (an "oxidizing flame"). I wonder if the rough edges and/or turbulence around the surface flaws is causing the polycarbonate to burn. The color is due to an excess of fuel. Inspection of the flaw and hole show no visible signs of combustion. If the polycarbonate is burning then very little plastic is actually being consumed in the process.
There is another interesting characteristic of these two plumes. Which direction do the plumes suggest the gases in the tube are moving at this particular instant in the firing cycle? The plumes suggest that, though the flame front is propagating to the right, the gases in the tube are moving to the left. The length of the larger plume suggests that the gas speed is similar to the flame speed (~30ft/sec) but in the opposite direction. Apparently, the combustion gases to the left of the plume have cooled significantly, dropping the pressure below atmospheric, resulting in a large "suck-back" of air into the tube. This is reminiscent of the flame and gas movement in a pulse-jet engine.
I wonder if I reached into the chamber with a nail on a stick and made small scratches every foot or so along the length of the tube if that would allow me to determine the direction and speed of the gas's movement at different places in the chamber as it is fired? Each small scratch would be the source of a plume, the plume direction and length would indicate the gas speed and direction at a particular instant.
Incompletely Mixed of Fuel
For this video I attempted to get the chamber to fire without fully equilibrating the fuel. The fuel was injected slowly and allowed to diffuse for ~30 minutes, roughly two diffusional half-lives for this chamber. You can view the QuickTime video here(1.3MB).
The second and third frames have orange-ish flames that do not appear to be associated with imperfections in the tube. Presumably this coloration is due to incomplete mixing of the fuel.
As in the first video, some of the frames have plumes that appear to be associated with flaws in the tube. The fourth frame has a single small plume, the fifth has a couple plumes.
A Very Rich Mixture
For the next video I injected sufficient fuel to bring the chamber to about 7% propane. The mixture is very rich and near the upper combustion limit or propane in air. You can view this video here (3.4MB).
Note that in the image above only the first frames are shown, there are an additional ~90 frames showing the flame front progressing through the tube.
The rich mixture gave a robust blow-out flame at the spark end which lasts ~130mSec. The flame front then proceeded down the tube very slowly, taking about 3 seconds to reach the far end. At the far end it burned through the plastic wrap plug. This very rich mixture gave a flame speed in the tube, after the initial blowout, of only about 1 foot/second, more than an order of magnitude slower than the shots with stoichiometric fuel. The slow burn speed may be a result of the need to suck fresh air into the tube, from the chambers spark end, to support combustion. Since the injection end cap was never blown off, air can only enter the tube via the spark gap end.
Not sure if I learned anything (other than working with video is a PITA) but movies of stuff blowing up and/or burning are always fun.
He did - They're linked in the post, just above the frame break down for each one.
Why it makes a raspberry noise, I don't want to guess.
Does that thing kinda look like a big cat to you?
"air diffusion into the tube to sustain combustion - that's why it took so long"
I bet not. Laminar flame front speeds in poor fuel-air mixtures are slow.
Just for fun, try capping the ignition end with something solid. I bet you'll see faster prorogation with all fuel mixtures.
Maybe, maybe not. Granted the laminar speed is only ~1ft/sec but that is at 1 ATM and typical temperatures.
What I suspect happened is that;
1. Spark end plug blew out
2. A fairly large proportion of the fuel / air mix in the tube was ejected, casueing the large extrenal flame.
3. The gases collapse back into the tube. If something like 50% of the gases were ejected then the tube now contains ~4% propane but only 10% oxygen.
3. The flame starts to migrate up the tube
4. The rich mixture is due to lower than normal oxygen content in the gases instead of higher than normal propane. Indeed, 4% propane in 10% oxy may not even be within the combustion limits.
5. More air is required to sustain combustion, but the air has to chase the flame front down the tube.
The more I think about it the more I suspect that that particular results is pretty much a fluke. I should probably try it a couple more times.
I think this is the relevant point, the behavior of the rich mix really has more to do with the strange behavior of the chamber and less to do with the affect of a rich mixture. In a closed chamber the rich mix would look a lot more like a stoichiometric mix. Maybe not exactly the same, but a heck of a lot more similar.
Anyone have any thoughts on whether the polycarbonate tube will survive the ~120 PSIG that a closed chamber will reach?
Soooo Jimmy, Why all this experimenting with combustion chambers? What are you in search of?
When life gives you lemons...throw them back they suck!
Wow, nice job... I can tell you spent a lot of time on this and worked hard
thumbs up to you!
Forum Rules - follow them or else!
“Happiness is the meaning and the purpose of life, the whole aim and end of human existence”
Building up to the complete tech gun and still trying to gather data to help with a mathematical model of a combustion gun.
The tech gun is done but hasn't been fired yet. If it works like it is supposed to, I should be able to get a complete Latke like CB study from a couple shots of the gun (instead of the 50~100 shots it took Latke) . That should make it a lot easier to answer questions like what is the optimal CB with versus without a fan, or multiple sparks, or different spark locations, or different projectile masses, or ...
Yes. Why wouldn't it?
It seems to be the material of choice for DDT experimentation /shock loads.
Well, because it isn't pressure rated and the tubes purpose is an application with no pressure.
But, it turns out that the water rocket people occasionally use these tubes for rockets. So, it looks like they'll take ~120 PSIG. The tricky part will be figuring out how to close off one or both ends. For water rockets, they just glue (or epoxy) an end cap on one end and a nozzle on the other.
Nice work, sir!
On a related note... It will probably take me another year to pull it off (unfunded projects take a lot of time), but right now I'm gearing up for 29" diam x 20' long combustion tests. Welders already have the drawings of the test chamber. I doubt I'll be able to match the visible flame for "cool factor" but I should be able to provide some excellent time/pressure/temperature histories.
I'm still jealous...
Any thoughts on how you are going to do temperature? I've considered trying a thermocouple but the heat capacity and response times seem to be totally unuseable for combustion gases in a closed chamber or gun.
Who is online
Registered users: Bing [Bot], D_Hall, Yahoo [Bot]