I decided it would be useful to characterize new propellants. And the best way to do that that is with a strand burner. The basic function of a strand burner is to measure the burn rate of a propellant strand under different pressures. The propellant variable a & n can only be determined through testing. While it could be done with static tests of a motor using different grains and/or nozzles, the strand burner should be easier and yield better data.
With my new digital data acquisition hardware, the strand burner is even easier to construct and use. Many strand burners use thermocouples or heat sensitive burn wires to indicate the start and end time of a propellant strand burn. With the digital data acquisition, I can record the strand burn start and end time by the chamber pressure rise, and then fall.
To start with, I am using my 2" EMT for a chamber. While the EMT is not the best material, it is what I had handy. In the future I will use a heavier gauge steel chamber. I will limit chamber testing to about 600 psi with this chamber. I welded a 12" long section of the EMT to a 3" wide by 8" long piece of 1/4" thick steel plate. This plate forms the back bulkhead of the chamber and also creates the upright of the stand to hold the strand burner. I then welded a steel plate to form the base of the stand, it holds the strand burner in a horizontal position.
The forward bulkhead was built the same way I do my motors. A .8" thick plug was turned on the lathe, and a groove was cut for an o-ring. I then drilled two small holes for the wires needed to fire an igniter to start the burn. The wires were inserted into the holes, then sealed on the chamber side with RTV. I then sealed the outside of the holes with high strength epoxy.
The plug is held in place with 12) 10-32x3/8" screws. The chamber casing was drilled and tapped for the screws. The plug is simply inserted into the chamber, then the screws are inserted into the casing and the plug allowed to press against the screws.
Above is a picture of the strand burner, the forward closure plug is on the bottom, the 26 volt regulated power supply is to the right.
Here is the business end of the strand burner with the forward closure in place.
The plumbing end of things is pretty straight forward. I drilled and tapped the back plate to 1/4" NPT. Then a short nipple leads to a T with a standard pressure gauge, another nipple leads to another T with the pressure transducer, the last nipple connects to a valve and a quick coupler for chamber pressurization.
Here is the regulated power supply. It uses 4) 9 volt batteries to output 26 volts. The regulated power is needed to keep the voltage stable to the pressure transducer. I plan on using the power supply on other transducers, so I used a RCA style connector to quickly change out the power supply to other units. I also installed a simple on/off toggle switch.
Before getting ahead of myself, I should mention a few issues I have had with the strand burner.
First of all. When I first designed the end plug, I tried using small bolts as penetrators to the inside of the chamber. I put a couple of turns of electrical tape around the bolt, just over the area that would be inside the plug. I then used small o-rings, washers and then double nuts to seal the bolt ends. I have used this technique in the past (on my home built submarine), and it worked well. But, I was concerned about the high pressures involved here, and the bolt threads would be hard to get a good seal against. My concern was justified, as both of the small bolts leaked. That's when I decided to try the epoxy. The problem I had with the epoxy was, it was so thick it didn't want to flow into the hole around the wire. So I tried poking it in with a needle. As the epoxy cured, I noticed small bubbles coming to the surface. I tried to poke them out with the needle, but as the epoxy started to cure, that was no longer possible. As a result, one of the wire holes has a small air leak when pressurized. I think I will try drilling the epoxy out, then using a syringe to inject the epoxy into the bottom of the hole, displacing the air out the top as it fills. I think that should take care of the problem. It should also be noted, only one hole is really needed. You could use the metal itself as one of the conductors.
Another problem I had, and anticipated. The EMT does not have a perfect surface for an o-ring to seal against. There is a weld seam that is the real problem. I tried to smooth the seam with a roto tool, then using fine grit sandpaper to polish the area. On my first pressure test, the o-ring around the weld seam leaked, not very much, but more than I was comfortable with. So I went back in and smoothed the surface even more. It seems to be leak free now. In my motors with a similar design, I use a small amount of RTV in addition to the o-ring. I really don't want to use RTV in the strand burner as it would really slow down the test rate.
Another concern I have. I'm not entirely convinced my weld holding the chamber casing to the back plate will hold. At one time I considered myself a good welder. But now, all I have for a welder is a small 120 volt unit, and I can't seem to get my welds perfect. The first pressure test I did showed about 7 small pinhole leaks at the weld. That's not very good for a 6 inch weld. So I went over them again, it looks messy but I think it will help some. To make sure the weld was sealed, I mixed my high strength epoxy and poured about 1/3" into the back of the chamber. The epoxy isn't there for strength, just to seal any pin holes. If I run a maximum chamber pressure of 700 psi, I will have a total force on the bulkhead of about 2,200 pounds. The weld should hold that, but just in case. I make sure the ends are facing away from me. I would like to hydro test the chamber, but that may have to wait.
As for data acquisition. I pondered what type of wire to use to carry the signal. I know a good shielded wire would be best. But I didn't have any. What I did have was 1,000' of RG 6 cable (TV wire). So I cut off 100' and gave it a try. I really worked very well, I didn't have any noise in the signal and the calibration remained rock solid for the hour or more I had it running. One more issue. The DataQ A/D board has a nasty tendency to want more power than a laptop computer can deliver. Of course, the laptop was what I was planning on using. DataQ has a tech page on converting the unit to remote power, which should cure the problem. But their information is sketchy at best, and required a couple of surface mount (very, very small) components to be installed. I found another company that had a similar, but simpler approach to the DataQ's external power. I tried the modification on my board, and it didn't work. So I switched it back to original and it again worked. So I'm still searching for a way to use my laptop with the DataQ. Any thoughts would be appreciated, I'll post the answer here once it's solved.
Update: Jan. 30, 2004
Last night I drilled out the leaking wire penetrator and re-epoxied the wire into the hole. A quick pressure test this morning indicated it was no longer leaking. So I decided it was time to give the strand burner its first test. I pressured the chamber up to 210 psi with air, and all was well.
So I went about getting everything ready for a test. The only strand I had was standard KNO3/sugar, and it was a little larger diameter than it really should be. But, I was going to fire the burner without adding any outside pressure, just to see how much the propellant would bring the pressure up on its own.
I cut the strand to a length of 1.9", it weighed 13 grams. The strand was then outside inhibited with a thin layer of RTV and allowed to cure. I then hot glued the strand to the forward plug. Earlier this evening I made up a batch of new igniters, a dozen of them just a few inches long for the strand burner. I used a newly made igniter and just twisted the igniter wires to the penetration wires on the strand burner. The wires were wrapped with a couple of turns of electrical tape to help protect them.
I set up my video camera zoomed in on the analog pressure gauge. I Hooked up the signal wire to my computer and set up my launch controller to fire the igniter. The DataQ software was then calibrated to my pressure transducer. Continuity check on the igniter was good, video camera was recording, DataQ software was recording, I was ready to go.
5,4,3,2,1 ignition! I expected something, you know, a sound at least. Nothing... but, a quick look at my computer screen showed the pressure rapidly rising to over 400 psi, then slowly dropping off. Hey, it worked.
I went in and stopped the video camera, then bled off the pressure from the strand burner, very little smoke came out, that surprised me. A quick check of the recorded video indicated about 410 psi. After making three copies of my data file (I didn't want to lose the data), I looked over the information the transducer had reported.
Here are the numbers from the test.
Strand Length: 1.9"
Strand Weight: 13 grams
Maximum Pressure: 402 psi
Burn Time: 3.4 Seconds
Average Burn Rate: .558"/Second
Average Pressure 201 psi
The burn rate surprised me a little. It was faster than expected. The grain was very dry, it was some months old and had been in my desiccant box the entire time. The grain was also made from the ball mill process, which does lead to faster burn rates. All in all I was pleased with the way it worked. Now I need to get a nitrogen tank for higher pressure tests, and I need to make up some strands for testing.