Larry Cottrill wrote:What I really need to do is set up some simple, reasonably sensitive way to measure thrust
Run the Elektra vertical, on the bathroom scales. That's (rougly) what the German rocket pioneers did, well before WW II. What's wrong with that?
Well, last night I tried this. It turned out to be one of the most frustrating and difficult times I've had so far in getting a pulsejet to run.
I learned one thing that is nice to know: It is possible to start Elektra I with air from my shop vacuum. I was able to get it running by using the nozzle of the vac practically right up against the intake flare, and then just backing it away fairly slowly after good firing started up. However, rightly or wrongly, I convinced myself that it wasn't running up to full power, so I went back to the leaf blower and never saw it self sustain again for the next couple of hours.
Of course, I had grand plans of getting a five minute run with measured thrust and measured fuel consumption. It was close to 90 degF in late afternoon with very high humidity. I set up the bathroom scale in the driveway, locating it carefully so it showed level with a spirit level. I weighed the propane tank and recorded the weight, then set the concrete block on and clamped the engine stand to it with the C-clamp. After attaching the fuel hose, I recorded the total dead weight as 31.0 lb [it's not a whole block, maybe 2/3 of the whole]. I managed to remember to get a photo of the setup at this point.
I figure I can read the bathroom scale, with care, to the nearest quarter lb, though values need to be taken with a grain of salt, since the scale hasn't gone back to the US Bureau of Standards for calibration for a few years now ;-)
I decided right away to give the shop vac a decent try, and after much trial and error got it to work when I held it in close enough. This resulted in a fine smooth run, with no perceptible variation in operating frequency, as previously reported. However, the scale showed only about a pound of increased loading, no matter how much I fiddled with the needle valve setting! Very disappointing, needless to say. Though the run seemed nice and loud, I jumped to the conclusion that either the engine was 'running light' [i.e. not at all 'pumped up' as our friend Rossco might put it] OR that an astonishing fraction of this engine's energy is devoted to driving gas out the intake [at a highly ineffective 60 deg angle to the intended thrust line]. At any rate, it didn't amount to much.
Note that the present configuration is still a 1-inch [25mm] ID tailpipe about a foot long with a 35-inch long 1.25 inch OD thin-walled sleeve slid all the way over it, for a total length of 38.5 inches [1 metre for all practical purposes]. Note also that output from the intake is significant. When you get your first explosions during starting, the impact into your air supply feels to your hand almost like the recoil of a small calibre pistol. This is especially observable with the shop vac nozzle, but can be felt even with the vastly more massive leaf blower in hand. Also, when the engine is running, you can feel quite a push a few inches away from the intake, as well as observable flame up through the center [of course, my fuel arrangement may be feeding this flame, especially at the low fuel pressures I use]. The output is not dangerously hot, however, so there must be a fair amount of fresh air content being kicked back out.
At any rate, I made the decision to use the leaf blower from this point forward. Try as I might, I never did get more than a few seconds of good pulsing combustion at any time, and it never would carry over into sustaining operation. I recognized at one point that the Model T coil was sounding pretty weak, and re-charged the battery, but while I now got easier bangs and roars, I still couldn't hit that 'just right' condition where it would take off and keep going. I had to stop when I got low on propane and vapor pressure, with ice all around the base of the tank.
When I finally gave up on it and disassembled everything, I found to my amazement that the scale was still reading 30 lb, with nothing on it! I tried stepping on it a couple of times and found that it would only swing up slightly and hang, then drop back to 30 when I stepped off. After shaking it a bit, it fell back to 0. When I then stepped on it, it zoomed up to about 50 and stopped. Now friends, I knew I had lost a lot of sweat out there, but surely not two thirds of my former weight! I tried shaking the thing fairly briskly in various positions, and managed to sift out a fair amount of dust and sand-like crushed rock particles of various grades. This restored the scale to normal operation. So, that was it -- my use of the blowers had caused infiltration of dirt into the scale mechanism, effectively blocking it! So, was this already happening by the time I got those first smooth runs? I'll never know.
Obviously, the crushed rock driveway is not the ideal place to try this kind of measurement, with the tools available -- just another of the many technical shortcomings of the Cottrill Cyclodyne Corporation's Propulsion Laboratory. I may try it again with the scale clothed in clear plastic to keep it clean, unless I manage to get some better equipment put together in the meantime.
What was the starting problem, really? I found it oddly awkward to work with the engine set up in a vertical position, possibly because this setup forced me to use the blower left-handed. I wonder, too, if there is a tendency for the engine to flood with vapor in this vertical orientation; I seemed to get the best bangs after shutting off the fuel for a while, allowing the blower to purge the machine. Or, could the extreme humidity be contributing to the problem? Or, all of the above? Hard to say, especially since I did manage to get a good start [though possibly not a full-power run] with the shop vac with relatively little experimental effort.