-
#1220 by Silversheep2011 on 27 Aug, 2015 12:07
-
Rmmfguy
Thanks for making those video available for public viewing
And on behalf of many here, we appreciate the time and sacrifice of effort that when into building the test rig and the ongoing testing.
Something that I noted as an observation between the two tests during power ’on’ was a change in the 'audible humming tone"
NSF-1701 Flight Test #1 try listening between 4:50 to 5:00
NSF-1701 Flight Test #2 try listening between 4:41 to 4:51
On test 1 I can hear something sounding like a higher and harder metallic note.
On test 2 I can hear a slightly more lower quieter and softer and more even note.
Rough Analogy: It makes me think of arc welding steel as much as it’s a visual skill you learn It’s also a skill you learn naturally by using your ears.
i.e. A high arc gap sounds bit like first test .[beginner's or start of weld] Second one sounds bit more like when arc gap and current is getting dial in and laying down weld.
It could mean that the big end plate was vibrating as a drum in first test run, and on the second run with magnetron’s end change it's now served a dual purpose of stiffing up the big end plate up. And stopping or damped down some low frequency audible resonance. Or some else of interest.
Suggestion: why not go the garage when all is quiet and people have gone to bed and use you ears [no need to video] as see if you can track down the placement main source of the "hum" It might just prove to be a helpful observation….
You know there is some resemblance to a “big speaker” after all…
-
#1221 by chucknorris101 on 27 Aug, 2015 14:04
-
Early days for drawing any conclusions. Of course, you try the big and obvious things first. Repositioning the magnetron was one such trial. I'd suggest another big and obvious one would be to flip the frustum over so now it's small side up. Does the thrust reverse? That is what Iulian found.
I think rfmwguy's setup is actually perfect for the flip test as well - the flanges that may have caused trapping of hot air in this direction are symmetrical on the other side - so any buoyant effects should be the same (I imagine it might be different for just a solid frustum shape, one end up or the other, but im not a physicist) -
#1222 by rfmwguy on 27 Aug, 2015 14:15
-
Once again, thanks for all for analysis and commentary. I find it inspirational. I went to bed early last night after a couple of days of intense shop work and testing. Here are some thoughts this AM:
1) There is some unexpected movement that is worth pursuing at a higher measurement resolution; well below 200 mg.
2) FT1 and FT2 produces what I believe to be thermal lift as the magnetron goes through its 6 minute total testing cycle, equivalent to approximately 500 mg.
3) The thermal lift is more linear in FT2. One possibility is the (heat source) magnetron is on the bottom of the assembly in FT1 and heated air builds, then surges upwards around the bottom plate. On top in FT2, it simply rises vertically, in a more controlled, semi-linear fashion.
4) Galinstan has an unexpectedly high amount of surface tension and high viscosity. Hat tip to Mr. Peter for pointing this out. On the positive side, this helps dampen oscillations. On the negative side, it attenuates small movements.
5) The panoramic mirror adds unecessary distortion in an attempt to create greater range of deflection.
6) The laser pointer mounted on the beam is not of a quality high enough for accurate readings in the milligram force range.
7) My mechanical test stand is suitable for further measurements with the following improvements:
a) High quality laser, tripod mounted off the beam, directed at a single surface, flat mirror mounted where the laser pointer is now on the balance beam.
b) Another single surface, flat mirror mounted on tripod where panoramic mirror is now located.
c) design a new laser target (have to think about this some more)
d) Replace galinstan with salt water per Mr. Peter's suggestion to lessen galistan drag.
e) HD video camera
Final notes: While I do not have the budget and time to spend creating a set of lab-quality experiments, I do believe DIY projects have something to offer to the body of knowledge, if not help stimulate interest in science. Private and public institutions exude secrecy, which can lead to wild speculation and unsubstantiated rumors. DIY projects can help reverse that trend and lead to more open-sourced projects we desperately need in many disciplines; the internet is a gift we should put to good use.
Pic Caption: FT2 setup this AM. Looking down with magnetron mounted on big base, small base is down. Electrodes dipping down into copper cups will Galinstan. Note location of black laser pointer.
-
#1223 by Rodal on 27 Aug, 2015 14:20
-
Early days for drawing any conclusions. Of course, you try the big and obvious things first. Repositioning the magnetron was one such trial. I'd suggest another big and obvious one would be to flip the frustum over so now it's small side up. Does the thrust reverse? That is what Iulian found.
I agree that the next step should be to flip the truncated cone so that the opposite end is pointing up.
Concerning the power settings: it was a good idea for RFMWGUY to plan to test the power at different settings before we knew what the test results were going to be. Why? Because the claimed results from Yang and Shawyer are by several orders of magnitude higher than those reported by Tajmar and by NASA. But now that we know that the results are at the very threshold of the measuring ability of this set-up, wouldn't it make sense to only test at 100% (maximum) power setting?
What is the point of testing at lower power setting when even the 100% maximum power setting gives results that are at the threshold of the measurement capability?
I would only test at 100% maximum power setting from now on. I would conduct the same or more number of power on/ power off time settings. This would allow us to look at some statistical distribution, by only testing at 100% maximum power and having several tests power on/ power off/power on/power off. Testing at only 100% power setting, and conducting a number of such tests would allow to further differentiate between the slow drift that appears unrelated to power, and any other movements. -
#1224 by Rodal on 27 Aug, 2015 14:26
-
...1) There is some unexpected movement that is worth pursuing at a higher measurement resolution; well below 200 mg....
I agree. Since the movement is at the threshold or below the threshold of what can be measured at the moment, I would only test at 100% power from now on....2) FT1 and FT2 produces what I believe to be thermal lift as the magnetron goes through its 6 minute total testing cycle, equivalent to approximately 500 mg....
I completely agree. The slow drift looks like a slow thermal artifact and not related to the electromagnetic power on/power off. We need to distinguish between the slow drift that appears unrelated to the electromagnetic power and other movements that may not be so explained away.
////////////////////////////////////
I hope that you will continue with your experiments
Congratulations again for a very impressive experiment !
-
#1225 by Stormbringer on 27 Aug, 2015 14:31
-
Are those carpenter's laser leveling systems from Home Depot or Lowes sharp enough for this sort of thing? I would assume they would be and they aren't that expensive and later you can use it for what it was made for. for example: http://www.lowes.com/ProductDisplay?partNumber=587559-353-8203-CL&langId=-1&storeId=10151&productId=50285087&catalogId=10051&cmRelshp=req&rel=nofollow&cId=PDIO1
and you can get models that project cross hairs/ cross lines instead of a dot with the ability to switch between dots and crossed perpendicular lines. -
#1226 by aero on 27 Aug, 2015 15:01
-
Early days for drawing any conclusions. Of course, you try the big and obvious things first. Repositioning the magnetron was one such trial. I'd suggest another big and obvious one would be to flip the frustum over so now it's small side up. Does the thrust reverse? That is what Iulian found.
I agree that the next step should be to flip the truncated cone so that the opposite end is pointing up.
Concerning the power settings: it was a good idea for RFMWGUY to plan to test the power at different settings before we knew what the test results were going to be. Why? Because the claimed results from Yang and Shawyer are by several orders of magnitude higher than those reported by Tajmar and by NASA. But now that we know that the results are at the very threshold of the measuring ability of this set-up, wouldn't it make sense to only test at 100% (maximum) power setting?
What is the point of testing at lower power setting when even the 100% maximum power setting gives results that are at the threshold of the measurement capability?
I would only test at 100% maximum power setting from now on. I would conduct the same or more number of power on/ power off time settings. This would allow us to look at some statistical distribution, by only testing at 100% maximum power and having several tests power on/ power off/power on/power off. Testing at only 100% power setting, and conducting a number of such tests would allow to further differentiate between the slow drift that appears unrelated to power, and any other movements.
@rfmwguy - correct me if I am wrong, but I thought the magnetron had only 2 power settings, "off" and 100%. It is the microwave oven control that runs the magnetron at 100% power, but only 30% of the time achieving what is called 30% power level. And it is 30% for cooking, but for the thruster, it is 100% for 30% of the time.
And on a side note, why 30%, why not 50%? Have you been able to calculate the resonate frequency of your foobie stick set-up? It might be interesting if it is possible to drive your balance beam at resonance. -
#1227 by Croppa on 27 Aug, 2015 15:07
-
...1) There is some unexpected movement that is worth pursuing at a higher measurement resolution; well below 200 mg....
I agree. Since the movement is at the threshold or below the threshold of what can be measured at the moment, I would only test at 100% power from now on....2) FT1 and FT2 produces what I believe to be thermal lift as the magnetron goes through its 6 minute total testing cycle, equivalent to approximately 500 mg....
I completely agree. The slow drift looks like a slow thermal artifact and not related to the electromagnetic power on/power off. We need to distinguish between the slow drift that appears unrelated to the electromagnetic power and other movements that may not be so explained away.
////////////////////////////////////
I hope that you will continue with your experiments
Congratulations again for a very impressive experiment !
It seems to me that the advantage of running at lower power is that it cycles on/off without rfmwguy being close to the setup. As I understand it the on cycles are full power anyway and only the overall proportion of on time is reduced. There are definite oscillations associated with the start of these runs (you can see these more easily in the unstretched montage shown here, as well as the raw video). I suspect these are air currents associated with rfmwguy moving past the apparatus after triggering the run but can't say for sure.
As other have noted, a great improvement to the current setup would be the ability to trigger it from a distance, although I imagine that's not trivial.
-
#1228 by TheTraveller on 27 Aug, 2015 15:46
-
Dave,
Instead of guessing in the dark at what load your magnetron is driving and whether your frustum has resonance or not at the magnetron's freq output bandwidth, you need to make a magnetron antenna probe that matches your magnetron's antenna, as attached, put it where your magnetron's antenna is installed and do a VNA S11 return loss dB sweep. From that you will get the resonant frequencies and the VSWR / impedance match.
It bothers me that your frustum is not getting warm as the magnetron's output energy has to be heating something. Worst case is it is all getting rejected as there is no resonance, lousy VSWR and the high level of reflected Rf is heating your magnetron.
Doesn't matter that you think this is not the correct why to do this. You need to do a survey of your frustum's resonant frequencies. Eagleworks did it. You need to characterise your frustum so you know what you are dealing with and what sort of load your magnetron is driving.
-
#1229 by Rodal on 27 Aug, 2015 15:47
-
...1) There is some unexpected movement that is worth pursuing at a higher measurement resolution; well below 200 mg....
I agree. Since the movement is at the threshold or below the threshold of what can be measured at the moment, I would only test at 100% power from now on....2) FT1 and FT2 produces what I believe to be thermal lift as the magnetron goes through its 6 minute total testing cycle, equivalent to approximately 500 mg....
I completely agree. The slow drift looks like a slow thermal artifact and not related to the electromagnetic power on/power off. We need to distinguish between the slow drift that appears unrelated to the electromagnetic power and other movements that may not be so explained away.
////////////////////////////////////
I hope that you will continue with your experiments
Congratulations again for a very impressive experiment !
It seems to me that the advantage of running at lower power is that it cycles on/off without rfmwguy being close to the setup. As I understand it the on cycles are full power anyway and only the overall proportion of on time is reduced. There are definite oscillations associated with the start of these runs (you can see these more easily in the unstretched montage shown here, as well as the raw video). I suspect these are air currents associated with rfmwguy moving past the apparatus after triggering the run but can't say for sure.
As other have noted, a great improvement to the current setup would be the ability to trigger it from a distance, although I imagine that's not trivial.
Cycling on/off at less than 100% full power constantly on all the cycle time has this undeniable drawback:
the response at the moment is so small that it is at the threshold or below the threshold of what can be measured. Therefore one should only test with 100% power constantly on all the cycle time in order to try to maximize the response. Anything less than 100% power on during the ON cycle will just result in amplifying the relative level of "noise" (thermal and other artifacts) and minimizing the relative level of any possible signal that is not an artifact.
-
#1230 by kencolangelo on 27 Aug, 2015 16:26
-
As other have noted, a great improvement to the current setup would be the ability to trigger it from a distance, although I imagine that's not trivial.
It should be really easy to trigger remotely. Ultimately the microwave control panel has a Start button, a membrane switch, you can just solder onto the circuit board and run two long wires wherever you like attached to a remote start momentary contact button. Preferably something dramatic. http://www.amazon.com/Lock-Switch-Mushroom-Button-Station/dp/B008ZY9CXE.
This person found a microwave and replaced the pad entirely with a hack of their own.
http://hackaday.com/2011/06/09/repairing-a-broken-microwave-keypad/
-
#1231 by Abyss on 27 Aug, 2015 16:39
-
You need to do a survey of your frustum's resonant frequencies. Eagleworks did it. You need to characterise your frustum so you know what you are dealing with and what sort of load your magnetron is driving.
Agreed. As more experiments are run it is clear that the 'magic' parameters for large measurable forces must be in a very small range. (as an aside, to me that makes it questionable that Shawyer found them in the first place).
As the search begins, it will be more and more important to characterize each step of the way to ensure you are working with what you think you are working with. -
#1232 by TheTraveller on 27 Aug, 2015 17:09
-
You need to do a survey of your frustum's resonant frequencies. Eagleworks did it. You need to characterise your frustum so you know what you are dealing with and what sort of load your magnetron is driving.
Agreed. As more experiments are run it is clear that the 'magic' parameters for large measurable forces must be in a very small range. (as an aside, to me that makes it questionable that Shawyer found them in the first place).
As the search begins, it will be more and more important to characterize each step of the way to ensure you are working with what you think you are working with.
Agreed.
Shawyer's 1st EMDrive, that he reported on, was highly tunable as attached.
So he clearly KNEW of the variables and designed in methods to deal with them, which says to me this was not his 1st experimental build but his 1st successful experimental build.
Shawyer did have at least a decade or 2 of microwave experience before he built his 1st device, so he had a decent real world microwave background knowledge to work from.
His Experimental unit could tune the top plate, the feed waveguide impedance, and the amount of the bottom dielectric insertion into the frustum for maximum Force generation. Says to me he knew well the devil he was working with. And this was in 2000 - 2002.
-
#1233 by seggybop on 27 Aug, 2015 17:19
-
Pic Caption: FT2 setup this AM. Looking down with magnetron mounted on big base, small base is down. Electrodes dipping down into copper cups will Galinstan. Note location of black laser pointer.
Is that blue masking tape holding the laser pointer in place? If so, that will be really prone to slowly drooping.
I think it may be a good idea to affix it with something more rigid, like hose clamps. -
#1234 by tchernik on 27 Aug, 2015 17:40
-
Agreed.
Shawyer's 1st EMDrive, that he reported on, was highly tunable as attached.
So he clearly KNEW of the variables and designed in methods to deal with them, which says to me this was not his 1st experimental build but his 1st successful experimental build.
Shawyer did have at least a decade or 2 of microwave experience before he built his 1st device, so he had a decent real world microwave background knowledge to work from.
His Experimental unit could tune the top plate, the feed waveguide impedance, and the amount of the bottom dielectric insertion into the frustum for maximum Force generation. Says to me he knew well the devil he was working with. And this was in 2000 - 2002.
I agree. Shawyer's experience surely showed him that he needed to be able to tune the most parameters he could.
And he came with a solution that could tune the geometry (vital for achieving resonance or the "right stuff") and possibly the magnetron's frequency. In this way he just needed to build one test device for covering a big swath of combinations, instead of building one fixed model, seeing it didn't work well or at all, and then build another and another. -
#1235 by zen-in on 27 Aug, 2015 17:55
-
...
If that is the case, we'll have to look at this as more than an instantaneous burst of motion.
As a side note, when I came back near the experiment at the end of the test run, I felt no warmer air than when I started, but did notice an unusual "ambience'" to the air. Best way I can describe it as metallic, ionized air. This is not scientific obviously, just a minor observation I did not notice during static testing.
That metallic ionized air may be ozone. If it collects under the flat metal pieces, due to static effects, it may have a bouyant force. Because of the static attraction to the metal the ionized molecules will accumulate under the plate and that population will have a lower pressure because of similar charge repulsion. The ionized air does not have to be physically contained to cause an upward force. Maybe by interposing a grounded screen just under the metal plate this bouyant force will be removed. -
#1236 by RERT on 27 Aug, 2015 17:56
-
rfmwguy -
I just spent some time trying to reconcile the results you have with some kind of expectation of how the setup should respond to a thrust.
The apparent slow movement of the balance beam seems simplest to fit to thermal (probably air current related) forces building as the temperature rises. That's also consistent with the apparent direction of the force being upwards in both cases. There could still be thrust, but thermal effects will clutter the results. What's more, I think that's intrinsic to the balance-beam approach, where the measured thrust is necessarily vertical. It's also hard to net them out in your setup, because the position of the magnetron means that thermal forces will be asymmetrical as you invert the frustrum.
If the effect is real, there is no reason to expect it to build over minutes. When you placed the weights on the beam, the response was comparatively immediate. Therefore, a 'kick' on switch-on followed by a slow build due to thermal might be something to look for. I'll go back and peer at that for a while!
My son the engineer commented that he would prefer to be looking for a horizontal force, otherwise thermals will be a confounding factor. Maybe the rotating table of Shawyer's wasn't such a bad idea....
R.
-
#1237 by TheTraveller on 27 Aug, 2015 17:58
-
Agreed.
Shawyer's 1st EMDrive, that he reported on, was highly tunable as attached.
So he clearly KNEW of the variables and designed in methods to deal with them, which says to me this was not his 1st experimental build but his 1st successful experimental build.
Shawyer did have at least a decade or 2 of microwave experience before he built his 1st device, so he had a decent real world microwave background knowledge to work from.
His Experimental unit could tune the top plate, the feed waveguide impedance, and the amount of the bottom dielectric insertion into the frustum for maximum Force generation. Says to me he knew well the devil he was working with. And this was in 2000 - 2002.
I agree. Shawyer's experience surely showed him that he needed to be able to tune the most parameters he could.
And he came with a solution that could tune the geometry (vital for achieving resonance or the "right stuff") and possibly the magnetron's frequency. In this way he just needed to build one test device for covering a big swath of combinations, instead of building one fixed model, seeing it didn't work well or at all, and then build another and another.
During the process Shawyer burnt out 3 magnetrons of 5 he used (probably due to reflected Rf) and partly melted a section of his waveguide assembly. So he learned at the school of "Microwave Hard Knocks 101". -
#1238 by aero on 27 Aug, 2015 17:58
-
Thinking more about driving the balance beam at resonance. It should be possible if the natural frequency of the balance beam is reasonable. What is the natural frequency? When you lift one end a few inches and release it, how much time does it take to complete one cycle returning to near the starting point? And what is the quality factor of the balance beam? How many cycles does it take before the maximum amplitude decays by 3 db?
Knowing the natural frequency, what is the power setting that gives a pulse at the same point in the frustum cycle? Run that power setting to see if it will induce resonance and a detectable sweeping of the laser pointer?
If it is possible to attribute resonance of the measurement system to force from the thruster, then it is just a mater of mathematics and classical mechanics to determine the force. Heating shouldn't be a significant issue, or maybe it will be. -
#1239 by TheTraveller on 27 Aug, 2015 18:01
-
rfmwguy -
I just spent some time trying to reconcile the results you have with some kind of expectation of how the setup should respond to a thrust.
The apparent slow movement of the balance beam seems simplest to fit to thermal (probably air current related) forces building as the temperature rises. That's also consistent with the apparent direction of the force being upwards in both cases. There could still be thrust, but thermal effects will clutter the results. What's more, I think that's intrinsic to the balance-beam approach, where the measured thrust is necessarily vertical. It's also hard to net them out in your setup, because the position of the magnetron means that thermal forces will be asymmetrical as you invert the frustrum.
If the effect is real, there is no reason to expect it to build over minutes. When you placed the weights on the beam, the response was comparatively immediate. Therefore, a 'kick' on switch-on followed by a slow build due to thermal might be something to look for. I'll go back and peer at that for a while!
My son the engineer commented that he would prefer to be looking for a horizontal force, otherwise thermals will be a confounding factor. Maybe the rotating table of Shawyer's wasn't such a bad idea....
R.
Battery powered, totally self contained, no wires, rotary table is the only way to go.
BUT NOT SO EASY TO BUILD AND GET FULL, REAL TIME, DATA COLLECTION OF ALL OPERATIONAL PARAMETERS.
