EmDrive Builder will Answer Questions

[rfmwguy]

Seems an older thread was locked down after I offered to answer legitimate questions on EmDrives as I understand them. Not sure if this is a verboten topic here or wise-guys abuse the topic, but for those honestly interested in what I know about it, my design, build and testing of 2 engineering models, fire your questions here...unless the powers to be want the topic to be excluded.

Edited November 5, 2016 by rfmwguy
typo

[Scotius]

OK, i will start - did you do any calculations of power input\thrust ratio? How much thrust can be squeezed out if we'd stick EmDrive chamber inside a cubesat?

[rfmwguy]

My friend Paul Kpycla is building a picosat EmDrive here: https://hackaday.io/project/5596-em-drive

Mine exceeds the 10 cm form factor and would not be suitable for cubesats. Basically, I had 750W RF producing 18.4 mN.

[sal_vager]

Hi @rfmwguy.

We'd just like to avoid the issues that arose with previous EM drive discussions, by all means please share your experiences with these devices.

[rfmwguy]

Just now, sal_vager said:

Hi @rfmwguy.

We'd just like to avoid the issues that arose with previous EM drive discussions, by all means please share your experiences with these devices.

Understood. EmDrive topics on other forums have the same issue at times. I'll do my best to focus on legitimate questions and not on disruptive comments.

[magnemoe]

34 minutes ago, rfmwguy said:

My friend Paul Kpycla is building a picosat EmDrive here: https://hackaday.io/project/5596-em-drive

Mine exceeds the 10 cm form factor and would not be suitable for cubesats. Basically, I had 750W RF producing 18.4 mN.

Power requirements would also make it unsuitable for cubesats anyway. 
18 mN is high trust compared to the NASA test but they also use less power. 
How is trust tested? would not putting it in an box eliminate trust from heated air? ventilate on sides radial to trust,. Yes heating would be an issue but it don't have to be run for long just so long trust from heated air is eliminated. With so high trust tiny error sources would not be an issue anymore. 
Other error would be electrical or magnetic coupling. 
 

[rfmwguy]

3 minutes ago, magnemoe said:

Power requirements would also make it unsuitable for cubesats anyway. 
18 mN is high trust compared to the NASA test but they also use less power. 
How is trust tested? would not putting it in an box eliminate trust from heated air? ventilate on sides radial to trust,. Yes heating would be an issue but it don't have to be run for long just so long trust from heated air is eliminated. With so high trust tiny error sources would not be an issue anymore. 
Other error would be electrical or magnetic coupling. 
 

Thrust was tested on a torsion beam. I have about 40 vids on youtube, this one was a test stand walkaround: 

 

[rfmwguy]

NASA/Eagleworks EmDrive full disclosure just occured: https://www.reddit.com/r/EmDrive/comments/5b9zfh/emdrive_data/

[Green Baron]

 

Looking forward to the paper and following discussion, and of course successfull reproductions ...

(pls use si units :-))

Edited November 5, 2016 by Green Baron

[Aethon]

Hey rfmwguy.  Welcome.

What happened to the other thread?  I hadn't noticed anything that would cause it to be closed.

[magnemoe]

Nice, 
watching 1701 Emdrive Flight Test #2B - 9/24/15 https://www.youtube.com/watch?v=HPm2oPUPi2Q now, honestly don't understand it. 
What is thermal lift?
Looks like you have serious issues with heating. some might be related to expansions because of heating, however this should be symmetric, outside on arm get an length of arm benefit but it should not be very large. And yes that effect would not be resolved by my box nor an vacuum chamber. only trust from heated air.
On the other hand the EM-drive should give thrust pretty instantly, saw you did some tests with pulsing. 

Edit: weight distributions changes either along the length axis of bar should just move the target marker up and down, an change along the engine trust axis would have more impact but not sure how this would work out. 
Still an 18 mN trust should give an decent twist on the long arm. 

Edited November 5, 2016 by magnemoe

[Scotius]

Just moderation team being cautious. Oh, well - it's their job. Anyway, it's sad news about Traveller's condition I hope he will be able to win his fight with cancer again. Please, send him best wishes.

[rfmwguy]

20 minutes ago, Aethon said:

Hey rfmwguy.  Welcome.

What happened to the other thread?  I hadn't noticed anything that would cause it to be closed.

No idea.

17 minutes ago, magnemoe said:

Nice, 
watching 1701 Emdrive Flight Test #2B - 9/24/15 https://www.youtube.com/watch?v=HPm2oPUPi2Q now, honestly don't understand it. 
What is thermal lift?
Looks like you have serious issues with heating. some might be related to expansions because of heating, however this should be symmetric, outside on arm get an length of arm benefit but it should not be very large. And yes that effect would not be resolved by my box nor an vacuum chamber. only trust from heated air.
On the other hand the EM-drive should give thrust pretty instantly, saw you did some tests with pulsing. 

Edit: weight distributions changes either along the length axis of bar should just move the target marker up and down, an change along the engine trust axis would have more impact but not sure how this would work out. 
Still an 18 mN trust should give an decent twist on the long arm. 

1701 was tested on a teeter-totter and thermal lift is the heated air around the unit causing it to lift up. A small amount of thrust with the original design was measured @ 177 microNewtons. !701A was a torsion beam and thermal lift was not a factor as the twist was in the horizontal plane. 1701A was a new, solid copper cavity and measured 18.4 mN best. Beam displacement made by precision Laser Displacement Sensor with micrometer accuracy.

Pulsing was more of a thermal management issue as the Magnetron heats up to 400F pretty rapidly.

 

[magnemoe]

56 minutes ago, rfmwguy said:

No idea.

1701 was tested on a teeter-totter and thermal lift is the heated air around the unit causing it to lift up. A small amount of thrust with the original design was measured @ 177 microNewtons. !701A was a torsion beam and thermal lift was not a factor as the twist was in the horizontal plane. 1701A was a new, solid copper cavity and measured 18.4 mN best. Beam displacement made by precision Laser Displacement Sensor with micrometer accuracy.

Pulsing was more of a thermal management issue as the Magnetron heats up to 400F pretty rapidly.

Understand. how much did the bar move under trust? have no idea about the resolution of an torsion pendulum 
vut 18 mN would be a bit below 2 gram
Cooling sounds like might be an problem however you could use liquid cooling and run it pretty hot 

You have this from your friend https://www.youtube.com/watch?v=ZSZT5plA4a4 yes its magnetic levitating so any force would get it rotate. Heating would hardly be an issue here, however electromagnetism might be an factor, rotating the engine would answer that, so would using an longer axis to move the magnets outward. 
Putting the engine on an longer arm too. 
Looks like we have an new type of reaction wheel  had been more kerbal to put on an cubesat than an engine along center of mass 

 

 

[rfmwguy]

19 minutes ago, magnemoe said:

Understand. how much did the bar move under trust? have no idea about the resolution of an torsion pendulum 
vut 18 mN would be a bit below 2 gram
Cooling sounds like might be an problem however you could use liquid cooling and run it pretty hot 

You have this from your friend https://www.youtube.com/watch?v=ZSZT5plA4a4 yes its magnetic levitating so any force would get it rotate. Heating would hardly be an issue here, however electromagnetism might be an factor, rotating the engine would answer that, so would using an longer axis to move the magnets outward. 
Putting the engine on an longer arm too. 
Looks like we have an new type of reaction wheel  had been more kerbal to put on an cubesat than an engine along center of mass 

 

 

I am speaking from memory, not the data itself, but the torsion beam was about 107 inches in total length, the LDS measured displacement in millivolts that I calibrated directly with calibrated weights. IIRC, the displacement of the beam was less than about 5-6 mm.

Liquid cooling adds an error component, moving fluids and impellers. This thing has to be totally vibration-free to be accurate.

[Nuke]

6 hours ago, rfmwguy said:

My friend Paul Kpycla is building a picosat EmDrive here: https://hackaday.io/project/5596-em-drive

Mine exceeds the 10 cm form factor and would not be suitable for cubesats. Basically, I had 750W RF producing 18.4 mN.

thats the one that i am most interested in. been watching it for awhile but it seemed the updates kinda fizzled out, at least last i checked nothing new to report. id definately want that particular drive configuration tested by a 3rd party before launching it. it looks nice on paper and its rather miniscule rf power supply is a pretty good bit of electrical engineering. i seem to recal a video of it spinning in a vacuum jar while hovering on a magnetic levitator. though i kind of wondered if there wasnt some interplay between the device and the levitator such that it in essence became a defacto electric motor. id love to see the thing tested on a much better rig. looks like its gonna happen.

Edited November 5, 2016 by Nuke

[magnemoe]

1 hour ago, rfmwguy said:

I am speaking from memory, not the data itself, but the torsion beam was about 107 inches in total length, the LDS measured displacement in millivolts that I calibrated directly with calibrated weights. IIRC, the displacement of the beam was less than about 5-6 mm.

Liquid cooling adds an error component, moving fluids and impellers. This thing has to be totally vibration-free to be accurate.

Pretty noticeable. how does you calibrate it? push with device who showed force and measure the impact?

Water cooling would be for an operational device.  An cooling rib would make sense but avoid the heat pipe cpu coolers as they contain evaporating and moving fluids, unlikely it would have any impact but it can be used against you. 
 

5 minutes ago, Nuke said:

thats the one that i am most interested in. been watching it for awhile but it seemed the updates kinda fizzled out, at least last i checked nothing new to report. id definately want that particular drive configuration tested by a 3rd party before launching it. it looks nice on paper and its rather miniscule rf power supply is a pretty good bit of electrical engineering. i seem to recal a video of it spinning in a vacuum jar while hovering on a magnetic levitator. though i kind of wondered if there wasnt some interplay between the device and the levitator such that it in essence became a defacto electric motor. id love to see the thing tested on a much better rig. looks like its gonna happen.

Not vacuum just an glass clock to avoid moving air to affect it. 
Posted about it above and the motor effect would be an error source, moving the engine around and using an longer axis to increase distance to magnets would reduce that. 

[Nuke]

oh ok i was just recalling from memory. hes going to be using a much better test rig on the device though.

 

https://hackaday.io/project/5596-em-drive/log/48016-emdrive-goes-for-serious-testing

[Shpaget]

Have you tried to run it in vacuum?

If yes, is there a difference in thrust when running the engine in atmosphere and in vacuum?

How hard was the vacuum?

If no, how can you be sure the thrust doesn't come from the air being pushed around?

[rfmwguy]

2 hours ago, Nuke said:

thats the one that i am most interested in. been watching it for awhile but it seemed the updates kinda fizzled out, at least last i checked nothing new to report. id definately want that particular drive configuration tested by a 3rd party before launching it. it looks nice on paper and its rather miniscule rf power supply is a pretty good bit of electrical engineering. i seem to recal a video of it spinning in a vacuum jar while hovering on a magnetic levitator. though i kind of wondered if there wasnt some interplay between the device and the levitator such that it in essence became a defacto electric motor. id love to see the thing tested on a much better rig. looks like its gonna happen.

Paul is reworking heatsink his output chip. Don't know all the details but think it went over temp in vacuum. No airflow = higher temp. Sorry, that's all I know.

27 minutes ago, Shpaget said:

Have you tried to run it in vacuum?

If yes, is there a difference in thrust when running the engine in atmosphere and in vacuum?

How hard was the vacuum?

If no, how can you be sure the thrust doesn't come from the air being pushed around?

No, this homeboy doesn't have access to a chamber yet. Thermals are vertical on all surfaces, so torsion beam would try and lift. Guy wires per video make lifting beam difficult and would not register on Lds. Sealed RF injection point to prevent jetting. Ballooning on teeter totter was significant. Force measured was opposite lift of about 177 microN. Basically it reversed, stalled or attenuated lift. That was a wire screen frustum at much lower Q. Regardless, saw the effect first with teeter totter setup.

1 hour ago, Nuke said:

oh ok i was just recalling from memory. hes going to be using a much better test rig on the device though.

 

https://hackaday.io/project/5596-em-drive/log/48016-emdrive-goes-for-serious-testing

His first vacuum test was done by Dr Tajmar at at Univ of dresden

Edited November 5, 2016 by rfmwguy
Updated

[rfmwguy]

Kocyla emdrive testing: https://hackaday.io/project/5596-em-drive/log/48016-emdrive-goes-for-serious-testing

Edited November 5, 2016 by rfmwguy
Typo

[Scotius]

Will professor Tajmar oversee the tests? I wonder if it will change his mind about feasibility of EmDrive? I remember critical talks on NASASpaceflight about low quality of his test engine.

Anyway, i have another question. As of now, test engines are already better than various ion engines used in space. What should we do to get even better performance? Higher power input? Bigger chamber? Several small chambers linked together?

[rfmwguy]

5 hours ago, Scotius said:

Will professor Tajmar oversee the tests? I wonder if it will change his mind about feasibility of EmDrive? I remember critical talks on NASASpaceflight about low quality of his test engine.

Anyway, i have another question. As of now, test engines are already better than various ion engines used in space. What should we do to get even better performance? Higher power input? Bigger chamber? Several small chambers linked together?

Tajmar is overseeing testing at his lab. Everything you mentioned is valid except bigger chamber. Chamber size is based upon frequency used. A bigger chamber would mean a much lower frequency. That has not been experimented with yet to my knowledge, only higher/smaller.

[Nuke]

about that frequency, is there any indication if higher frequencies lead to better drives? i figure you could get drives about half the size of the 24 ghz cubesat thuster before you start hitting the far infrared. going the other way with huge cavities at lower frequencies seems like it would bite into your power to weight ratio and lead to rather bulky engines. getting things smaller seems to me you could swarms of micro probes that can be shotgunned to some location in the solar system (say the asteroid belt) for a detailed survey.

[rfmwguy]

9 minutes ago, Nuke said:

about that frequency, is there any indication if higher frequencies lead to better drives? i figure you could get drives about half the size of the 24 ghz cubesat thuster before you start hitting the far infrared. going the other way with huge cavities at lower frequencies seems like it would bite into your power to weight ratio and lead to rather bulky engines. getting things smaller seems to me you could swarms of micro probes that can be shotgunned to some location in the solar system (say the asteroid belt) for a detailed survey.

Higher frequencies can reduce size but power output and stability can be a factor. No one is certain about what the sub 1ghz drives might do but mass would definitely increase. 25 GHz is highest so far. When it becomes accepted technology as I think it will be there should be a broad spectrum of experimentation.