-
#3680 by zen-in on 08 Jul, 2016 20:46
-
...
Yes I have and (it looks like Daves with just the single diode voltage doubler ckt ) it runs stable under 1MHz BW with little splattering or drift (takes about 15-20 seconds to start to stabilize). The key was the water jet cooling to keeping the signal from drifting. Dave got to see the thermal drift and was lucky that his cavity resonance was a lower TE013 and it could drift into it and lock.
The spectral harmonics have decreased as have the amplitude modulation splattering, but I still have some "noise" in the spectrum but it is down starting under 12-14db.
Shell
Is that the DC filtered PS spectra or the original? I see just one spectra in that video. I have an older Tektronix 491 spectrum analyzer that has a dispersion/resolution control. I can also vary the sweep rate and with the phospher persistance. I have found that playing with those controls can give me a better idea of what the RF spectra I am looking at really is. For example is it a narrow band signal that is wobbling in frequency or just a lot of simultaneous tones that are close together? With your video there may be some frequency aliasing happening. Can you integrate the spectra over time? Is there any way of changing the dispersion and resolution? It's possible some of the spectral spiles you are seeing are from reflections and are an artifact of the instrument. I have done side-by-side comparisons of spectrum analyzers and observed some will show image signals that are internally generated. A video display may aliase the image signals, especially if the CF is not changing. On my 491 the images go in one direction at 2x or 3x and the real signals go in the opposite direction. Newer spectrum analyzers suppress images very well. That's the difference between a $15k spectrum analyzer and a $50 one. -
#3681 by Tellmeagain on 08 Jul, 2016 21:22
-
LOL, I have 5 microwave skeletons. Yes, I know the microwave secondaries provide a heater tap but I needed a little more voltage (for regulation) and to keep the issues caused by the dual series HV transformers (http://sound.westhost.com/xfmr2.htm) out of the heater output and regulator.Since high voltage relays rated for the power levels involved tend to be very specialized (transmitter applications and such) and therefore really expensive, another way may be better:
Transformers from old microwave ovens are cheap and easy to come by, so one can just use two of them. Let one transformer supply only the filament (with the high voltage winding left not connected), while a second transformer supplies only the high voltage (with the filament winding going nowhere). That way the problem of switching the high voltage can be moved to the primary (mains) side, where it's easy to solve with commonly available affordable relays.
If you've ever taken apart a microwave oven (and I'm on #2 now), you will see that the power supply transformer already has two secondaries, one a low-voltage AC to keep the filament warm. The second is half-rectified into half-wave pulsed HVDC.
Shell
Did you decide not to use the Panasonic ultrasonic power supply? -
#3682 by SeeShells on 08 Jul, 2016 21:23
-
That video is of Dave's unaltered power supply driving his magnetron. As to what your seeing on Dave's magnetron the answer, your seeing AM modulations introduced by the introduction of the AC from the heater and the phase shifted from that from the voltage doubler, the splatter back from the heater, and the interactions of all effecting the magnetrons internal cavities generating the RF. It is a wide band spectral soup consisting of both even and odd harmonics with a AM modulation shifting the entire group. Then shift it all down in frequency as the magnetron expands from the heat generated VWSR and being kicked back from the frustum.
...
Yes I have and (it looks like Daves with just the single diode voltage doubler ckt ) it runs stable under 1MHz BW with little splattering or drift (takes about 15-20 seconds to start to stabilize). The key was the water jet cooling to keeping the signal from drifting. Dave got to see the thermal drift and was lucky that his cavity resonance was a lower TE013 and it could drift into it and lock.
The spectral harmonics have decreased as have the amplitude modulation splattering, but I still have some "noise" in the spectrum but it is down starting under 12-14db.
Shell
Is that the DC filtered PS spectra or the original? I see just one spectra in that video. I have an older Tektronix 491 spectrum analyzer that has a dispersion/resolution control. I can also vary the sweep rate and with the phospher persistance. I have found that playing with those controls can give me a better idea of what the RF spectra I am looking at really is. For example is it a narrow band signal that is wobbling in frequency or just a lot of simultaneous tones that are close together? With your video there may be some frequency aliasing happening. Can you integrate the spectra over time? Is there any way of changing the dispersion and resolution? It's possible some of the spectral spiles you are seeing are from reflections and are an artifact of the instrument. I have done side-by-side comparisons of spectrum analyzers and observed some will show image signals that are internally generated. A video display may aliase the image signals, especially if the CF is not changing. On my 491 the images go in one direction at 2x or 3x and the real signals go in the opposite direction. Newer spectrum analyzers suppress images very well. That's the difference between a $15k spectrum analyzer and a $50 one.
I don't have a video of mine the capture software will not work with the OS I have. It's on my list to upgrade the OS... along with a gazzilion other things.
It does look like this picture and stable except it takes about 10-20 seconds to settle the center frequency as the magnetron heats up.
Shell
Speeellng corecton
-
#3683 by SeeShells on 08 Jul, 2016 21:26
-
I did decide not to because I wanted better control over the power output and the rheostat full wave rectified variable PS gave me that.
LOL, I have 5 microwave skeletons. Yes, I know the microwave secondaries provide a heater tap but I needed a little more voltage (for regulation) and to keep the issues caused by the dual series HV transformers (http://sound.westhost.com/xfmr2.htm) out of the heater output and regulator.Since high voltage relays rated for the power levels involved tend to be very specialized (transmitter applications and such) and therefore really expensive, another way may be better:
Transformers from old microwave ovens are cheap and easy to come by, so one can just use two of them. Let one transformer supply only the filament (with the high voltage winding left not connected), while a second transformer supplies only the high voltage (with the filament winding going nowhere). That way the problem of switching the high voltage can be moved to the primary (mains) side, where it's easy to solve with commonly available affordable relays.
If you've ever taken apart a microwave oven (and I'm on #2 now), you will see that the power supply transformer already has two secondaries, one a low-voltage AC to keep the filament warm. The second is half-rectified into half-wave pulsed HVDC.
Shell
Did you decide not to use the Panasonic ultrasonic power supply?
Shell -
#3684 by SeeShells on 08 Jul, 2016 21:29
-
http://forum.nasaspaceflight.com/index.php?topic=39772.msg1558072#msg1558072Quote
If you've ever taken apart a microwave oven (and I'm on #2 now), you will see that the power supply transformer already has two secondaries, one a low-voltage AC to keep the filament warm. The second is half-rectified into half-wave pulsed HVDC.
.
The reason I used another stand alone transformer for the heater was if I varied the voltage on the AC feeding the dual high voltage transformers I'd also vary the heater voltage as well.
Shell -
#3685 by zen-in on 08 Jul, 2016 22:03
-
That video is of Dave's unaltered power supply driving his magnetron. As to what your seeing on Dave's magnetron the answer, your seeing AM modulations introduced by the introduction of the AC from the heater and the phase shifted from that from the voltage doubler, the splatter back from the heater, and the interactions of all effecting the magnetrons internal cavities generating the RF. It is a wide band spectral soup consisting of both even and odd harmonics with a AM modulation shifting the entire group. Then shift it all down in frequency as the magnetron expands from the heat generated VWSR and being kicked back from the frustum.
...
Yes I have and (it looks like Daves with just the single diode voltage doubler ckt (deleted video) it runs stable under 1MHz BW with little splattering or drift (takes about 15-20 seconds to start to stabilize). The key was the water jet cooling to keeping the signal from drifting. Dave got to see the thermal drift and was lucky that his cavity resonance was a lower TE013 and it could drift into it and lock.
The spectral harmonics have decreased as have the amplitude modulation splattering, but I still have some "noise" in the spectrum but it is down starting under 12-14db.
Shell
Is that the DC filtered PS spectra or the original? I see just one spectra in that video. I have an older Tektronix 491 spectrum analyzer that has a dispersion/resolution control. I can also vary the sweep rate and with the phospher persistance. I have found that playing with those controls can give me a better idea of what the RF spectra I am looking at really is. For example is it a narrow band signal that is wobbling in frequency or just a lot of simultaneous tones that are close together? With your video there may be some frequency aliasing happening. Can you integrate the spectra over time? Is there any way of changing the dispersion and resolution? It's possible some of the spectral spiles you are seeing are from reflections and are an artifact of the instrument. I have done side-by-side comparisons of spectrum analyzers and observed some will show image signals that are internally generated. A video display may aliase the image signals, especially if the CF is not changing. On my 491 the images go in one direction at 2x or 3x and the real signals go in the opposite direction. Newer spectrum analyzers suppress images very well. That's the difference between a $15k spectrum analyzer and a $50 one.
I don't have a video of mine the capture software will not work with the OS I have. It's on my list to upgrade the OS... along with a gazzilion other things.
It does look like this picture and stable except it takes about 10-20 seconds to settle the center frequency as the magnetron heats up.
Shell
Speeellng corecton
I don't see how 60 Hz (half wave rectified AC) will generate splatter. There is such a thing as intermod but you need a non linear element like a diode, metal junction, etc and the mixing frequency has to be reasonably close. At best you will get products that are 2.5 GHz +/- (n * 60 Hz) where -10 < n < +10. If the 60 Hz is AM modulating the magnetron output you would see spikes on either side of the 2.5 GHz magnetron output. However most spectrum analyzers don't have the resolution to see those spikes and so you would have to use a demodulation feature. A 60 Hz AM modulation would produce negligiable broadening of the magnetron output. I have not played with magnetrons and don't claim to be an expert. This opinion is just basic RF physics. I think the "splatter: you are seeing in Dave's magnetron output may be due to thermal effects, an RF mis-match, or may just be images in the spectrum analyzer. It is worthwile if you can get your hands on another ($$) better quality spectrum analyzer, for a second opinion. -
#3686 by SeeShells on 08 Jul, 2016 22:45
-
That video is of Dave's unaltered power supply driving his magnetron. As to what your seeing on Dave's magnetron the answer, your seeing AM modulations introduced by the introduction of the AC from the heater and the phase shifted from that from the voltage doubler, the splatter back from the heater, and the interactions of all effecting the magnetrons internal cavities generating the RF. It is a wide band spectral soup consisting of both even and odd harmonics with a AM modulation shifting the entire group. Then shift it all down in frequency as the magnetron expands from the heat generated VWSR and being kicked back from the frustum.
...
Yes I have and (it looks like Daves with just the single diode voltage doubler ckt (deleted video) it runs stable under 1MHz BW with little splattering or drift (takes about 15-20 seconds to start to stabilize). The key was the water jet cooling to keeping the signal from drifting. Dave got to see the thermal drift and was lucky that his cavity resonance was a lower TE013 and it could drift into it and lock.
The spectral harmonics have decreased as have the amplitude modulation splattering, but I still have some "noise" in the spectrum but it is down starting under 12-14db.
Shell
Is that the DC filtered PS spectra or the original? I see just one spectra in that video. I have an older Tektronix 491 spectrum analyzer that has a dispersion/resolution control. I can also vary the sweep rate and with the phospher persistance. I have found that playing with those controls can give me a better idea of what the RF spectra I am looking at really is. For example is it a narrow band signal that is wobbling in frequency or just a lot of simultaneous tones that are close together? With your video there may be some frequency aliasing happening. Can you integrate the spectra over time? Is there any way of changing the dispersion and resolution? It's possible some of the spectral spiles you are seeing are from reflections and are an artifact of the instrument. I have done side-by-side comparisons of spectrum analyzers and observed some will show image signals that are internally generated. A video display may aliase the image signals, especially if the CF is not changing. On my 491 the images go in one direction at 2x or 3x and the real signals go in the opposite direction. Newer spectrum analyzers suppress images very well. That's the difference between a $15k spectrum analyzer and a $50 one.
I don't have a video of mine the capture software will not work with the OS I have. It's on my list to upgrade the OS... along with a gazzilion other things.
It does look like this picture and stable except it takes about 10-20 seconds to settle the center frequency as the magnetron heats up.
Shell
Speeellng corecton
I don't see how 60 Hz (half wave rectified AC) will generate splatter. There is such a thing as intermod but you need a non linear element like a diode, metal junction, etc and the mixing frequency has to be reasonably close. At best you will get products that are 2.5 GHz +/- (n * 60 Hz) where -10 < n < +10. If the 60 Hz is AM modulating the magnetron output you would see spikes on either side of the 2.5 GHz magnetron output. However most spectrum analyzers don't have the resolution to see those spikes and so you would have to use a demodulation feature. A 60 Hz AM modulation would produce negligiable broadening of the magnetron output. I have not played with magnetrons and don't claim to be an expert. This opinion is just basic RF physics. I think the "splatter: you are seeing in Dave's magnetron output may be due to thermal effects, an RF mis-match, or may just be images in the spectrum analyzer. It is worthwile if you can get your hands on another ($$) better quality spectrum analyzer, for a second opinion.
I would LOVE to have better tools to dig into this, it has been one of my greatest frustrations in building this at home DYI Kitchen Sink experiment. So if someone would like to help this old gal out to reach a small goal which hasn't changed this last year in my gofundme link... then thank you.
That said, much still can be done to understand the basics of the electronic hardware we are using to generate RF and produce cavities.
On your thoughts in how the magnetron splatters please think about this. I have a voltage doubler that will produce a voltage doubled "pulse" and that pulsed output phase shifting is highly dependent on the load it drives. (I can see this shifting on my usb scope.) In this video they are locked on and triggered so the pulse seems stable. ... it's not.
If we add into the pulsed voltage doubled HV to the magnetron the AC to the filament which the 60HZ is quite stable and see the harmonics between the two interact with the microwave creation in the magnetron cavity we begin to see how the frequencies can splatter and shift.
My goal was to thermally stabilize the magnetron by a water cooling jacket and provide the tube with a clean HV and heater DC. I was being reminded of the horrible "hum" that I'd hear in old radios with leaky filter caps introducing 60Hz AC into the tubes then then out to the speakers. (Ya I know that dates me
)
This help?
My Best,
Shell -
#3687 by rq3 on 08 Jul, 2016 22:52
-
Yes I have and (it looks like Daves with just the single diode voltage doubler ckt ) it runs stable under 1MHz BW with little splattering or drift (takes about 15-20 seconds to start to stabilize). The key was the water jet cooling to keeping the signal from drifting. Dave got to see the thermal drift and was lucky that his cavity resonance was a lower TE013 and it could drift into it and lock....
That's quite close to what I've done.
Using a "matched" set of microwave transformers (very important, if not smoke may ensue) in series, driven by a variable rheostat, which gives me a clean full wave rectified and filtered variable >4KV output.
The heater is driven by another transformer and then full wave rectifying the output. Filtering that for a clean DC voltage, to be voltage and current control no matter the loading on the dual transformers. The filament voltage and current can be switched off after a set time to further stabilize the water cooled magnetron. All we are interested in is heating the heater
to suck off electrons whether it's DC or AC, it doesn't matter it will just be more stable with no AC added to the magnetron's output.
...
Have you compared the two power supplies (AC Vs filtered DC) by recording the spectral output of the magnetron? Where is the noise and what kind of noise is it? Do you get a narrow band low phase noise output with the filtered DC supply?
The spectral harmonics have decreased as have the amplitude modulation splattering, but I still have some "noise" in the spectrum but it is down starting under 12-14db.
Shell
A spectrum analyzer is a radio receiver, usually superheterodyne, with all the faults and foibles inherent with that design. Overdrive of the front end WILL provide evidence of measurement artifacts that are not real. A zero IF (intermediate frequency) digital spectrum analyzer is even worse, since you must also deal with spectrum folding (Fourier imaging) due to the sample rate.
On any spectrum analyzer, the input attenuator is your best friend. If you can reduce intermods in relation to the input level by reducing the input signal, you are over-driving.
Most spectrum analyzers would like to see an input level, AT THE DETECTOR, of about -30 dBm for best signal to noise ratio and intermodulation distortion. You can't just stuff an RF (radio frequency) signal in to the instrument, and believe what you see on the display. -
#3688 by zen-in on 08 Jul, 2016 22:53
-
...
On your thoughts in how the magnetron splatters please think about this. I have a voltage doubler that will produce a voltage doubled "pulse" and that pulsed output phase shifting is highly dependent on the load it drives. (I can see this shifting on my usb scope.) In this video they are locked on and triggered so the pulse seems stable. https://www.youtube.com/watch?v=I2k2g00onL0?t=331 ... it's not.
If we add into the pulsed voltage doubled HV to the magnetron the AC to the filament which the 60HZ is quite stable and see the harmonics between the two interact with the microwave creation in the magnetron cavity we begin to see how the frequencies can splatter and shift.
My goal was to thermally stabilize the magnetron by a water cooling jacket and provide the tube with a clean HV and heater DC. I was being reminded of the horrible "hum" that I'd hear in old radios with leaky filter caps introducing 60Hz AC into the tubes then then out to the speakers. (Ya I know that dates me)
This help?
My Best,
Shell
I don't know what you mean by splatter. When I hear splatter I think of the time I started frying bacon and eggs one morning ...
Power line hum in any kind of audio electronics (I think it's mostly the harmonics of 60 Hz we hear) is understandable. Both are in the audio frequency range and it's just AM modulation of an audio signal.
-
#3689 by rq3 on 08 Jul, 2016 22:56
-
That video is of Dave's unaltered power supply driving his magnetron. As to what your seeing on Dave's magnetron the answer, your seeing AM modulations introduced by the introduction of the AC from the heater and the phase shifted from that from the voltage doubler, the splatter back from the heater, and the interactions of all effecting the magnetrons internal cavities generating the RF. It is a wide band spectral soup consisting of both even and odd harmonics with a AM modulation shifting the entire group. Then shift it all down in frequency as the magnetron expands from the heat generated VWSR and being kicked back from the frustum.
...
Yes I have and (it looks like Daves with just the single diode voltage doubler ckt (deleted video) it runs stable under 1MHz BW with little splattering or drift (takes about 15-20 seconds to start to stabilize). The key was the water jet cooling to keeping the signal from drifting. Dave got to see the thermal drift and was lucky that his cavity resonance was a lower TE013 and it could drift into it and lock.
The spectral harmonics have decreased as have the amplitude modulation splattering, but I still have some "noise" in the spectrum but it is down starting under 12-14db.
Shell
Is that the DC filtered PS spectra or the original? I see just one spectra in that video. I have an older Tektronix 491 spectrum analyzer that has a dispersion/resolution control. I can also vary the sweep rate and with the phospher persistance. I have found that playing with those controls can give me a better idea of what the RF spectra I am looking at really is. For example is it a narrow band signal that is wobbling in frequency or just a lot of simultaneous tones that are close together? With your video there may be some frequency aliasing happening. Can you integrate the spectra over time? Is there any way of changing the dispersion and resolution? It's possible some of the spectral spiles you are seeing are from reflections and are an artifact of the instrument. I have done side-by-side comparisons of spectrum analyzers and observed some will show image signals that are internally generated. A video display may aliase the image signals, especially if the CF is not changing. On my 491 the images go in one direction at 2x or 3x and the real signals go in the opposite direction. Newer spectrum analyzers suppress images very well. That's the difference between a $15k spectrum analyzer and a $50 one.
I don't have a video of mine the capture software will not work with the OS I have. It's on my list to upgrade the OS... along with a gazzilion other things.
It does look like this picture and stable except it takes about 10-20 seconds to settle the center frequency as the magnetron heats up.
Shell
Speeellng corecton
I don't see how 60 Hz (half wave rectified AC) will generate splatter. There is such a thing as intermod but you need a non linear element like a diode, metal junction, etc and the mixing frequency has to be reasonably close. At best you will get products that are 2.5 GHz +/- (n * 60 Hz) where -10 < n < +10. If the 60 Hz is AM modulating the magnetron output you would see spikes on either side of the 2.5 GHz magnetron output. However most spectrum analyzers don't have the resolution to see those spikes and so you would have to use a demodulation feature. A 60 Hz AM modulation would produce negligiable broadening of the magnetron output. I have not played with magnetrons and don't claim to be an expert. This opinion is just basic RF physics. I think the "splatter: you are seeing in Dave's magnetron output may be due to thermal effects, an RF mis-match, or may just be images in the spectrum analyzer. It is worthwile if you can get your hands on another ($$) better quality spectrum analyzer, for a second opinion.
It generates splatter because it's an AM (amplitude modulated) oscillator being driven to 100% modulation depth with an approximation of a square wave. The Fourier frequencies extend (theoretically) to infinity. You get the fundamental carrier of the magnetron, +/- all of the Fourier frequencies. -
#3690 by FattyLumpkin on 08 Jul, 2016 23:55
-
Shell, have you started testing yet? Am still waiting for results (baited breath)!
FL
-
#3691 by SeeShells on 09 Jul, 2016 00:12
-
Shell, have you started testing yet? Am still waiting for results (baited breath)!
FL
Sigh, right now I'm going through profiling and fine tuning everything! Very time consuming. If I find something that is not working just right, like I just did with the laser yesterday I correct it so it doesn't impact the tests. So sorry it does take time.
Shell
-
#3692 by 1 on 09 Jul, 2016 00:24
-
Also worth remembering that the magnetron itself, like all vacuum tubes, is an inherently non-linear device. At best, it's operated in a regime that's approximately linear; i.e. steady state. Making no claims to having intimate knowledge of anyones setup, but I find it quite plausible that high voltage pulses could cause frequency dispersion.
-
#3693 by zen-in on 09 Jul, 2016 02:47
-
It generates splatter because it's an AM (amplitude modulated) oscillator being driven to 100% modulation depth with an approximation of a square wave. The Fourier frequencies extend (theoretically) to infinity. You get the fundamental carrier of the magnetron, +/- all of the Fourier frequencies.
If you are still talking about 60 Hz then no that is incorrect. The Fourier coefficients of a square wave do not all have the same amplitude. They fall off very rapidly. To get a sideband 1 MHz away from the nominal magnetron center frequency (choose any frequency +/- 10 MHz) there would have to be an harmonic 17,000 times the 60 Hz. Anyway there is no square wave. The half-wave rectified AC is half a sine wave, not a square wave. Define "splatter". What is the mechanism according to the established physics of RF energy that creates this "splatter"? Also AM modulation does not create phase noise or a broadband signal.
I think most of what is seen in rfmwguy's spectral plot is from the magnetron RF overdriving the spectrum analyzer he is using. You were right in your earlier post. He needs to put more attenuators on the input and maybe put the unit in a well shielded box. -
#3694 by Tellmeagain on 09 Jul, 2016 05:02
-
It generates splatter because it's an AM (amplitude modulated) oscillator being driven to 100% modulation depth with an approximation of a square wave. The Fourier frequencies extend (theoretically) to infinity. You get the fundamental carrier of the magnetron, +/- all of the Fourier frequencies.
If you are still talking about 60 Hz then no that is incorrect. The Fourier coefficients of a square wave do not all have the same amplitude. They fall off very rapidly. To get a sideband 1 MHz away from the nominal magnetron center frequency (choose any frequency +/- 10 MHz) there would have to be an harmonic 17,000 times the 60 Hz. Anyway there is no square wave. The half-wave rectified AC is half a sine wave, not a square wave. Define "splatter". What is the mechanism according to the established physics of RF energy that creates this "splatter"? Also AM modulation does not create phase noise or a broadband signal.
I think most of what is seen in rfmwguy's spectral plot is from the magnetron RF overdriving the spectrum analyzer he is using. You were right in your earlier post. He needs to put more attenuators on the input and maybe put the unit in a well shielded box.
I think what you said is convincing. I googled some microwave oven magnetron spectrum images online,
Besides the one shell found, here is another one,
and https://www.cst.com/Applications/Article/Magnetron-And-Microwave-Oven-Design-To-Solve-Wi-Fi-Interference-Issues
-
#3695 by 1 on 09 Jul, 2016 06:43
-
To be fair, the rectified signal must contain a square wave component as it can be mathematically expressed with such. Although, I have a bit of confusion how a half wave rectified AC signal turns into 60Hz. I suspect we might be discussing a full wave rectified signal. That said, I must agree with zen-in that the harmonics aren't causing the signal seen.
For a full wave signal, we can express that signal as a sine wave multiplied by a square wave of the same period. Negative components of both multiply to become positive.
The Fourier transform of such is then given by the convolution of the transforms of each individually
Likewise, a half wave signal can be expressed as a multiplication of a sine wave with a square wave containing a DC offset (to bring the wave from 2V to 0 instead of V to -V).
Re-normalizing the amplitude, this transform looks similar:
What's important is the summation terms in both. As the cosine terms necessarily cannot become greater than 1, we see that the amplitude of each harmonic falls off as the power of that harmonic squared. The 10,000th harmonic of 60Hz (even less for 17k) at best contributes one part in 400 million to the overall energy of the signal. This corresponds to a -86db drop from peak; and although I can't read the spectrum analyzer display very clearly, I suspect it broadens far earlier than that.
-
#3696 by Tcarey on 09 Jul, 2016 07:48
-
LOL, I have 5 microwave skeletons. Yes, I know the microwave secondaries provide a heater tap but I needed a little more voltage (for regulation) and to keep the issues caused by the dual series HV transformers (http://sound.westhost.com/xfmr2.htm) out of the heater output and regulator.Since high voltage relays rated for the power levels involved tend to be very specialized (transmitter applications and such) and therefore really expensive, another way may be better:
Transformers from old microwave ovens are cheap and easy to come by, so one can just use two of them. Let one transformer supply only the filament (with the high voltage winding left not connected), while a second transformer supplies only the high voltage (with the filament winding going nowhere). That way the problem of switching the high voltage can be moved to the primary (mains) side, where it's easy to solve with commonly available affordable relays.
If you've ever taken apart a microwave oven (and I'm on #2 now), you will see that the power supply transformer already has two secondaries, one a low-voltage AC to keep the filament warm. The second is half-rectified into half-wave pulsed HVDC.
Shell
Shell, What voltage do you need for your filament circuit? I think I saw the current draw was about 5 amps?
It seems to me that for Dave and perhaps you, that using a Lipo battery to drive the heater circuit would be an easy way to eliminate the heater wire heating issues that Dave is facing without having to deal with the much larger power requirements of the HV circuit. Since the heater circuit is switched off once the maggie gets operating, the total AH required wouldn't be that high. A single Lipo battery is about 3.7 volts. If that is enough to light off the maggie then it would be relatively easy to implement. RC aircraft batteries are rated at much higher current draws so that part is not an issue. I'm just thinking out loud here. -
#3697 by SeeShells on 09 Jul, 2016 13:45
-
To be fair, the rectified signal must contain a square wave component as it can be mathematically expressed with such. Although, I have a bit of confusion how a half wave rectified AC signal turns into 60Hz. I suspect we might be discussing a full wave rectified signal. That said, I must agree with zen-in that the harmonics aren't causing the signal seen.
Regarding just the power supplies using a iron core transformer which describes most of the unmodified used here in tests run so far and also includes the new style inverter driven from Panasonic.
For a full wave signal, we can express that signal as a sine wave multiplied by a square wave of the same period. Negative components of both multiply to become positive.
The Fourier transform of such is then given by the convolution of the transforms of each individually
Likewise, a half wave signal can be expressed as a multiplication of a sine wave with a square wave containing a DC offset (to bring the wave from 2V to 0 instead of V to -V).
Re-normalizing the amplitude, this transform looks similar:
What's important is the summation terms in both. As the cosine terms necessarily cannot become greater than 1, we see that the amplitude of each harmonic falls off as the power of that harmonic squared. The 10,000th harmonic of 60Hz (even less for 17k) at best contributes one part in 400 million to the overall energy of the signal. This corresponds to a -86db drop from peak; and although I can't read the spectrum analyzer display very clearly, I suspect it broadens far earlier than that.
That source pulsed DC and along with the AC heater cathode noise (which can be amplified in the tube) we can start to see the levels of complexity that this system operates in. Even how the noise dynamics of the heater anode can be amplified in the tube we start to see the complexity of the microwave output waveform of a kitchen microwave.
A very good paper on the dynamics of the magnetron. Warning... PDF
https://deepblue.lib.umich.edu/bitstream/handle/2027.42/71064/JAPIAU-22-4-433-1.pdf;jsessionid=893A38118FC2A9FD53B427931826CF7A?sequence=2
https://www.jlab.org/ir/MITSeries/V6.PDF
These papers from The UofMi and MITSeries references (along with what seems countless others) educated me to the point where I currently am, using a clean HV DC power supply and low voltage 6.3DC 10 amp DC voltage regulated filament current and water jacket cooling on the magnetron.
Shell -
#3698 by SeeShells on 09 Jul, 2016 13:55
-
I might have the same, have 2 USBs one is a SMA and the other a SA and both show the same basic center Fo with even and odd harmonics out of the standard microwave power supply magnetron. http://www.allaboutcircuits.com/textbook/alternating-current/chpt-7/more-spectrum-analysis/
While I am not an expert in magnetron tubes, I do have thoughts on this. In plain language Shell, this is a free-running oscillator by design. Rather than the expense of phase locking a fundamental to the molecular resonance of H2O, designers chose to fire a broad-band signal around 2.45 GHz that would contain multiple "spikes" so when thermal drifting, there would always be a component at or near molecular resonance of water, which heats the food.
That video is of Dave's unaltered power supply driving his magnetron. As to what your seeing on Dave's magnetron the answer, your seeing AM modulations introduced by the introduction of the AC from the heater and the phase shifted from that from the voltage doubler, the splatter back from the heater, and the interactions of all effecting the magnetrons internal cavities generating the RF. It is a wide band spectral soup consisting of both even and odd harmonics with a AM modulation shifting the entire group. Then shift it all down in frequency as the magnetron expands from the heat generated VWSR and being kicked back from the frustum.
...
Yes I have and (it looks like Daves with just the single diode voltage doubler ckt ) it runs stable under 1MHz BW with little splattering or drift (takes about 15-20 seconds to start to stabilize). The key was the water jet cooling to keeping the signal from drifting. Dave got to see the thermal drift and was lucky that his cavity resonance was a lower TE013 and it could drift into it and lock.
The spectral harmonics have decreased as have the amplitude modulation splattering, but I still have some "noise" in the spectrum but it is down starting under 12-14db.
Shell
Is that the DC filtered PS spectra or the original? I see just one spectra in that video. I have an older Tektronix 491 spectrum analyzer that has a dispersion/resolution control. I can also vary the sweep rate and with the phospher persistance. I have found that playing with those controls can give me a better idea of what the RF spectra I am looking at really is. For example is it a narrow band signal that is wobbling in frequency or just a lot of simultaneous tones that are close together? With your video there may be some frequency aliasing happening. Can you integrate the spectra over time? Is there any way of changing the dispersion and resolution? It's possible some of the spectral spiles you are seeing are from reflections and are an artifact of the instrument. I have done side-by-side comparisons of spectrum analyzers and observed some will show image signals that are internally generated. A video display may aliase the image signals, especially if the CF is not changing. On my 491 the images go in one direction at 2x or 3x and the real signals go in the opposite direction. Newer spectrum analyzers suppress images very well. That's the difference between a $15k spectrum analyzer and a $50 one.
I don't have a video of mine the capture software will not work with the OS I have. It's on my list to upgrade the OS... along with a gazzilion other things.
It does look like this picture and stable except it takes about 10-20 seconds to settle the center frequency as the magnetron heats up.
Shell
Speeellng corecton
In disassembling a couple of mags, they have an electron ring with multiple chambers (cavities). I suspect each chamber or cavity oscillates at a slightly offset fundamental, thus the spattering/mix of fundamentals.
My theory only (as I haven't read this online) is that the reason each "spike" appears is not due to voltage deviations, but chamber design. Count the chambers, count the spikes and know that sum and difference freqs will appear. This is probably the reason your much better regulated supply still shows broadband dispersion.
To satisfy those with differing opinions, this is my theory only and I am not claiming authority on mags. Also, the mags are potentially hazardous, no one should tear one apart without knowledge of hazardous materials (fortunately I have that). Regarding spec-an front end overload, I simply take off-air measurements and have located the antenna a foot to 6 feet away plus engaged and disabled the built in 20dB LNA...all you get is lower signal strength, but the appearance of the near-band spikes remains. I also have a 30dB attenuator I have put in line...same thing...you probably have the same SMA atten I have.
Quick primmer.
Shell -
#3699 by Monomorphic on 09 Jul, 2016 14:12
-
It was the cheap power adapter I was using to power the laser displacement sensor that was causing major problems. So I broke out my handy bench-top power supply and now it's silky smooth.
Also included is the chart with first "tap and wind" tests. This is just a screen capture, I did not log the data. The first bump is from a light tap, the second is from me blowing air. I am also moving around in the room, and have yet to tighten the oil dampening paddle, so I expect these results - which I am already very happy with - to get even better.
)
to suck off electrons 
FL