Double the windings on the vc, but make the second half wound in the opposite direction, and extend it out of the gap towards the back. It will buck the vc coil's influence on the magnetic circuit, will not provide force, will halve the efficiency.
As long as the primary coil is sufficiently under hung there will be sufficient flux path on the face to keep all it's field within the circuit. The second half of the vc could remain stationary, but that will not attack the secondary effects of vc velocity modulation of the inductive energy.
Gap faces should never be solid metal, either laminated .5 mm plates or ferrite extending thick enough to be away from the vc dB/dt. That will also stop eddy induced inductance modulation.
The ttc from 18 sound is close to that, they put the second coil to use by adding a second gap with opposite flux direction. However as depicted, they overhung it so the cancellation is not perfect.
Remember, the overhung coil going in remains fully coupled to the magnetic circuit, but overhang leaving the gap does not.
Cost is their driver, but not ours..
Jn
Last edited:
Hey Chris,
In the broadcast world all transmit system impedances are 50 ohms. FM Yagis sold for broadcast use (receive or transmit) are also usually 50 ohms like this Kathrein-Scala unit (also available in 75 ohms) which is built like a tank and costs the better part of a kilobuck: http://www.kathreinusa.com/wp-content/uploads/2017/03/CL-FM.pdf
When I really have to pay attention to impedances is building 300 or 600 ohm open-wire feeders in amateur radio. 1500 watts into 50 ohm coax is ~274 VRMS or 387 VPeak. The same 1500 watts into an 600 ohm open wire feeder results in 949 VRMS or 1342 VPeak. As you can appreciate, material insulation and stand-off distances are critical at this voltage. Add to this the fact that unlike DC, RF can capacitively couple an arc through an insulator, and you can quickly get a mess.
Cheers!
Howie
And what I meant was RF can create a charged plasma through an insulator.
An arc being a continuous stream of electrons creating a low-impedance path, whereas a plasma is relatively higher impedance with a cloud of ionizing atoms providing the conductive path. I'm sure JN would be able to provide a more concise definition of the difference between an arc and plasma. If enough current is available to a plasma it can avalanche into an arc.
In optical disc sputtering we employed plasma power supplies which used the impedance differential to detect and quench arcs and maintain a sputtering plasma. It was the development of these high speed control loop supplies which enabled inexpensive disc production, with full cycle metalization cycles of less than 1 second. Before that discs were metalized en mass in very large, slow vacuum chambers with cycle times of an hour or more.
More ancient trivia...
Howie
Howie, I make DC plasmas all the time in a sputter coater for some of our metal depositions (I try to avoid the practice much outside of work though. ). RF generated plasmas are definitely needed for insulating materials to capacitively couple the energy past the barrier.
Plasma is an electrically neutral gas of positively charged ions and their respectively-stripped electrons, to be thorough. And arc would be dumping a whole lotta current through that conductive plasma. (Then we have to get into space charge regions and Debye lengths, et cetera, et cetera)
). RF generated plasmas are definitely needed for insulating materials to capacitively couple the energy past the barrier. Plasma is an electrically neutral gas of positively charged ions and their respectively-stripped electrons, to be thorough. And arc would be dumping a whole lotta current through that conductive plasma.
(Then we have to get into space charge regions and Debye lengths, et cetera, et cetera)
Double the windings on the vc, but make the second half wound in the opposite direction, and extend it out of the gap towards the back. It will buck the vc coil's influence on the magnetic circuit, will not provide force, will halve the efficiency.
As long as the primary coil is sufficiently under hung there will be sufficient flux path on the face to keep all it's field within the circuit. The second half of the vc could remain stationary, but that will not attack the secondary effects of vc velocity modulation of the inductive energy.
Gap faces should never be solid metal, either laminated .5 mm plates or ferrite extending thick enough to be away from the vc dB/dt. That will also stop eddy induced inductance modulation.
The ttc from 18 sound is close to that, they put the second coil to use by adding a second gap with opposite flux direction. However as depicted, they overhung it so the cancellation is not perfect.
Remember, the overhung coil going in remains fully coupled to the magnetic circuit, but overhang leaving the gap does not.
Cost is their driver, but not ours..
Jn
Sounds like I am back to trying the dual voice coil I have in the subs. May not be as perfect as a designed MFB coil could be.... but usable.
-RNM
Womp, womp, wahhhhhhh. Haha.
I've been out of the microelectronics fab for a while due to the nature of my work, although we're going to be setting up a small clean room here at my new institution. Biggest stuff I work on is 4"-6" (100-150 mm) wafers for research, though. Not your monster!
We are spraying ceramic trays, which is a kind of kiln furniture, used for firing MLCC's.
But we would still need a tracking bridge to get rid of the thermal VC resistance shifts, with the bridge extracting the true EMF voltage by subtracting the i*Z_vc component, wouldn't we? For a David Birt style servo a bifilar wound main VC would be nice (AC signal nulls in the output, DC test current magnetically cancels itself and has no effect).
With your proposals wrt compensated underhung VC and improved magnetic design, do I understand it correctly that this would lower and, more importandly, stabilize the VC "static" impedance as well as it stabilizes the EMF voltage it reports back, making the effort of building a velocity control loop around this much easier?
Last edited:
Ebeam evap is a pretty common practice in small fabs especially when we want uniform <10 nm films. Sputter coating forms too large of grains, and I can show pictures any time under SEM of discontinuous grains of gold from our sputter coater at 4-10 nm but it's still very useful for reducing charging. Not sure what the big guys are doing, I'm betting some form of ALD (atomic layer deposition, essentially alternating gas phase chemistries that form monolayers on the surface to get extremely uniform/conformal coatings)
Last edited:
Yes, with the SEM the target builds up a charge on it and over loads the detectors. Just enough coating... a molecule
thick of gold is enough to discharge the charge buildup to ground. I had responsibility to keep a one-off custom SEM (the guy who was the principal designer left LLNL), running and had the chemist/operator show me how to use the sputtering machine. I learned the SEM from the schematics left behind. Then I used it to coat some Mogami RCA jacks when no one was in the SEM room. I sputtered all sorts of metals and exotics onto them so I could tests them in audio apps. Was fun. But they said I was using up too much gold and so I had to stop.
I wanted to know what difference could be measured/detected between ultra pure gold coatings and 'doped' gold used commercially for longer wear.
Fond memories.
-Richard
Last edited:
BTW -- the Chemistry group asked me to see what I could do... the chemists said the SEM wasnt as detailed as it used to be. The images were not so sharply defined anymore. I found that the amps were DC coupled and had low offset and drift … but no servo on them... so over time, it needed rezeroing on several discrete bipolar transistor amps. There were trim pots for it. I suggested they servo those amps with a mod. Stays sharp. ( they had not heard of dc servo then).
Dont know if it was increased amp distortion (2H) that caused the loss of resolution/focus. Didn't pursue it further. Once they knew they needed a E-Tech to keep it sero'ed and tuned up, I never went in the SEM room afterwards. It just reminded me of increased distortion reducing audible sharpness/definition also.
-Richard[/QUOTE]
Dont know if it was increased amp distortion (2H) that caused the loss of resolution/focus. Didn't pursue it further. Once they knew they needed a E-Tech to keep it sero'ed and tuned up, I never went in the SEM room afterwards. It just reminded me of increased distortion reducing audible sharpness/definition also.
-Richard[/QUOTE]
Last edited:
That's about it.
Analog out of digital equipment is most often 2Vrms/+6dBV for 0dBFS for hifi home components (cd players, hifi DACs and the like) and half that for soundcards, phones, etc.
For broadcasting, this BBC paper explains quite well the alignement levels and how the new -23 LUFS standard is coming into play.
That's interesting I've seen 5 1/2 digit voltmeters from the 60's made with tubes, the DC was very stable. I guess physicists need to get out more too. BTW if you want DC response how does the servo help?
Last edited:
- Status
- Not open for further replies.