LT Spice lets you create inductors with the John Chan hysterisis model. I did it for a well-gapped RM10 core, and got about 105dB. When I wound the coil and sent if off to my expert friend in London, his test gear confirmed the result.
This was at 17kHz. You'd probably want a nickel-zinc ferrite for a high Q at 1MHz. You can pull the data for the John Chan model out of the manufacturer's data sheets, though it does take a bit of ferreting around.
-- Bill Sloman, Sydney
If you can make the measurement at just one frequency why not? They certainly have deep nulls.
Prolly want to use something old and big, if you can find it -- 50mA is a bit much for today's crystals. Check out Surplus Sales of Nebraska and see if they still exist, and if so, if they still sell crystals.
-- Tim Wescott Wescott Design Services http://www.wescottdesign.com I'm looking for work -- see my website!
How much capacitor did you need?
It's probably self-capacity of the winding. Multilayer windings are
*terrible*.One interesting side-effect of this: if you look at the turns ratio of a generic audio transformer (like one of those little 1k:8 ohm jobbies), with respect to frequency, it drops off sharply above the self-resonant frequency: the gain from primary to secondary actually /increases/, because the bulk of the primary winding is effectively shorting itself out!
Tim
-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website: http://seventransistorlabs.com
I worked in that department for a few months, back in 1984. Stanford had a pretty useful rotation program where grad students could try out as many as three research groups before buying in. (Of course the professors had a sa y too.)
My supervisor for that fairly brief period was Peter Banks, who was Sally R ide's advisor BITD. I gave up on the aero/astro thing for several reasons-- mainly because I preferred having my project to myself, but also because my office neighbour had spent four years designing and building a small "geta way special" satellite, only to have it wind up in the Atlantic when its De lta booster failed.
Peter was a great guy--he first proposed a tethered satellite experiment in 1969 that finally flew in 1986 iirc. That was one of the other reasons. ;)
Cheers
Phil Hobbs
Well, that's encouraging.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
Around 300pf to resonate at 1MHz.
Multilayer is bad for Q to. I know the SRF is caused by the self capacitance, I have never ran into I minor SR peak, it is about 1/7th the amplitude of the major peak. And I think if I measured very close, they are harmonically related. Uh, Uh! I just found even though my HP8640B is set to 1.4MHz, my scope signal is at 2.8MHz, the minor peak is the second harmonic. This is a poor showing for my HP8640B! I tried my HP651B SigGen and it does not show the 1,4MHz peak. Take note Larkin, no minor peak.
I think/thought that was capacitive coupling between Pri and Sec that effect. Mikek
-- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus
You can help both self capacitance by winding the layers properly. Winding one layer end to end is the worse as it puts sections of wire next to each other that have higher potentials. Winding the layers one turn at a time reduces the potential and so the effective capacitance.
6 9 12 15 3 5 8 11 14 1 2 4 7 10 13 ...-- Rick C
On a sunny day (Wed, 10 May 2017 16:17:28 -0700) it happened John Larkin wrote in :
Sounds like possible. I once even had a 100 kHz xtal... in my home build nixie frequency countyer Very big.
4 or 5 1MHz xtals to make a small bandfilter. You see that in ham radio projects a lot.
I'm sure it's not the signal generator; that pesky coil must be doubling it! The dastardly thing! :^)
Yeah, that way it stays progressive, like a single solenoid, but where each turn is somewhat wider than it's "supposed" to be, and each turn is actually three turns. :)
More extreme methods use few-layer pancake sections and stack them: pi-wound chokes.
It's a construction method that's not wholly obsolete. I discovered this beastie in a CRT monitor:
CCFL inverter/ballasts of course are a prominent modern application.
The generally applicable method is to have as much distance as possible, between the two ends of the full winding. Making full layers, and doubling back on each layer, is the worst possible method.
Consider a two layer winding: the start and end are in the same place, turn for turn. The start and end can be unwound together as one (well, if there's no tape between them to hang things up), one going "left" around the bobbin, at the same time and place the other end is going "right". Thus, down to the length of a single turn, the winding is a parallel-wire transmission line.
You should therefore expect the TDR of such a winding to show an electrical length (propagation + reflection) equal to the total copper length. For typical magnet wire, the impedance will be quite low, perhaps 20 ohms. It will be a bit dispersive, because the turns aren't perfectly paired, but overlap their neighbors a bit. On top of this, there will be the undulations due to the turns going in opposite directions (left- and right-handed, as viewed from their respective ends), with a time period equal to the electrical length of a single turn.
Additional layers behave as these undulating transmission lines stacked end to end. For a high impedance winding (like a RFC), obviously you want the total impedance quite high, therefore you want many turns to lie inbetween the start and finish. You also want to minimize copper wire length, and layer width and count, and maximize spacing between turns (lattice winding), which keeps the electrical length short, and the between-turns and between-layers transmission line impedance high.
For the same reason (transmission line theory), in a transformer, capacitive coupling from primary to secondary (but not from end to end of a given winding -- phew!) is desirable, to the extent that it matches leakage inductance as the desired operating frequency. (That is, the cutoff frequency of a transformer is 1 / (2*pi*sqrt(LL*Cp)), and bandwidth is maximal when source and load are sqrt(LL/Cp) ohms. You can use this in reverse to find the maximum tolerable winding length, and primary-to-secondary gap, for a transformer design!)
Tim
-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website: http://seventransistorlabs.com
I think it is the generator (but hope it's not), when I switched to my older lower cost, less bells and whistles, HP651B, there was no sign of it. I will add, when I made my driving loop, I was to lazy to search out a
50 ohm resistor, so I shorted the output with a clip lead to make the driving loop. I would hope that would not cause distortion, But...Mikek
-- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus
For high-Q coil, you should aim to approximately square geometry, length equals diameter. To minimize the capacitance effects of multilayer, the classical method is honeycomb winding, tedious, but good.
Please remember to use gloves when handling the Litz, to prevent grease from the fingers getting to the wire.
-- -TV
a quartz crystal is a good idea or you want to use the "magnified L" circuit.
Put a cap directly across the L and then a smaller cap to your signal.
Set the parallel resonance is below your trap frequency so the combination looks like an L that is considerably larger than the actual L.
This "magnified L" is series resonated with your small C to form the trap.
This makes the trap narrower but less deep.
m
Oh, only 100? That is not pushing the technology very hard, then. Should be easy to do with generic materials.
Jon
Probably each layer has its own self resonance, and then the capacitance between layers has another resonance. Generally, you want to space the turns out, and put some space between the layers to cut these parasitic capacitances. That, of course, reduces the inductance. So, it is all a compromise.
Jon
As a Canadian, they probably wouldn't allow you to be an astronaut.
One fun thing about doing instrumentation design is that you can be an electronic dilettante, and not work on one thing for decades. Like, I don't wanna grow up, I want to be a Toys-R-Us kid.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Agreed. Over my career I've changed fields about every 5 or 6 years, which is enormous fun. These days I'm doing way too many front ends, which is one reason I'm making some canned designs to sell.
Some of them are pretty interesting still--this noise canceller I just finished up for the U. of Oxford folks has some new wrinkles that are likely to make it into a product at some stage. The MAT14s have so much better matching and better thermal tracking than the HFA3xxx arrays that it's worth working pretty hard to coax them out to higher frequency.
The spherical-cow model says that it can be tweaked to 80 dB cancellation out to nearly 10 MHz, but of course that involves subtracting two signals to a (tweaked) accuracy of 1 part in 10**4, and since I'm not using SuperSpice I can't trust it to that accuracy. ;)
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
I get to see lasers, jet engines, accelerators, semi fabs, tanker ships, superconductive magnets, all sorts of stuff, without making any of them a career. Fun. It's also nice to help non-electronic people to make their stuff work better.
There are times when the thing that you need is a trimpot.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
I have used formulas and SW from this page. There is at one web calculator using these too. The writers have made their own studies and compared those with the old ones.
They seem to say that the self res>
They let that Canadian manipulator arm be an astronaut though...
On a sunny day (Thu, 11 May 2017 12:11:51 -0700) it happened John Larkin wrote in :
Are you referring to NASA?
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.