EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: cdev on December 26, 2015, 05:12:24 pm
-
Hi!
I am an SWL hoping to make a HF preselector and need an inductor that's continuously variable throughout its range.
Would it be worth the extra energy to use a cut piece of copper sheeting or tape as the coil conductor? I've read somewhere in the past that by using a tape-like conductor I could then expect more "Q" than a plain wire. The coil would likely be wound fairly tightly around a piece of plastic like a PVC pipe or re-purposed pill bottle or similar.
My goal is to get as maximum a "Q" i.e. "peakiness" as is possible in a practical project.
This is to tune out unwanted very strong other HF shortwave stations. (Religious and other unwanted programming) :palm:
-
"Hi-Q" inductors at HF frequency means very large structure, a price to pay for adjustable hi-q variable value.
http://www.eham.net/articles/9641 (http://www.eham.net/articles/9641)
-
Do not use PVC pipe to wind it, the PVC will be a lot more lossy than other materials and will drop the Q considerably. You probably will want a low loss plastic as a former, or wind it so that it is either formerless and self supporting, or use enamelled wire and varnish to keep it firm. If you can get silver plating solution ( any number of methods to plate a thin coat of silver onto a wire) or silver coated wire that will reduce loss as well. thicker wire ( bare or enamelled insulation) is better as well, though if you can get copper tape it will work.
-
Relative to what? -- If you have lots of fine wire, building a Litz assembly might be better (depending on how much tedium you're up for!). If you have heavy (solid) wire, or tube, you can build a large solenoid with good Q.
Tape or foil is tricky because it's thin and high aspect ratio. The thinness is good for skin effect, but it only helps if you can interleave it, otherwise it acts like two solid wires (one for each edge, where most of the current flows) with a shield inbetween that carries little to no current!
Power inductors are often wound from flat wire, edgewise. This gives a lot of current density in a small package, but at high frequencies, makes things quite a bit worse relative to the amount of copper used. This is acceptable for a power inductor, but probably not optimal for RF.
As mentioned -- if you can grow large, do it!
Tim
-
As mentioned -- if you can grow large, do it!
I have the space to make a larger coil In practical terms I think the ideal size for a box would be maybe 15 cm wide and 5 cm tall.
It seems to make more sense to use a non-metallic case. I can use bare copper wire with some plastic string in between the windings as a spacer- to keep the winding evenly spaced.
Then have a tap that moves along it. I don't need knobs as much as I need the higher selectivity.
-
Do not use PVC pipe to wind it, the PVC will be a lot more lossy than other materials and will drop the Q considerably. You probably will want a low loss plastic as a former
Do you specifically mean any kind of plastic? I was thinking perhaps the orange plastic in pill bottles might work well for me. Its quite rigid, even brittle. Also, some of them are fairly large- Most importantly, they are quite strong and a hole drilled in them wont grow, I may have some clear plastic bottles too.
How to ascertain if a prospective coil form is lossy. That's an interesting problem.
-
Don't know offhand what pill bottles are. Is there a recycling number on the bottom?
PVC isn't a big deal below VHF, but it gets more important the higher you go, and also if the voltage / impedance is high (which is probably the case for a highly selective tuner).
The microwave test is fairly representative: stick it in the microwave for a little while (plus a glass of water on the other side, for load). If it gets warm (or melts, or worse..), it's probably not very good.
Tim
-
Good preselectors are multi stage and not just a single high Q tank circuit. The compact military ones that I used consisted of a ECW coil with a long ferrite slug that moved in and out of the fibre glass coil former. The ferrite slug was approx 10mm diameter x 100mm long and driven by a worm drive. The redifon professional receiver auto preselector were different. They used IIRC a two stage preselector that had fixed inductors and variable capacitors that were motor driven. The HF bands were covered in discrete 'chunks' and so the inductors were switched in as required. It was a very high performance preslector as we often transmitter on a frequency very close tomourvreceive frequency for operational reasons. The standard Plessey PR2250 preslector was inadequate for such.
None of the designs that I have seen used unusual inductor construction such as you are describing.
Single stage predilection is usually a pretty poor performer if wide tuning range is required. I can post the schematic of the Lowe PR150 preslector if it helps. That works quite well and is 2 stage but has multiple preslector banks to cover the different bands. Sadly a single wide band preselection stage will perform poorly in your applicatio. PR150 units sell for over $300 in the UK as the design works well. Normal RF rated inductors are used in combination with reed relays. I have a couple of PR150's that I will be selling but sadly the price is eye watering due to rarity. DIY is the cheapest option.
Another option is a high Q H field antenna. Such an antenna would normally be a tuned loop but sadly such would only cover one limited frequency span in order to maintain the selectivity. On lower frequencies the Ferrite tuned loop offers both high Q and excellent directivity (also good for eliminating strong unwanted signals)
If you have a seriously powerful local transmitter that you wish to avoid, you could build a nice sharp notch filter to effectively attenuate just that transmission.
Good luck.
Fraser
-
Hi Fraser,
Thank you for all of this info. Yes, a multistage preselector using either adjustable inductors or capacitors or both- for optimum performance are what I am hoping to be able to build, so any schematics are quite welcome.
I have seen photos of some elaborate ones that use a plunger to tune several inductors at the same time.
I am not averse to building a multi-stage preselector. I have most of the parts needed and can usually try things out fairly quickly, so its immediate gratification to some extent.
Good preselectors are multi stage and not just a single high Q tank circuit. The compact military ones that I used consisted of a ECW coil with a long ferrite slug that moved in and out of the fibre glass coil former. The ferrite slug was approx 10mm diameter x 100mm long and driven by a worm drive. The redifon professional receiver auto preselector were different. They used IIRC a two stage preselector that had fixed inductors and variable capacitors that were motor driven. The HF bands were covered in discrete 'chunks' and so the inductors were switched in as required. It was a very high performance preslector as we often transmitter on a frequency very close tomourvreceive frequency for operational reasons. The standard Plessey PR2250 preslector was inadequate for such.
None of the designs that I have seen used unusual inductor construction such as you are describing.
Single stage predilection is usually a pretty poor performer if wide tuning range is required. I can post the schematic of the Lowe PR150 preslector if it helps. That works quite well and is 2 stage but has multiple preslector banks to cover the different bands. Sadly a single wide band preselection stage will perform poorly in your applicatio. PR150 units sell for over $300 in the UK as the design works well. Normal RF rated inductors are used in combination with reed relays. I have a couple of PR150's that I will be selling but sadly the price is eye watering due to rarity. DIY is the cheapest option.
Another option is a high Q H field antenna. Such an antenna would normally be a tuned loop but sadly such would only cover one limited frequency span in order to maintain the selectivity. On lower frequencies the Ferrite tuned loop offers both high Q and excellent directivity (also good for eliminating strong unwanted signals)
If you have a seriously powerful local transmitter that you wish to avoid, you could build a nice sharp notch filter to effectively attenuate just that transmission.
Good luck.
Fraser
Loop antennas are what made me realize that preselection is what I need. I have had very good reception at times with loops. Sometimes a loop is just the ticket.
But sometimes I also get indoor noise - it probably comes from my monitor or computer, and its much less of an issue with an outdoor antenna.
Basically, Ive had two "loops" - two simple high Q tuned circuits, loosely coupled to one another, have had fairly decent performance using something like that as a preselector. But I think that it would work better if the loops were isolated from one another more.
So, yes, any schematics or ideas you have are very welcome! Basically my area of interest is HF and maybe MW and below- not VLF for now. Just want to be able to get a bit more stuff than I have been able to up to now, with a bit more consistency.
-
Hi,
Many years ago I investigated hi-Q inductors for the (U.S.) AM broadcast range. I even went so far as to buy a used HP 4342A to measure some coils.
As far as coil forms go, the best, by far, is air. If you have to use some support, use the minimum you can get by with. I used "legs" with grooves in them:
(https://farm6.staticflickr.com/5669/23923656231_2832d44608_c.jpg)
The best commonly available dielectric is HDPE, but not all HDPE is created equal and you'll never get anyone to give you any specs for it. Better yet is Teflon, but for large-ish coils that can get expensive.
Some tests of dielectrics are reported here:
http://www.bentongue.com/xtalset/24Cmnts/24Cmnts.html (http://www.bentongue.com/xtalset/24Cmnts/24Cmnts.html)
See table 4 near the very end of the paper.
And more information than you probably would ever want can be found here:
http://www.g3ynh.info/zdocs/magnetics/index.html (http://www.g3ynh.info/zdocs/magnetics/index.html)
http://www.g3ynh.info/zdocs/comps/index.html (http://www.g3ynh.info/zdocs/comps/index.html)
The microwave oven test is often recommended for testing dielectrics, but it doesn't actually work. See here:
http://www.g3ynh.info/zdocs/comps/part_6.html (http://www.g3ynh.info/zdocs/comps/part_6.html)
(scroll down to section 2-27).
As far as winding for hi-Q goes, you want to start with a paper by Medhurst on the subject:
http://www.g3ynh.info/zdocs/refs/Medhurst/Medhurst1947.html] [url]http://www.g3ynh.info/zdocs/refs/Medhurst/Medhurst1947.html (http://[url) [/url]
The "rules of thumb" you often see given for "optimum" coils are not all that helpful. Being a software guy, I wrote a program that for a given inductance used brute force to examine *all possible* combinations of commonly available wire diameters, reasonable spacing, reasonable coil diameter, etc. and calculated Q based on the formulas in the above papers. The program then spits out the highest-Q coil dimensions for a given inductance.
My first try resulted in a coil 8" diameter with 10 gauge bare copper that had a measured Q of between 800 and 1000. Unfortunately, it also functions as a very good antenna and more accurate measurements required an RF-quiet room which I started but never completed.
Generally speaking, at least for the coils in my inductance range (around 200-250uH), the spacing between the wires should be equal to the wire diameter. For small-diameter wires, an easy way to achieve this is to wind two wires next to each other, then remove one of the wires. That eliminates one of the variables (wire spacing). The larger the diameter of the wire, the better, but the more costly the coil will be, though for small coils it isn't a big deal (my 8" coil cost $80 a number of years ago). The larger the diameter of the coil, the better; Tim is right on this--think BIG! Length of the coil should be approximately equal to the diameter. You want bare wire rather than enamelled.
If you give me your desired inductance and max physical coil size, I can run some numbers for you.
--Steve
-
Is PVC pipe OK up to 500 kHz? Values from 60uH to 10uH?
-
I don't recall seeing loss measurements below 500kHz. Given that you can get HDPE in small sizes (look for recycling number 2) of containers, I would guess that would be a better route. A few years ago a lot of the Crystal Radio guys went crazy for 4" polystyrene sewer (DWV) pipe that could (sometimes) be gotten from the likes of Home Depot/Lowes because that was significantly better than PVC. My *guess* is that almost anything is better than PVC, but as I say, I've not seen any measurements below 500kHz.
--Steve
-
I wouldn't think < 10MHz would be a problem, even for PVC at high Q.
Tim
-
Glass performs pretty well ? Used jam jars and such like with a glue strip in four locations to hold the coil in place.
Then again, the use of RF ferrite cored chokes is very common in receiver pre selectors. Nice and compact and they do the job if used correctly.
Remember that large scale air core inductors and their associated connecting leads can become an antenna in themselves unless screened well.
I will dig out the Lowe PR150 schematics tomorrow ( in bed sick today).
-
Thank you for all this info. There is a lot of information here.
-
I don't have coil measurement info handy, but in the cited Ben Tongue paper, the table I mentioned has a column that shows the effect of the dielectric on the Q of a capacitor at 1500 kHz.
PVC gives a Q of 59, HDPE gives Q=1120. Glass is not listed and I don't have a measurement for glass close at hand.
--Steve
-
Most glasses will be not quite as good as HDPE. A surprisingly good choice is dry wood. Should be quite a bit better than HDPE. The trick is to keep it dry.
-
Some neat stuff I found Googling on inductors-
:)
"
William Thomson, 1st Baron Kelvin (1824-1907)
“ I fear I have taxed your patience too long; but I cannot sit down without again reflecting on the men whose lives have been patterns to the world. There is none more remarkable perhaps than the man of this country, Joseph Henry, who ended his days here. He and Faraday were patterns of scientific investigators. In some degree they went parallel and made similar scientific discoveries. Henry, indeed, preceded Faraday in the great discovery of the electro-magnetic induction between unmoved conductors. Henry gave the warmest welcome to all practical applications of his discoveries. He sought to make none himself, not because he superciliously despised the applications of science to the public good, but because his own convictions constrained him to go on in pure science; because he felt, as Professor Eowland has said so well in respect to Faraday, that it would have taken him from his work to have devoted himself at all to the practical applications of his discoveries. But what a beautiful trait of character it is to see what a kindly welcome he gave to those who did make the practical applications. He saw what might be done, but deliberately left it to others. ""
Lord kelvin on Joseph Henry..
----
Early test equipment!
https://www.princeton.edu/ssp/joseph-henry-project/galvanometer-picts/ (https://www.princeton.edu/ssp/joseph-henry-project/galvanometer-picts/)
https://www.princeton.edu/ssp/joseph-henry-project/joseph-henry/ (https://www.princeton.edu/ssp/joseph-henry-project/joseph-henry/)
https://www.princeton.edu/ssp/joseph-henry-project/joseph-henry/copper-ribbon/ (https://www.princeton.edu/ssp/joseph-henry-project/joseph-henry/copper-ribbon/)
https://www.princeton.edu/ssp/joseph-henry-project/ohm/The_Galvanic_Circuit_Investigated_Mathem.pdf (https://www.princeton.edu/ssp/joseph-henry-project/ohm/The_Galvanic_Circuit_Investigated_Mathem.pdf) (by a Dr. G. S. Ohm)
"Albany electromagnet !"
https://www.princeton.edu/ssp/joseph-henry-project/albany-electromagnet/ (https://www.princeton.edu/ssp/joseph-henry-project/albany-electromagnet/)
There's lots more..