A double-tuned 7th-harmonic filter for 7.16MHz calls for T30-6 cores, but only T37-6 cores are on hand. What's the trade-off? How does the larger size affect filter performance? TIA.
73,-- Don Kuenz, KB7RPU Genius is nothing but a greater aptitude for patience. - de Buffon
A double-tuned 7th-harmonic filter for 7.16MHz calls for T30-6 cores, but only T37-6 cores are on hand. What's the trade-off? How does the larger size affect filter performance? TIA. Answered my own question. A T30-6 core has an AL value of 36 versus
30 for a T37-6 core. So, you need a few more turns on a T37-6 core to keep the status-quo.73,
-- Don Kuenz, KB7RPU Genius is nothing but a greater aptitude for patience. - de Buffon
AL=3.6
AL=3.0
The dimensions are also somewhat different. I don't have the Amidon spec book and don't want to go through this ordeal process to get one: So, I'll guess(tm) that there will be a change in Q with a difference core and change in number of turns, but no clue whether up or down or how much.
I don't have an instant answer because I don't know the type of filter and the calculated values of the various components. I suggest you build an LTspice model of the filter using the original values and adjust the Q of each toroid coil to match the change in material and see what happens. For example, here's one I did about 5 years ago for an FM broadcast band notch filter for a 2m repeater receiver:
Ideal inductors:
Real inductors with lower Q:
If you want to tinker with this filter, the files are all in:
Build a model of your filter and compare the responses with different Q inductors.
If you don't like LTspice for filters, try AADE v4.5
or Elsie 2.77
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
Note that, if the core isn't stuffed full of wire, you can adjust the inductance by scrunching the turns closer together or further apart. Downside: an incomplete winding has more stray flux. May want to keep more distance between inductors (than, like, stacked together as close as possible), but heck, if it's a soft filter it probably doesn't matter much.
At such low A_L and relatively low frequency, I'm guessing it won't be a problem to adjust it by whole turns and capacitors.
Tim
-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website: https://www.seventransistorlabs.com/
My filter uses trimmer caps, so whole turns are OK. My ALs come from an ancient Amidon datasheet that measures AL in uH per 100 turns, but toroids.info measures AL in uH per 1000 turns. That's why the ALs differ by a factor of 10. Jeff, the OD of a T37 core is 0.068" larger than the OD of a T30. The filter calls for 10 turns on a T30 core. When you plug that data into toroids.info it turns out that an equivalent inductor on a T37 core needs 11 turns.
73,-- Don Kuenz, KB7RPU Genius is nothing but a greater aptitude for patience. - de Buffon
My filter uses trimmer caps, so whole turns are OK. My ALs come off an ancient Amidon datasheet that measures AL in uH per 100 turns, but toroids.info uses an oddball AL whose metric is ? mH / Turns^2 ? Jeff, the OD of a T37 core is 0.068" larger than the OD of a T30. The filter calls for 10 turns on a T30 core. When you plug that data into toroids.info it turns out that an equivalent inductor on a T37 core needs 11 turns.
73,-- Don Kuenz, KB7RPU Genius is nothing but a greater aptitude for patience. - de Buffon
Those are all made by Micrometals, and as of the last couple years, they have proper datasheets now:
A_L is always a "per turns squared" unit, that's the physics. Why there's a tradition of labeling things as "at 100 turns" is beyond me. It's probably as old as the parts themselves. It's like using gauss versus tesla, a 10^4 difference. (Note that MM uses cgs units still, but they do at least use the scaled base unit for A_L.)
Everything here is simple ratios. You can plug the numbers into Google, and get out whatever unit you like. Just dimensional analysis: if the dimensions multiply or divide together to get another unit, you are, at worst, off only by a constant (like a pi or something).
For that, it can help to track turns as a dimension itself, in which case you need to associate turns with another underlying unit, like A_e/N (which is reasonable in the sense that, for every turn of wire, you get that many times the cross sectional area under the wire).
If you aren't worried about nonlinear properties (power loss, saturation, etc.), or deriving properties from base units (A_L = A_e * mu / l_e, note that mu = mu_r * mu_0, where mu_0 ~= 1.257 uH/m in SI units), you won't need anything besides A_L.
Tim
-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website: https://www.seventransistorlabs.com/
0.068" is a tiny change. Since both cores are the same material, there isn't going to much of a change in Q and since you have trimmer caps, the substitution should be "close enough".
-- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
The difference in turns is around 10%. For small number of turns it may be a challenge to get near enough. Also the AL spread may cover the difference.
-- -TV
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.