LPF without ferrite toroids

[Dom13c]

As most of you guys who replied back to my posts earlier can remember, I have a ferrite issue all the time. I think there is a lot of guys out here who've been in ham radio for like years. All what I wanna know is that have any of you guys ever built a lpf without those "Brought from Mars(at least for me)" ferrite cores. Say ,by using pvc/nylon washer former to form the coil? If so, how good it is ( efficiency/performance)  when compared to those ferrite wound coils? Also, how to shield the coils if I do make them on a PVC pipe former.


The problem isn't quite the unavailablity, it's the price of 'em. For an example, I can buy 100 piece of BC547, 3 piece IRF510 with the money that it takes to import a single T30-2 toroid!!

[Whales]

"Air cored inductors" (or air cored transformers) are the terms you are after.  Plastics are similar to air for most magnetic core applications.

If you are in a situation that normally uses a ferrite core: expect to have to do a lot of air-core windings to get the same inductance.  It may be impractical, depending on what you need.  EDIT: hundreds to thousands of times more windings in actual fact.  Core inductance is almost directly proportional to (relative) permeability.

What country are you in?  People might be able to give advice on secret core suppliers.

[vk4ffab]

As most of you guys who replied back to my posts earlier can remember, I have a ferrite issue all the time. I think there is a lot of guys out here who've been in ham radio for like years. All what I wanna know is that have any of you guys ever built a lpf without those "Brought from Mars(at least for me)" ferrite cores. Say ,by using pvc/nylon washer former to form the coil? If so, how good it is ( efficiency/performance)  when compared to those ferrite wound coils? Also, how to shield the coils if I do make them on a PVC pipe former.


The problem isn't quite the unavailablity, it's the price of 'em. For an example, I can buy 100 piece of BC547, 3 piece IRF510 with the money that it takes to import a single T30-2 toroid!!

Sure, air cores are fine in low pass and bandpass filters etc, they are more practical at VHF and UHF but if space is not an issue you can wind them for HF, by using the lowest practical amount of inductance and refactoring the capacitance and matching in and out Z to suit.

[vk4ffab]

I have added 2 pictures, this is 9 turns of 1mm enamelled wire on a 22mm form giving 1.9uH is about right for a 5th order 40m butterworth lowpass filter. By playing with the number of turns and the diameter of the winding you can get the inductance you require. And by using some clear nail polish or some kind of glue you can keep the windings neat and tidy, giving greater Q. And when it comes to filters, its all about the Q of the inductors.

Its just physically larger than a T50-6, so if space on the board is not a problem or the mechanics of an air core is not a problem, wind away and you will be fine. There is nothing magical about using Iron or Ferrite cores.

[radiolistener]

if you want to minimize loss and improve Q factor, you're needs to remove all core and use air or even vacuum as inductor core

Inductor loss also depends on wire resistance. Due to skin effect there is no needs to use silver wires, you can use just silver plated copper wire in order to reduce heat loss.

[Qmavam]

We don't know where you are, but digikey and mouser have a selection of toroids/
Here's a digikey search of toroids, prices from $0.10 to $2.84.
https://www.digikey.com/products/en/magnetics-transformer-inductor-components/ferrite-cores/936?k=ferrite&k=&pkeyword=ferrite&sv=0&v=240&pv1012=6&sf=1&FV=ffe003a8&quantity=&ColumnSort=0&page=1&pageSize=100
 I don't know how much power, so don't the size you need and I don't know what material to use for what frequency.
Shipping costs are always a problem.
                                     Mikek

[Dom13c]

"Air cored inductors" (or air cored transformers) are the terms you are after.  Plastics are similar to air for most magnetic core applications.

If you are in a situation that normally uses a ferrite core: expect to have to do a lot of air-core windings to get the same inductance.  It may be impractical, depending on what you need.  EDIT: hundreds to thousands of times more windings in actual fact.  Core inductance is almost directly proportional to (relative) permeability.

What country are you in?  People might be able to give advice on secret core suppliers.

I'm from India. Agree with the impracticality of the number of turns of wire but that's the only choice I'm left with. I live near Kolkata and have searched several times at nearby markets  but got nothing. the case is that there are very few people out here in India who are aware of ham radio . so the ham related producats are vere scarce here.they hardly keep these things due to the slim market demand

[Dom13c]

We don't know where you are, but digikey and mouser have a selection of toroids/
Here's a digikey search of toroids, prices from $0.10 to $2.84.
https://www.digikey.com/products/en/magnetics-transformer-inductor-components/ferrite-cores/936?k=ferrite&k=&pkeyword=ferrite&sv=0&v=240&pv1012=6&sf=1&FV=ffe003a8&quantity=&ColumnSort=0&page=1&pageSize=100
 I don't know how much power, so don't the size you need and I don't know what material to use for what frequency.
Shipping costs are always a problem.
                                     Mikek



I'm after QRP level (10-30 watt)HF only(3-30 MHz).  I'm from India .  yup, the overseas shipping cost really sucks and thank you for the reference of digikey. I've been there before but after checking the shipping price I just told myself "nope. Ain't a way I can spend this much just for shipping"  QRP- guys also sells them for decent price but IT'S ALL 'BOUT SHIPPING.

[Dom13c]

I have added 2 pictures, this is 9 turns of 1mm enamelled wire on a 22mm form giving 1.9uH is about right for a 5th order 40m butterworth lowpass filter. By playing with the number of turns and the diameter of the winding you can get the inductance you require. And by using some clear nail polish or some kind of glue you can keep the windings neat and tidy, giving greater Q. And when it comes to filters, its all about the Q of the inductors.

Its just physically larger than a T50-6, so if space on the board is not a problem or the mechanics of an air core is not a problem, wind away and you will be fine. There is nothing magical about using Iron or Ferrite cores.

(Attachment Link) e

Board space really isn't an issue. But the stray capacitance and shielding might be an issue. Any thought on how to reduce them?

[David Hess]

Board space really isn't an issue. But the stray capacitance and shielding might be an issue. Any thought on how to reduce them?

Air cores are not self shielding like torroids and other forms so you have to play the usual games with orientation, spacing, and shielding to prevent coupling between them.

Coil form materials can be tested in a microwave oven to estimate relative loss.  (1)  Some commonly available plastics have excessive loss and definitely should not be used for power inductors like tank coils.

(1) Place several samples in the microwave oven along with a cup of water.  Then check the relative temperatures of the plastics after heating to get an idea of the loss of each one.

[hagster]

That test only really makes sense at microwave frequencies. At those frequencies you can normally get tiny surface mount air core of ceramic wound inductors. I assume the op was working at much lower frequencies.

[David Hess]

That test only really makes sense at microwave frequencies. At those frequencies you can normally get tiny surface mount air core of ceramic wound inductors. I assume the op was working at much lower frequencies.

It is still useful for comparing relative loss and loss is more difficult to measure at low frequencies anyway.  Where this really comes into play is tank coils and transmitter filters where the high power levels can melt the coil form if the material has too much loss.

[bd139]

Another option is winding your inductors on tap washers.

[vk4ffab]

Contact Farhan at rfsignals.com or Sunil at Indiakits.com i think it is, they should be able to get you cores if you trust the Indian mail system to be able to deliver them to you. I think Farhan uses Diz from Kitsandparts.com to ship amidon iron cores to India and Sunil is about to get shielded IF inductorslocally, I ordered a pile off him a while ago because getting shielded IF inductors from 300nH to 2.2uH range are hard to get in VK.

[vk4ffab]

I have added 2 pictures, this is 9 turns of 1mm enamelled wire on a 22mm form giving 1.9uH is about right for a 5th order 40m butterworth lowpass filter. By playing with the number of turns and the diameter of the winding you can get the inductance you require. And by using some clear nail polish or some kind of glue you can keep the windings neat and tidy, giving greater Q. And when it comes to filters, its all about the Q of the inductors.

Its just physically larger than a T50-6, so if space on the board is not a problem or the mechanics of an air core is not a problem, wind away and you will be fine. There is nothing magical about using Iron or Ferrite cores.

(Attachment Link) e

Board space really isn't an issue. But the stray capacitance and shielding might be an issue. Any thought on how to reduce them?

Really at HF you have so much wriggle room I dont think it matters all that much. You will have some added capacitance in the coil meaning you need less capacitance in each node to achieve the same level of filtering. On LPF it does not matter at all, I put the corner frequency on a 40m LPF at say 7.3mhz when designing the filter and the added capacitance will take it up to 3.4mhz where I want it. For BPF I use a trimmer cap of say 100pf in each node to be able to tune the filter where desired and account for my inductors not being perfect.

Shielding i would not worry about, but offsetting the air cores at 90 Deg to each other to reduce coupling should be done as a matter of principle. Oh and there are some really good air core calculator tools online that will help you design air cores to suit your needs based on the size wire you can access and what forms you have available etc. Takes the guess work out of the job.

[Dom13c]

Another option is winding your inductors on tap washers.

Thought of it but I lack an inductance meter. So measuring the actual inductance value would be a PITA for me

[Dom13c]

I have added 2 pictures, this is 9 turns of 1mm enamelled wire on a 22mm form giving 1.9uH is about right for a 5th order 40m butterworth lowpass filter. By playing with the number of turns and the diameter of the winding you can get the inductance you require. And by using some clear nail polish or some kind of glue you can keep the windings neat and tidy, giving greater Q. And when it comes to filters, its all about the Q of the inductors.

Its just physically larger than a T50-6, so if space on the board is not a problem or the mechanics of an air core is not a problem, wind away and you will be fine. There is nothing magical about using Iron or Ferrite cores.

(Attachment Link) e

Board space really isn't an issue. But the stray capacitance and shielding might be an issue. Any thought on how to reduce them?

Really at HF you have so much wriggle room I dont think it matters all that much. You will have some added capacitance in the coil meaning you need less capacitance in each node to achieve the same level of filtering. On LPF it does not matter at all, I put the corner frequency on a 40m LPF at say 7.3mhz when designing the filter and the added capacitance will take it up to 3.4mhz where I want it. For BPF I use a trimmer cap of say 100pf in each node to be able to tune the filter where desired and account for my inductors not being perfect.

Shielding i would not worry about, but offsetting the air cores at 90 Deg to each other to reduce coupling should be done as a matter of principle. Oh and there are some really good air core calculator tools online that will help you design air cores to suit your needs based on the size wire you can access and what forms you have available etc. Takes the guess work out of the job.

Let's see how far I can go with that approach. 

[BravoV]

Air cores are not self shielding like torroids and other forms so you have to play the usual games with orientation, spacing, and shielding to prevent coupling between them.

Regarding shielding, does air core toroid better ?

[David Hess]

Air cores are not self shielding like torroids and other forms so you have to play the usual games with orientation, spacing, and shielding to prevent coupling between them.

Regarding shielding, does air core toroid better ?

No, it makes it worse because now it is sensitive along its entire plane.  The magnetic core with its higher permeability than air directs the magnetic flux through the core excluding the area outside of the core.

[vk4ffab]

I have added 2 pictures, this is 9 turns of 1mm enamelled wire on a 22mm form giving 1.9uH is about right for a 5th order 40m butterworth lowpass filter. By playing with the number of turns and the diameter of the winding you can get the inductance you require. And by using some clear nail polish or some kind of glue you can keep the windings neat and tidy, giving greater Q. And when it comes to filters, its all about the Q of the inductors.

Its just physically larger than a T50-6, so if space on the board is not a problem or the mechanics of an air core is not a problem, wind away and you will be fine. There is nothing magical about using Iron or Ferrite cores.

(Attachment Link) e

Board space really isn't an issue. But the stray capacitance and shielding might be an issue. Any thought on how to reduce them?

Really at HF you have so much wriggle room I dont think it matters all that much. You will have some added capacitance in the coil meaning you need less capacitance in each node to achieve the same level of filtering. On LPF it does not matter at all, I put the corner frequency on a 40m LPF at say 7.3mhz when designing the filter and the added capacitance will take it up to 3.4mhz where I want it. For BPF I use a trimmer cap of say 100pf in each node to be able to tune the filter where desired and account for my inductors not being perfect.

Shielding i would not worry about, but offsetting the air cores at 90 Deg to each other to reduce coupling should be done as a matter of principle. Oh and there are some really good air core calculator tools online that will help you design air cores to suit your needs based on the size wire you can access and what forms you have available etc. Takes the guess work out of the job.

Let's see how far I can go with that approach. 

Get a cheap lcr meter as soon ss you can, makes life so much easier. You could probably make one with an arduino, im sure i have seen such projects online before.

[Dom13c]

I have added 2 pictures, this is 9 turns of 1mm enamelled wire on a 22mm form giving 1.9uH is about right for a 5th order 40m butterworth lowpass filter. By playing with the number of turns and the diameter of the winding you can get the inductance you require. And by using some clear nail polish or some kind of glue you can keep the windings neat and tidy, giving greater Q. And when it comes to filters, its all about the Q of the inductors.

Its just physically larger than a T50-6, so if space on the board is not a problem or the mechanics of an air core is not a problem, wind away and you will be fine. There is nothing magical about using Iron or Ferrite cores.

(Attachment Link) e

Board space really isn't an issue. But the stray capacitance and shielding might be an issue. Any thought on how to reduce them?

Really at HF you have so much wriggle room I dont think it matters all that much. You will have some added capacitance in the coil meaning you need less capacitance in each node to achieve the same level of filtering. On LPF it does not matter at all, I put the corner frequency on a 40m LPF at say 7.3mhz when designing the filter and the added capacitance will take it up to 3.4mhz where I want it. For BPF I use a trimmer cap of say 100pf in each node to be able to tune the filter where desired and account for my inductors not being perfect.

Shielding i would not worry about, but offsetting the air cores at 90 Deg to each other to reduce coupling should be done as a matter of principle. Oh and there are some really good air core calculator tools online that will help you design air cores to suit your needs based on the size wire you can access and what forms you have available etc. Takes the guess work out of the job.

Let's see how far I can go with that approach. 

Get a cheap lcr meter as soon ss you can, makes life so much easier. You could probably make one with an arduino, im sure i have seen such projects online before.

I will indeed 

[Georg - PY5ZSE]

sorry for bumping up the old thread...

There is a Chinese manufacturer of iron powder core toroids, TangDa. They sell via AliExpress, one link would be:
https://www.aliexpress.com/item/32963410519.html?spm=a2g0x.12010612.8148356.2.412a1d45HCy9c7. Shipping is free or at least cheap.

I bought toroids once from China a couple of years ago, but found out they are not the same quality as the original ones from Micrometals / Amidon. The losses are higher, the Q of the coils I wound was between half and 2/3 of what was to be expected according to Micrometals' data sheets. In the rather narrow Bandpass filters I was building I could not use them for that reason, in a low pass or similar low Q application there should be no problem.

[JohnG]

Air cores are not self shielding like torroids and other forms so you have to play the usual games with orientation, spacing, and shielding to prevent coupling between them.

Regarding shielding, does air core toroid better ?

Absolutely, if properly wound.

In a previous job, I studied this with both FEM analysis and experiment, with good agreement between them. Unfortunately, I no longer have access to that data, but the effect is not small. The external magnetic field of a properly wound air core toroid is much less that that of a solenoid, both in magnitude and distance from inductor center. The comparison was done on the basis of inductors of equal value and equal Q-factor. If I recall correctly, the toriodal air core inductor ended up somewhat larger than the solenoid, but the effective size of the solenoid was much larger. What I mean by effective size is the volume of space in which the energy is stored in the magnetic field. This effective size is really what determines how close you can put anything conductive close to the inductor without affecting the inductor.

What does properly wound mean? Just like the picture you show. A single layer winding with non-overlapping turns, with the gap between the ends being small. One caution is that the gap between the ends will dominate the self-capacitance of the inductor, and this will be worse than the solenoid. This is the main tradeoff, in that if the gap becomes much bigger than the turn spacing, it will increase the effective size of the magnetic field, but if it is small, it increases the capacitance and decreases the self-resonant frequency of the inductor. If using this for RF, do not overlap the ends for this reason.

Cheers,
John