Winding multi-pole LPF coils onto the same PVC pipe

For my **broken link removed** I want to build a 40m LPF using a PVC pipe as a coil former. Note this transmitter has an auto transformer to match higher impedance antennas (wires, no feedline).

A problem I had with a LPF was that I thought could only be designed for 50R I/O impedance. But see this design of mine attached. It uses lots of poles but one value of capacitors and coils only. The coils are exactly 6T on this coils former 32mm. Apart from the great harmonics attenuation, this LPF can work satisfactorily at any I/O impedance, as long as the input impedance is the same as the output. This means that I can plug even higher impedance antennas to it and also tap it in the higher impedance tap on the transformer and the filter will still operate fine. See the diagrams, I provide for each impedance, two photos, one for the pass frequency and one for the second harmonic.

What do you think of it?

I am also thinking of building this onto the same PVC pipe instead of splitting the coils. Now I know it is better for the coils not to couple energy between them, but if they are far apart, even on the same pipe, it may not matter. I do not know how far apart though.
Also, I am thinking of winding near-by coils in different winding phase (direction). This should cancel the coupling between them, even if they are wound on the same pipe. Shouldn't it?

Out of band rejection is not the only characteristic of a low-pass filter.
The pass-band ripple of your low-pass filter in 50/50 ohms is about 1db, and is rising gradually from 2dB in 100/100 ohms, to 17dB in 800/800 ohms, and up to 27dB in 2.4k/2.4k.
Those high ripples are unacceptable values for most of the low-pass filters in the world.

vfone said:

Out of band rejection is not the only characteristic of a low-pass filter.
The pass-band ripple of your low-pass filter in 50/50 ohms is about 1db, and is rising gradually from 2dB in 100/100 ohms, to 17dB in 800/800 ohms, and up to 27dB in 2.4k/2.4k.
Those high ripples are unacceptable values for most of the low-pass filters in the world.

Click to expand...

You are right. But this transmitter is only operating on one frequency 7.030MHz. And the attenuation at this one is minimum at all impedance values.
I am interested mostly in the way the coils can be wound onto the same form. Would it be better for the adjacent coils to be wound in opposite phase so as one does not affect the other and could possibly be placed nearby?

Apart from the great harmonics attenuation, this LPF can work satisfactorily at any I/O impedance, as long as the input impedance is the same as the output.

Click to expand...

Depends on your definition of satisfactory operation. By nature of passive LC filters, the filter characteristic is valid for the designed source and load impedance and considerably different for others.

Also, I am thinking of winding near-by coils in different winding phase (direction). This should cancel the coupling between them, even if they are wound on the same pipe. Shouldn't it?

Click to expand...

No. Mutual coupling of filter inductors will kill the stop band attenuation, no matter if it has positive or negative sign. Placing adjacent inductors 90 degree rotated is the best way to reduce coupling.