Page 5 of 8

Re: HIGH PASS FILTER FOR OPTIMUM HF RECEPTION

Posted: Tue Feb 12, 2019 8:36 am
by glovisol
FILTER CHEBYSHEV HIGH PASS N=9 - 50/50 Ohm - UNBALANCED

From prototype to finished filter


The filter is built on a one side PC board used as a soldering reference ground plane for minimising parasitic inductances. All components are assembled on this board prior to mounting inside the die cast enclosure. I was worried about parasitic coupling among the coils, due to close proximity, but filter attenuation @ 1.5 MHz is very similar to that of the original prototype (#6) so this problem did not materialise. Another worry was what would happen with closing the lid, but no detrimental IN/OUT coupling was detected.

Coil formers were made with 20 mm dia. PVC electrical conduit pipe and no detrimental effects on coil Q was detected. Once the filter is mounted inside the enclosure and tested, it is advisable to glue the base of each coil former to the PC board with bi-component epoxy resin.

Later today antenna attenuation of LW and MW band will be tested on the RSPduo and reported.

Re: HIGH PASS FILTER FOR OPTIMUM HF RECEPTION

Posted: Tue Feb 12, 2019 10:09 am
by glovisol
FILTER CHEBYSHEV HIGH PASS N=9 - 50/50 Ohm - UNBALANCED

Practical results

Re: HIGH PASS FILTER FOR OPTIMUM HF RECEPTION

Posted: Tue Feb 12, 2019 4:45 pm
by glovisol
FILTER CHEBYSHEV HIGH PASS N=9 - 50/50 Ohm - UNBALANCED

Practical results


NOTE: For the measurements of previous post and of this one there is a 550 to 50 Ohm toroid transformer between Beverage antenna & filter's input.

Re: HIGH PASS FILTER FOR OPTIMUM HF RECEPTION

Posted: Wed Feb 13, 2019 3:31 pm
by glovisol
FILTER CHEBYSHEV HIGH PASS N=9 - 50/50 Ohm - UNBALANCED

Filter performance near cutoff with Network Analyser- 1

Re: HIGH PASS FILTER FOR OPTIMUM HF RECEPTION

Posted: Thu Feb 14, 2019 7:47 am
by glovisol
FILTER CHEBYSHEV HIGH PASS N=9 - 50/50 Ohm - UNBALANCED

Filter performance 2 to 10 & 2 to 29 MHz with Network Analyser- 2

Re: HIGH PASS FILTER FOR OPTIMUM HF RECEPTION

Posted: Thu Feb 14, 2019 10:27 pm
by mark.
Responding to the original post of the thread..
For the one scenario I didn't see, 500 ohm longwire and 50 ohm radio.

Would I build a 500 in/500 out filter with a 500/50 transformer on the output of the filter or a 50/50 filter with a 500/50 transformer on the input of the filter? Or can it be done through filter values and no transformers?

Re: HIGH PASS FILTER FOR OPTIMUM HF RECEPTION

Posted: Fri Feb 15, 2019 7:46 am
by glovisol
Hi Mark,

In the last posts I have finalised construction and performance measurements for the 50 Ohm to 50 Ohm filter. This filter is useful for connecting a 50 Ohm antenna feedline to a 50 Ohm input of the RSP receiver. If you wish you can place a 10:1 low noise transformer at the filter's input (560 to 50 Ohm) and connect through the filter a long wire antenna to the 50 Ohm input of the RSP receiver.

Two days ago I have started the construction of the Chebyshev filter N=9, r=1.5 dB, ZIN/ZOUT=0.6 - PROTOTYPE #7: this filter had been designed for connecting directly a 600 Ohm long wire antenna to the 1 KOhm Hi Z balanced input of the RSP2 or the RSPduo. Unfortunately I have found that the coils developed for the prototype (prototype #7) are too large and do not fit in the 1590 BS Hammond die cast enclosure I have purchased. Furthermore I have found, measuring prototype #7 with the Network Analyser, that SWR becomes too high past 4 MHz and this problem is caused by the 0.6:1 ratio between input and output. For this reason I have re-designed the filter for a 1:1 ratio, eg. 560 to 560 Ohm. This filter will connect to the long wire antenna directly and will have at the output a 1:2 ratio wideband, low noise transformer matching the 1 KOhm HI Z input of the RSP receiver. By changing this transformer, e.g. using a 10:1 ratio, 560 to 50 Ohm, it will be possible to couple a long wire antenna to a 50 Ohm input.


For the practical realisation of this filter I need smaller coils. I tested iron powder toroid coils made with type 1 material, e.g. Micrometals T50-1, but Q is too low: Q<120 @ 2 Mhz. So I have developed a new layout with air core coils (Q=200) and am presently building the filter which will fit into the box. I expect to be ready for measurements and do them over the weekend. I also inted to write a post dedicated to the practical realisation of low noise wideband toroid transformers together with Network Analyser measurement data.

Kind regards,

glovisol

Re: HIGH PASS FILTER FOR OPTIMUM HF RECEPTION

Posted: Fri Feb 15, 2019 12:39 pm
by illllm
can you please let me know what it is that you used to generate the signal so you cold test your filter?

Re: HIGH PASS FILTER FOR OPTIMUM HF RECEPTION

Posted: Fri Feb 15, 2019 2:40 pm
by glovisol
Hi Illlm,

I use a wideband noise generator sold on ebay for 15 USD approx. You can find details & pictures of the measurement setup here:

https://www.sdrplay.com/community/viewt ... f=5&t=3943

and here:

https://www.sdrplay.com/community/viewt ... f=5&t=3352

Remeber to use a supply voltage of maximum 10 V with the noise generator. Higher voltages are not safe as I discovered to my own cost.

glovisol

Re: HIGH PASS FILTER FOR OPTIMUM HF RECEPTION

Posted: Fri Feb 15, 2019 2:53 pm
by glovisol
HIGH IMPEDANCE FILTER CHEBYSHEV N=9 / 0.1 dB RIPPLE / 560 to 560 OHM

Development and construction of this filter with a new layout & winding coil method went much quicker than anticipated and I am able to upload some measurements taken on the finished filter assembly, before mounting into the die cast box. Filter performs well and offers a SWR below 1.5 up to 15 MHz and a SWR below 3 up to 29 MHz. All details in the next posts.