Topics

Easy Bitx Bandpass filter. How wide is normal?


RCC WB5YYM
 

Disclaimer!!!    I have no idea what I'm doing, but it sure is fun anyway.

Does anyone have a clue as to the bandwidth of the bandpass filter on a Bitx40 or Easy Bitx?  My breadboard version looks to be about 700khz, with primitive testing. Here is a youtube video.

Homebrew Bandpass Filter.

While waiting on a bandpass filter from QRP-Labs, I thought I would take a shot at building a filter using the information I found in the files section of this group. I bread boarded the circuit up using t50-2 toroids, several trimmer capacitors, and a few fixed capacitors. Not having a clue as to what I was doing, and not sure how to tune the circuit with the gear I have, I put a 50 ohm resistor on the output, and my antenna analyzer on the input tuned to the middle of the 40 meter band, and adjusted the filter, by removing turns from the transformers and adjusting the trimmer capacitors, for 50 ohms and resonance in the center of the 40 meter band. I patched it into the receiver so I could take it in and out quickly and found the receiver was much better with the homebrew bandpass filter in.  Just for the sake of curiosity (see disclaimer above). I hooked the scope to the output and manually swept the frequency from above the 400 meter band, and surprisingly found  that the filter had some pretty sharp cutoff at about 200khz above and below the 40 meter band. I tried to make a video, switching views between the analyzer which I was using to generate the 7mhz signal and the scope.  I'm not sure if this blind hawg found an acorn or not, but it sure brought the receiver to life. 


Dean Souleles
 

RCC - 

Nice job testing - using what you have!  Characterizing filter performance by sweeping the frequencies across the pass band is exactly the right approach. As you have learned bandpass filters (and low pass filters) for QRP are not too difficult to build if you follow some basic principles.  I know you have ordered a QRP-Labs kit and that will be great to compare.  Also - they provide all the construction details and values for all HF bands in the instructions for the BPFS - so if you have the toroids and caps you can build as many of them as you want using the same technique.

One thing you might want to look into is the NanoVNA - it is a great tool for analyzing filters - and lets you see the pass band and filter shape right on the display.  I went through a similar learning curve - testing in almost the same way you did - with a signal generator, scope and the filter terminated in 50 ohms.  I dumped all the values into a spreadsheet and made a graph - a couple of hours testing and fiddling.   I'm glad I did that because I learned a ton! But the NanoVNA can do all of that in seconds.

If you can pickup a used copy of Solid State Design for the Radio Amateur, by Doug Demaw - there is a great section on filters and table of values you can refer to also.  Demaw also wrote some great QST articles on the topic including this one:

The double-tuned Circuit: An experimenter's tutorial (robkalmeijer.nl)

Fine business OM - you know more than you let on,

73 - Dean
KK4DAS 


RCC WB5YYM
 

I was wondering if the Nano VNA would graph it out. I ordered a signal generator which arrived yesterday DOA, and has already been sent back. I need to take a look at the Nano Vna. Thanks.


Evan Hand
 

RCC,

Yes, the NanoVNA is a great tool for checking/displaying filter characteristics.  It will measure both the input and output characteristics, both of which are important in circuit designs.

I ordered mine on Amazon and was very pleased.  I have read of some that were not so lucky, so be sure to check on the number of stars and reviews if you buy from Amazon.  I have not tried eBay or any of the other discount sites for this type of equipment unless it was a recommendation of the original designer.

There are also multiple versions of the NanoVNA.  I am not up to speed on the differences, as I have two of the original designs, not the H or one of the newer versions.  You might want to check out the nanovna-users group here:
https://groups.io/g/nanovna-users

Have fun.
73
Evan
AC9TU


Dean Souleles
 

To be 100% sure you get an authentic one, you may want to buy from one of the authorized sellers:

More info is on nanovna.com - but here is the link to the list of sellers:

Buy a NanoVNA | NanoVNA

I bought mine early from the Ali express seller.  The list now has a couple of authorized Amazon sellers.  And I just purchased the sister product - the TinySA spectrum analyzer from R&L electronics - if you are in the US they are reliable.

Dean
KK4DAS


Evan Hand
 

Dean,
Thank you for the links!

73
Evan
AC9TU


RCC WB5YYM
 

Thanks for the information. I have returned the signal generator (off brand) that I got from Amazon that was DOA. Will check out the links to the Nano VNA and see what I can find there. 

Thanks to all. This group never disappoints. 


RCC WB5YYM
 

Dean, I have become a Nano VNA fan. I'm glad that my first piece of test gear did not work out, and I could go this route at just over half the cost. What I found out is that with all the trimmers in place, I have very good adjustment for the frequency and shape. This is my final adjustment until I move it to the board. The initial measurement was a peak at 7.189 and slope down each side from there. I am thinking this shape may be better to cover the whole 40 meter band. Thank you for the suggestion on the VNA. It will be something I will use often. 


Dean Souleles
 

RCC - 

Excellent work - and these are great learning tools, aren't they?

It is a terrific tool and you are well on your way..... keep tweaking and see if you can get your passband a little flatter.  Ideally you would like a very flat passband so you have equal insertion loss across the band.  That dip in the middle means you are losing signal right in the middle of the band.  At this stage I found it very helpful to make little adjustments and take good notes.

The next tool you should look at is the TinySA - for $49 it is an an amazing little spectrum analyzer and signal generator.   It is designed by Hugen  - one of the NanoVNA designers and it is also a terrific and inexpensive tool.  It is great at characterizing quality of you transmitted signal and will show you if spurs or harmonics are leaking out.  Be aware that you will also need a set of attenuators though - before you hook it directly to your transceiver.

Congrats on your progress and thanks for the great pictures!

73,

Dean
KK4DAS



Dean Souleles
 

RCC -

Another hint.... switch your scale to LOGMAG to see the dB insertion loss. 

Back to your original question - - filter bandwidth is often referred to as the 3dB bandwidth - this is where the signal level will be 1/2 of the peak.  Switching to LOGMAG will let you see that easily.    An ideal filter would have 0db loss across the pass band and slope to the -3dB point - with steeper slop giving better cutoff.   In reality it is a tradeoff between all the parameters.  The NanoVNA lets you see what happens as you adjust the  L and C and see what has the most effect.

73,

Dean
KK4DAS


.    


RCC WB5YYM
 

Thanks for the feedback. It is really helping me learn. Here is a picture from this morning. I can get the insertion loss a little lower, but I lose bandwidth on the filter. I think I am going to add a couple turns of wire to the primary side of the transformers and see what happens.  I am all ears for any tips/suggestions.