An Ugly POC

This isn't about some skin condition or other annoying medical problem.  No, POC means proof of concept in this context.  I've always had great interest in the VHF/UHF+ bands, but keeping long-boom stacked yagis in good shape takes more climbing and tower time than I'm willing to invest as a geezer.  During the Christmas holiday season, another outlet for experimentation in those bands surfaced:  Meshtastic.

Metastatic piqued my interest and curiosity for a number of reasons:

1.  The devices are essentially small micro-controller boards that also contain a radio chip. These boards aren't very expensive, almost always < $50, <$100 with a nice case.

 


2.  In the US, these devices operate in the 33cm ham band (900 MHz), but can also be used by unlicensed folks, provided certain EIRP restrictions are observed. 

3.  The firmware for these boards and client software for Android phones, Linux machines, and that other OS are all open-source. 

4.  There's been rapid growth in "the mesh" around my area.  It reminds me of the explosion of 2m/70cm repeaters that seemed to start in the 70's. 

5.  Meshtastic appeals to my inner prepper (as a possible ad-hoc communication network when the internet is down or has been deliberately been taken down).  These days, I wouldn't assume that those possibilities are remote. 

So during the holidays, I managed to grab a few "nodes" (boards) and an alleged 5.8dbi gain vertical antenna.  That went on a side mount, about 20 feet up my old TV tower.  Unfortunately, I literally live in the shadow of a 250' (largely) T-mobile cell tower.  There's also a TV broadcast tower about 2 miles away.  I looked at the spectral space around 900 MHz with a RTL-SDR dongle (and GQRX).  The broadband noise level wasn't overwhelmingly high, but noticeable.  Some seemingly random trash also was present as shifting peaks in the FFT display.  Since my site could be best described as BFE, it looked like I'd need a front-end band-pass filter to get as much receive sensitivity as possible.  One additional catch is that my base node is a 1 watt device, so the filter has to survive that and have minimal insertion loss.  Next question is build or buy...

At 900 MHz, you have a few options for band-pass filters: 

1. Small helical resonators (Toko or Temwell) can be found on Ebay for reasonable cost.  They appear to be 1 watt tolerant.  I bought a couple to try, but deemed their insertion loss (~2.5db, at best) too high and did not pursue this option further.

2. SAW or ceramic filters are available and are generally in the $20 (approximate) range.  Most will tolerate +20dbm (100 mW) and some may be OK between +20 and +30 dbm.  These looked like they could do the job, especially for the 20 - 22 dbm boards. 

3. Commercial cavity, comb line or interdigital filters.  These have great pass band shapes and can handle power levels well above the limit for unlicensed operation at 900 Mhz.  They are, however, quite expensive, even surplus.  The surplus stuff mostly is duplexers and some single cavities.   A few far-eastern companies produce cavity filters (4 pole or so) in the $60 - $100 range (about 5x - 10x less expensive than surplus).  Availability looks OK, but shipping to the US is sketchy, at best.   

4. Build one.  (No one is coming, you are on your own.)  I looked a quite a few options, here, but the most easily duplicated (without a milling machine) design looked to be a comb line filter.  

So off I went down the comb line filter road.  Along the way, Paul Wade's (W1GHZ) Altoid tin design caught my eye. (https://w1ghz.org/filter/Altoids_Tin_Filters.pdf  -  please read it!)  A design for 33 cm is included in this article, so I made an attempt to duplicate that simple 3-resonator design.  That's the ugly proof of concept.

I pretty much followed Paul's design with one exception.  In the original design, each resonator is tuned by sliding the center conductor in and out of resonators made from 0.141" semi rigid coax.   That tuning method seemed like it might require 3 hands, so I substituted three 1pf to 10pf ceramic trimmers (hamfest find - Johanson ceramic puppies that now list for $20 on ebay).  Behold the Ugly:


 Inter-resonator spacing is 17 mm and the feed taps are 11 mm from the closed end of each resonator.  (There's actually room for 5 resonators in an Altoids tin, so I might try that in the future.) 

 

 

In order to get an idea of how this POC worked, I first caibrated my VNA over the 850 - 1050 MHz range, including two short SMA patch cords.  (Yes, I know a NanoVNA's alleged upper limit is 900 MHz, but this gambit seemed to work out.).  I swept this filter over that range and compared its behavior to the responses I got from a SAW filter (Data-Alliance) and unicorn cavity filter (CallBoost).  Filter pass band and SWR are shown below, Ugly POC, SAW, and cavity filter, top to botttom.

 
 
   
Filter -3db BW (MHz) -6db BW (MHz) IL (db) SWR
W1GHZ Altoids Tin Combline 40 56 0.61 <=1.5
915 SAW 34 39 1.33 <=1.2
Callboost Cavity Filter 33 34 0.31 <=1.3

 

As you'd expect, the cavity filter is great: steep roll-off outside the pass-band, low insertion loss and SWR, no issues with reasonable power levels.  The SAW filter's pass band shape and SWR are acceptable, but its insertion loss is a bit on the high side for my taste.  The POC comb line filter, while not as sharp as the other two, does exhibit acceptable SWR and low insertion loss.  Power handling at the 1 watt level should be OK. Given the noise level I'm currently seeing, it would likely work well enough for broadband noise reduction.  Adding an extra two resonators would undoubtedly improve the pass band shape.

I hope you've figured out that this post isn't about Meshtastic or the minutia of band pass filter design for 900 Mhz.  It's about choosing among different options in solving a common RF engineering problem.  It's about balancing performance against hardware cost and availability.  It's easy enough, sometimes, just to throw money at a problem, but other factors can derail that approach.  Some solutions simply are not available to the RF experimenter because of exorbitant cost (US made filters in this case) or poor availability (tariffs add cost and shipping from the far East can be slow, unreliable, and sporadic).  Sometimes, building your own is the optimal way to go.  Also be sure that you exhaustively search the web for prior art - in this case, I found pretty much exactly what I was looking for thank to W1GHZ.  Don't be afraid to dig deep - you usually won't find good home brew info at the top of conventional search results.  Of course you could ask one of the AI chatbots for help, but I would compare the results from at least three different models and against published data like Paul's article.  If you can successfully learn how to build your own "stuff", then you are more likely to have your projects succeed in spite of today's logistical and political challenges.




 

 

 

 

 

Comments

Post a Comment

Popular posts from this blog

Playing with Phasing - part 1

Image
      Every so often it seems that some folks are lured into the less-populated territory of SSB reception and transmission using the phasing method.  Maybe it's the promise of simpler (?) circuitry, eliminating the need for crystal/mechanical filters, or perhaps as a gateway drug into SDR-land.   Rick Campbell's (KK7B) articles describing a single-signal direct conversion receiver (QST - Jan 1993), followed by a phasing transmitter (QST - April 1993) are seminal works that detail the implementation of the phasing method with modern (for the 1990's, at least) circuitry.  These papers also discuss design tradeoffs in terms that are easily understood.  Quite a few years ago, I actually built a R2/T2 transceiver, shamelessly copying those designs and adding a Si5351 quadrature LO along the lines of that described by ZL2CTM in his blog.  It worked OK, but the opamp-based, all-pass phase shifters required hand matching of the RC components fo...
Read more

A Half-baked Fix for Si5351 Quadrature Issues

Image
In a previous post, I briefly described issues I had preserving quadrature output from an Si5351 board.  Allegedly, once a 5351 is programmed to provide a fixed phase offset on a second output,  it should maintain that phase shift across frequency excursions, provided the "even divisor" remained unchanged - No PLL reset needed until the a large enough frequency excursion requires a change in the "even divisor".  PLL resets caused annoying "pops" in the output of direct conversion front ends that I typically use for phasing-type receivers.  I was loathe to include a PLL reset on every frequency change because of that.  So, I (and many others) only do a PLL reset when the required integer "even divisor" was changed.  Many fewer "pops", but semi-random loss of LO quadrature within a relatively small frequency range.  I added a polled front-panel switch to my hardware that I use to manually reset the  PLL when my "ear detector" de...
Read more

Revisiting the LT5517

Image
Some time ago (just before Linear Tech & Analog Devices merged), I requested a few samples of LT's 5517 quadrature demodulator and 5598 quadrature modulator.   They happily supplied a few samples of each, but the local LT rep also called me at work and asked if I'd also like to have a dc678a evaluation board ( https://www.analog.com/en/resources/evaluation-hardware-and-software/evaluation-boards-kits/dc678a.html#eb-overview ).  Of course, I said, "Hell yes!" and it appeared in the mail soon after (no charge).  But then it sat unloved and unused for years until I recently decided to see how well it worked at HF. The 5517 board is the green one in the upper left corner.  A 2-channel baseband preamp is the Manhattan-style board next to it.  The board below the 5517 board is a split supply boost converter that powers a 100x gain, 2-channel baseband preamp using THAT1510 instrumentation amps.  These devices are very similar to the widely used INA217....
Read more