WK4DS Amateur Radio Blog

Search Posts

David Saylors 8/26/26 David Saylors 8/26/26

My 17 meter band pass filter and POTA…

Today I met up with some other hams that routinely go out together and activate POTA parks and was able to fellowship with them for a while before and after I made some contacts. There was (Jonathan Frye AA6H, Allen Padgett KN4FKS and Danny Wooten AG4DW) and they were running what appeared to be FT8 and SSB stations today. Allen and Jonathan were operating SSB the whole time and made quite a few contacts themselves. It was kind of neat to plug in the nanoVNA and see one of the other stations transmitting and messing with my antenna reading and when they stopped I could see the graph again! HAHA!

I built this band pass filter after having a couple of activations with my friend in Florida, Chas (NA2B). We were is such close proximity that if I didn’t have a bandpass filter for the band I was working, I simply couldn’t hear. The sBitx is notorious for front end overload and when the antennas are just a few yards apart, you are going to get front end overload! Well, on the bands that I had bandpass filters for, I had no issues, it was only the bands I lacked filters that I had a problem. It is pretty easy to find bandpass filters for the non-WARC bands, but WARC bandpass filters are kinda rare.

Well, I never let that stop me before…haha. A while back I went down the rabbit hole of wanting to make my own bandpass filters and I figured I would start with the 17 meter band since it is harder to find. I now own a spectrum analyzer with a tracking generator so I can sweep the filter in real time to see what tuning does to the filter shape. That is kinda awesome in its own right, but you could easily do the same thing with a nanoVNA too.

So here is the current iteration of the 17m bandpass module I built. IT sure looks like I home-brewed it, doesn’t it? HAHA! That big inductor It is a classic Pi design with an inductor and capacitor in parallel to ground at the front and back and a series inductor and capacitor in the middle. The input capacitor is underneath that massive 5.31uH inductor in the middle.. I taped it to keep the coils from moving around while I tuned it and then put the ziptie on it to hold the coils in place once I got it like I wanted it. I pulled the values for this filter from the internet off of one of the many website calculators that are out there…I don’t remember which one though… It has been a minute…haha.

Adding this filter literally stopped all the 20 and 40 meter signals that I had from my neighbors today. It isn’t even all that great of a filter, it does have decent rolloff, but it isnt some knife edge vertical skirt filter. I grabbed a quick photo of the plot on my nanoVNA with the marker on the 20 meter band. I have a little more insertion loss (mine is 1.3dB as compared to a commercial unit at about .65dB), but it works almost as good as one that would cost 200$ and I have about 10$ in it. The filter was even better on the passband before I put it in the chassis. Since I used open air coils, they are interacting capacitivly with the aluminium body to ground. This added capacitance increased the passband insertion loss about .6dB, but it is so clean, I chose to just leave it as is and use it for now. The 20 meter band being over 40dB down compared to the 17 meter band, I was able to setup right next to Danny, who was on 20 meters at the time, without him coming into my radio at all from what I could tell. All the lower bands past 20 meters are lower than that too, so all of them are pushed out of the way so to speak. The higher bands are interesting to me. Before I mounted the unit in the aluminium chassis, it fell off on the high end too. Once mounted it looks like it is about 35dB on 10 meters and about 20dB on 15 meters, which is still good. Another problem that I encountered wtih the added capacitance is that it moved the passband up some. I had it to where the high side of the passband was right at the end of the 17m band before I mounted it. This gave me more attenuation on 15 meters, by probably 6dB or so, but you use what you have and move on with life… This would have also raised the 20 meter rejection some as well, by probably about 6dB or so.

This band pass filter project looks really simple.. The aluminum box, the two BNC connectors, some 16ga magnet wire, three capacitors and a piece of pc board material for a build plane and I am good to go. I lost the lid to the aluminum box hence no cover, but it appears to work just fine like this so I used it. It turns out that if you use toroid inductor cores, that this is a lot easeier to build as the toroids are LESS affected by the chassis ground plane. Notice I said less… they will still change values, just not as much as air wound inductors like I used here. I deliberately chose air wound inductors due to their high Q, which is what gives me the sharpness to the skirts you see in the plot below. I have it marked with “IN” and “OUT” but I honestly think it doesn’t matter which way it is inserted into the feedline as long as it is there. Another benefit of adding a passband filter is that is works both ways, it keeps out stray signals from strong nearby sources and it also only lets the part of your signal LEAVE that will fit through the passband. This makes a huge difference in your signal purity as you are adding a filter to the output of your radio.

nanovna plot 17 meter band pass filter ham radio

You can see the antenna right behind Danny’s (AG4DW) truck looking out the windshield of my truck today and I moved as far away from the other people as I could muster without driving in the yard… This area is really small to be fair. It just doesnt have a lot of room to setup too many stations. In reality I was probaby the last one that could fit without too much trouble being caused by other radios. Now that doesnt mean this park is overloaded with three stations. Not at all! There are 5 other locations to set up radios that I can think of right off hand with out even trying hard so we could setup several sites at this park very easily. Now, that turns it into something else thoguh, that might be enough activity to trigger the need for a permitted activity, but I dont know. I normally just hand out wtih these guys or go by myself for the most part so it is pretty low key like that.

Those band modules in the above photo are a precursor to what I did today. I opted to stay on the higher bands no matter what. I knew I could land the activation with FT8 and CW, but I had no idea just how hard it would be…haha. This poses a good challenge though as I know I can easily land the minimum ten QSOs to get the activation if I just start on 20 meters but lately the bands have been up and down and have been real inconsistent on the higher bands so to get my ten would take a bit more work on my part. Challenge accepted!

So, natuarally I started on 17 meters…haha. I actually planned to work my way up to ten meters, hence all those band modules you saw earlier. I found W1AW operating CW on 17 meters and busted his pileup to get in their log! That was my first QSO of the day too, I knew right then that it was going to be a good day. As you can see in the photo above, the station was setup in the truck cab to keep me out of the sun as much as possible. I also was using the Begali Traveler key and the HamGadgets Pico keyer as well. Also if you will notice, I dont have the common mode choke installed today. I was testing to see if getting the SWR below 2:1 made using it irrelevant or not. I had really good SWR all day at the radio so I dont think I need this thing adding losses to the system when I was having a good showing on the radio. (It is the thing laying next to the radio in the background.) Do you use one every time you go out or is it something you dont worry too much about? I ran this one through the nanoVNA and it said it had from -.06dB to -.19dB of loss just inserting it in the line. Look at the chart below to see what it looked like across the HF bands for this common mode choke on the nanoVNA.

160m: -.06dB

80m: -.09dB

60m: -.10dB

40m: -.12dB

30m: -.13dB

20m: -.15dB

17m: -.16dB

15m: -.16dB

12m: -.17dB

10m: -.19dB

While this might not sound like much loss (all of them being less than 2/10ths of a single dB), if the bands are in bad shape and you are trying to hear stations near the noise floor, this can make a difference. I routinely listen along the noise floor with headphones for stations that are really weak. Today was one of those days. The bands were really bad for the most part. CW was not being used much because of this. I did get a good run on FT8 on 17 and then 15 meters, but the CW portions were all in the tank.

tentec scout 555 hf trasnceiver pota cw ssb

I started on 17 meters as mentioned earlier, I started by hunting a little and found W1AW was making contacts so I worked them first thing. It only took me three or four trys to get them in the log so I was pretty stoked about the day at that. But then it took me 50 minutes to land the 5th QSO in the log! That is super slow for my normal operations. So I opted to switch to FT8 so I could get the activation locked in before I went off on the higher bands playing…

I worked five FT8 contacts in a little under 20 minutes just using hunt and pounce. This had my ten in the log so now the rest are “gravy” at this point. Next thing I did was re-tune the antenna to 15 meters and since the sBitx was already running, I hunted on FT8 there for a while. I netted 6 QSOs in something like 14 minutes. This is actually a pretty good rate for me on this radio so I was happy with that. After landing those in the log, I set the Scout 555 back up and called CQ on 15m CW for a while… I had zero replies… It was like the CW portion was simply turned off. Seriously, where did all the CW ops go? FT8 was booming! I saw a ton of DX on FT8, even though I couldn’t work them, but they were there. Anyway, after calling CQ for like 15 minutes with no replies I figured I would hop on 10 meters and listen for my beacon and then see if I could get at least one QSO there before calling it quits for the day.

It took me about 50 minutes of calling CQ into the void to finally land a contact between 15 and 10 meters. The last FT8 QSO in the log and the one on 10 meters CW are literally 50 minutes apart! That is a lot of calling CQ with no responses. I actually was listening to the radio while talking to Jonathan and just letting the keyer do its thing. When suddenly there was a reply out of the ether! KE2FUA up in New York heard my pitiful pleas for a call and felt sorry for me. Giving me the QSO on 10 meters I desperately wanted. With that in the log, I called QRT and broke down the station. Afterwards I was able to spend some time with the other operators on site and chat with them a little. This is the main reason I love activating with these guys, the camaraderie is incredible.

cloudland canyon state park nickajack trailhead

This view shows the tight conditions pretty good. The pavilion in the back has the SSB setup and the white truck had the other station with me kinda in the middle. As long as we used band pass filters or there was enough band space between us, I didnt have a problem. While I was on 10 meters I didn’t have a filter for that band, so I was open to interference and didn’t hear anything from the other stations. I know that at this time, one was on 17 meters and the other was on 20 meters. Band pass filters are a life saver as well as a little coordination between stations. When I arrived, the others were already setup and operating so I went over to each one and gathered some intel about how it was going for them and their current band so I could pick one that they were not currently on. They were both running band pass filters so I was able to start literally anywhere I wanted except 20 and 40 meters. I love the higher bands so that was when I cooked up the idea to do just the higher bands to see what I could net with that sort of restriction. It also allowed me to use my newly minted 17m band pass filter too…

begalit traveler cw key hamgadgets memory keyer

I cant say enough good things about this little keyer. The stark contrast between the cost of each item is not lost on me… I like things the work well and serve me without too much fuss. Both of these tools have done that well. I have finally used the keyer long enough to need to replace the battery in it. Which is a simple task, but does require a screw driver… I thought I would hate having a battery powered keyer, but to be honest, it is kinda nice being able to put it literally anywhere and it just works. I printed the operation manual and keep it in the case with the Scout 555 so if I need to do something in the menu that I dont remember how to do, I can just look it up right quick.

At the end of the activation, I had to break down early and go run some errands and I did itjust in time too as this was what greeted me when I pulled out of the parking lot onto the main road. Sometimes you just get the right amount of things to happen in the right order I guess…

You can help support this channel by using these Amazon Affiliate Links as well:

QRP/Portable Radios:

Antennas & Tuning:

CW Equipment:

Power & Accessories:

Organization & Transport:

BONUS ITEMS (Optional 16-20):

WK4DS - David

David Saylors 2/9/26 David Saylors 2/9/26

Ten-Tec Scout 555 60m Band Module: IF Filter Redesign & Final Testing [Part 4]

Today saw me finish the TenTec Scout 555 60 meter Band Module Project in the workshop, now it is on to phase 2, testing in the field… I had wanted to rework the filter in the IF stage as it was not great from before. I was able to get a signal out of the radio, but the filter shape left something to be desired…

This is the 4th episode in my Ten-Tec Scout 555 60-meter band module series:

- Part 1: Initial Conversion and Filter Design

- Part 2: Crystal Selection and Mixer Circuits

- Part 3: Field Testing and Troubleshooting

- Part 4: IF Filter Redesign (this post)

The first chart image is of my filter after I got the radio to work. This is not what a bandpass filter should look like. The lower photo is of a factory TenTec VERY narrow bandwidth 40 meter bandpass filter. Turns out, this is almost impossible to replicate with discrete components that are not adjustable…as you will see. They used two tunable inductors to achieve this filter shape. I almost went to the trouble of adding some trimmer capacitors to mine, but as you will soon see, I felt it wasn’t needed. I soon learned what a filter Q is and why it matters when your trying to build a filter. I used some online calculators to get the component sizes for the filters and after messing around with a couple, I found one of the calculators had came up with numbers that were actually able to be made in my shop.

Armed with this information I decide to breadboard this filter first so I could tinker with it before soldering it to the band module circuit board. This, as it turns out, is a terrible idea if you plan to simply transfer the parts to the board and solder it all together. There is SO much stray capacitance and inductance at RF frequencies with a breadboard that you can build up a circuit, but when it comes to making the final item on perf board or Manhattan style, that you WILL use different values.

Looks good doesn’t it? Well look at the plot below! It actually looked great! It did have some insertion loss that I was not super happy about, but that plot looks great! Yes, the one inductor is hand made, I didn’t have one small enough in my little kits to work here I so wound one for the job. Notice how long the leads are on those components, that will come into play very soon…

I may have gotten a little aggressive with the passband width… I think it was set to 200khz or maybe 500khz, but it was not much. I think the values I put into the online calculator were 5.4mhz center frequency, 500khz bandwidth and 3 LC stages in a “T” configuration. The bottom line on the chart is -70dB which is unbelievable! This thing was incredible! So I get the 60 meter module out and take it apart and strip out the old filter I had previously built wholesale leaving a clean slate for the new filter…

After I had the old filter gone and cleaned up a little, I cut me a pad as the new filter layout and the TenTec layout are different so I wanted to not have to permanently alter the circuit board other then removing the old components. This pad became the central connection point for the three stages. A little hot glue and I was in business! I simply tinned the whole top of the pad so I could land part leads where ever I wanted on it. This worked really well…much better than the filter I am about to build as it would turn out.

SIDE NOTE ABOUT MODIFYING SCOUT BAND MODULES:

What I am pointing out in these two photos is a home made Via that I added. You see, when you remove the canned tunable inductors from the board, turns out you break the grounds. TenTec decided that since these things were going to be here anyway, might as well use them for something. So they apparently decided to make them connect one ground plane to another in a couple of places on the board. Once you remove the cans, you lose a critical ground path…or two. I took a finger drill and a very small drill bit and drilled a hole from one trace to the ground plane on the opposite side(I am confident these are only two layer boards), then I soldered a bonding jumper wire in the hole so both planes would once again be connected again and all the stuff would continue to work properly. You also get to see the old homebrew filter in the below photo as a bonus.

Below is the result of me simply attempting to move the parts over to the circuit board and soldering them in place… What a mess. I started with the exact parts you saw on the breadboard, but when I connected the nanoVNA to the filter I found the center frequency had moved up 500khz! It was now pushing my desired frequency out of the passband!!! I was seeing something like 16dB of loss at the desired frequency.

This is the world of ultra high Q passband filters made with inductors and capacitors I guess. Once I removed all the stray capacitance from the breadboarded system, all the capacitor values went down enough to make the filter not usable. That is why you see the mess below, I didn’t have room for the trimmer caps I have on hand so it was good ole trial and error method instead to get the passband in the right spot. To be fair, this mess actually worked, it just had something like 12 or 13dB of insertion loss in the filter which made the output power go way down. I used the amp meter to check it and the current draw on transmit with this filter was 1.8 amps and the current draw with the factory 40 meter band module was 9.3 amps. So I took a break and grabbed some food and went back to the drawing board…

I am going to admit something here that might make some people unhappy with me…I decided to use Claude AI to see what it would come up with for a filter design. I gave it all the parameters I had used earlier and (since it is a chatbot after all) had a conversation about what my goal was. I decided to move the bandwidth out to 1mhz giving me a decent shot at making a working filter without adjustable parts and using only what I had available on site.

This AI model came up with a “T” filter but instead of the center 2 LC components being in parallel, they were in series and so I built it. It tested really bad as far as rejection goes as it only attenuated about 8 or 10dB across the whole band and looked more like a poorly designed high pass filter instead. This did give me some inspiration though. So I went back to the online calculators that are there to figure part values for you and changed my approach. I went to this website and just made a low pass filter and then a high pass filter and coupled them together with an inductor. Why did I choose an inductor? Because all the capacitors I tried kept increasing the losses in the filter output is why. As it turns out, if you attenuate the IF signal in the radio the output power get lowered and the radio goes “deaf”…

Below is what I came away with finally. Yes, only 3 inductors and two capacitors.

Below is the filter shape for this simple filter design too. It is a lot better than what I started with and I was able to build it with parts that I had available to me at the time. Could the passband be narrower? Yeah, it could, but I am not going to throw too much shade at this filter. This is what ham radio is all about in my book, experimentation. The marker is at 5.64mhz which is 250khz above the highest place I can transmit so I know this filter wont attenuate my signal and the radio will work as it should.

The original values for the filter ended up being wrong, but I simply did what I did in the previous filter build and just used it as a jumping off point and started switching out parts one at a time with different values. This tactic coupled with looking at the nanoVNA to see what the new part did to the filter shape gave me the results you see. At first I didn’t write down what I had done, but it quickly became apparent I needed to note what I had done and so things progressed much more quickly once I did.

The above photo ( bottom of the second page ) is what I started with on this part and the lower photo is what I ultimately ended up with for the actual parts I used to build the filter. You can see where I was trying all sorts of stuff to get a working filter, these are not the only sketches either…

This is what I ended up with ultimately to get the filter shape above. The impedances are wrong, I am sure as I didn’t take them into consideration at all, but the radio is working correctly now (from what I can tell) and transmits almost at full power. Shoot, they could be right and I just got them right by accident, I really don’t know to be honest.

This is the output power draw now for the 60 meter module on the same radio as the 40 meter module previously compared. I can live with that! I checked it when I got into the shack at the house and it is a little over 35 watts into a 1.3:1 SWR load. That will work just fine for me. (Note: The current reading was into a 50 ohm dummy load at the workbench so I don’t know the exact current draw in the shack.) This radio will produce more output power too, I have it dialed back on purpose to about 40 watts for the 20 meter band module that I use with this radio.

This is what the final filter looked like when I was finished building it. This looks a lot better than that mess I started out with above. Not overly complicated and generates a respectable filter shape. The radio sounds good too. So I am happy with this.

Side note about the above photo, I had noticed in my original filter design that the orientation of the inductors made a difference in the filter profile on the nanoVNA so I started marking them with a paint pen to remember the orientation. This filter didn’t seem to care too much about that for some reason but the original cared ALOT.

The side view photo shows something else the I have not mentioned in these write-ups. The filter has space constraints. The filter can not be taller than that metal post or the other circuit board with hit the parts and the parts can not go past the post either as that is the space for the output low pass filter from the other board. So parts count matters…

It feels oddly satisfying to see the finished module and to use it in the shack. All I did last night was listen to some stations having a QSO and how well it could hear them. I also dialed up the webSDR again and listened to myself calling CQ for a bit with no takers on a different channel. I know the radio is working as the listening station in Utah could hear me…

Tonight I hope to find a chance to jump into a QSO with someone and see how it sounds to them.

I hope this has inspired you to do something with your gear whether it be build a kit or modify a device to do something new or different or even to repair something you have that is not working now…shoot maybe you will learn something you didn’t know before! That is what has happened to me on this journey, I have learned a lot about how filters work and what affects them in use. I am starting to understand impedance matching the stages as well as insertion loss. A lot of things I never understood before are now becoming more clear. I also have a much clearer understanding about how the TenTec Scout 555 radio works as well…

Thanks for following along on this 4-part journey! If you build your own 60-meter module for the Scout 555, I'd love to hear about it. 73!

You can help support this website by using these Amazon Affiliate Links:

QRP/Portable Radios:

Antennas & Tuning:

CW Equipment:

Power & Accessories:

Organization & Transport:

BONUS ITEMS

WK4DS - David