Why I don’t use the sBitx radio for other modes anymore…

I have an interesting relationship with this radio…

When I first got this radio, I found it had some keying problems on CW but I just adapted to them and used it anyway. These problems are in how the software monitors the CW circuit and keys the radio and such. Turns out that as the radio heats up, the problem tends to get worse causing me to have to slow down the keyer to be able to send accurate code. Even with these tactics, I still send many mistakes towards the end of an activation due to the lag in the keyer.

The reason I currently dont use it for SSB is the audio is terrible and I need to trouble shoot that on the bench to see what is going on there. At first it had a weak microphone element so I made a preamp to correct for this but it still doesn’t seem to work all thatwell. Based on all this, I have not used SSB. (I do plan to revisit the SSB circuit to see if I can get it right)

As I also own several portable Ten Tec and Penntek radios that work wonderfully for SSB and CW, I threw in the towel and just started carrying one of those as well as the sBitx. It doesn’t take long to change out the radios since they are so small and light and I really like the break in on Ten Tec radios (the Penntek has wonderful break in too). Another huge plus is that the filtering on the Ten Tec radios is much better than the sbitx in my opinion. Now to be fair, the sBitx filtering is good, but the Ten Tec radios are just better. The Penntek TR-35 suffers the same problems that the sBitx suffers from. Strong, nearby stations will dull the receiver sensitivity.

What I have found that helps in the scenarios is to simply move. There is nothing an overloaded front end can do to help this problem. I do like the features that are available in CW on the sBitx though. Things like the memories and how easy they are to employ is awesome. Another thing is the waterfall is real nice that runs right alongside the CW decoder which has helped me a couple of times. I dont watch the decoder much, but it is nice to be able to see it match what I copied at times for call signs and such. Another great thing about the sBitx is the touchscreen. Not needing buttons is real nice. The only part of the experience that is honestly lacking in CW is the keying problem. Once someone solves this problem, this will be a game changer for a radio. I am still experimenting with different cooling solutions to keep the machine cooler to see if I can get the keyer to work better.

As you can see, this machine makes for a great FT8 field radio. The lack of a need for an external computer is a huge benefit when it comes to this. I just wished the dev team could sort of the CW keying problem this radio has that makes it less than ideal for this application at this time. One thing I plan to do soon is to set it up and see how running QRP power only helps alleviate the heat problem by not running 20 watts on 15 meters and building up the extra heat. This may solve most of my problems but the radio is designed for more transmitter power so I want to access that power.. Another option is to use a keyboard to send the code as it generates the code internally in software instead of polling the IO architecture to get the keying input. This is fun to me too but a nice key is also fun to use. So till the dev team does come up with a solve for the CW keying problem, I will relegate this radio to keyboard CW and FT8 (and maybe even SSB once I address the audio problem again)… But the main reason I wanted the radio to start with was the built in FT8 function that it has, so in that regard I have a great radio that I love to deploy!

If you are considering getting one of these radios, just be aware that it is far from perfect and it is literally made for experimentation. As long as you keep your expectations in reasonable bounds, this radio can be a lot of fun.

1 POTA, 2 Radios, 3 Modes and 4 Bands!

Today was a fun day…

The location is all too familiar, but I set out to do a sort of special operation today. I wanted to complete a POTA activation using two radios, 3 modes and 4 bands in one sitting. This doesn’t sound like a tall order but since I am operating inside the truck cab, this does involve a little logistics to not just have stuff piled everywhere while I am operating. As you will see in the photos below, I still ran into this problem somewhat, but it was manageable.

I setup at the disc golf parking lot again, but this time there were quite a few people there and some of them actually quizzed me about what I was doing. One gentleman, Jerry as noted in my log, actually chatted for a good bit about how he has been looking to get into ham radio but the Atlanta area is kinda tough on locating local hams for some reason. I gave him a card and told him to email me and I would give him as much info on it as I could find to get him on the right path…

As you can see in the photos, I set up ham sticks and this time I started on 40 meters FT8. I wanted to get the FT8 portion of my “sprint” if sorts out of the way first as the other two modes were going to be on the Ten Tec Scout 555. I had recently activated with it and it was acting up with chirping on CW and reports of RF noise on my audio on SSB. I attributed this to poor connection of the 20 meter band module as it was fine on 15 meters. Armed with this knowledge, before I left out for the park, I took the radio to the shop and use the Deoxit for gold contacts and a tooth brush to clean the contacts on the band modules. I even wetted one module pretty good and used it to “clean the contacts in the radio by plugging and unplgging the module a bunch of times.

This improvised procedure worked as I got good reports and the CW ran flawlessly as you will see later. Deoxit is magical stuff, if you dont have any, it is work your investment to grab a small can of it.

I really enjoy using this radio for my FT8 operating and with the upgraded finals and RF deck, it runs flawlessly now with SWRs in the 2:1 and even slightly higher range without problem. Today saw SWR levels on transmit of about 1.7:1 and it happily skipped right along making several contacts in a row at one point.

After completing 6 QSOs on FT8 I figured it was time to get the Ten Tec Scout 555 out of the case and see what I could scare up on CW.

The 40m Ham stick that I have must have a really high Q as it is very narrow banded. So I have it tuned for the CW portion of the band and the SWR in the SSB area can be quite bad at times. For this reason, I chose not to hunt any SSB contacts on 40 meters today. Once on CW (this was about 1/2 hour later as this is when I talked to Jerry) the band must have been closing or the band noise was getting so bad that I could not hear many stations. I was able to work Alabama and Tennessee before the call signs faded into the noise. That is something you will learn pretty quick about HF radio propagation, if you want to work closer in stations, use the lower bands, but if you want to reach the west coast from Georgia on a ham stick, use the higher bands like 15 and 20 meters. When 10 meters is open I have worked other continents with ease… Don’t discount those higher bands, they are truly magical.

In the above photo I have the CW key and keyer staged but not connected. The mouse is actually driving the FT8 machine and makes working FT8 so much easier to be honest.

In these two photos you can see what you have to work on constantly. The above photo shows the cable that came with my Ten Tec Scout 555 when I bought it. This is a common issue you will find with home shop made cables. These were stripped back way too far and as I used the cable, the conductors started breaking and I was beginning to worry about blowing fuses.

I happen to own a tool that is designed to remove these pins from the connector, so I am able to dismantle this connector properly. I dismantled it, then took it apart and cleaned it up, properly stripped and re-soldered the pins, then put heat shrink tubing on it all to insulated and protect it better. Now I am not so worried about it either shorting and melting the wire or blowing the fuses and shutting down the activation over something as simple as a cable…

Once I finished on 40 meters and had two modes in the bag on one band, I switched over to 15 meters CW to see what was happening there. There was a contest going on so it was a little crowded and I didn’t understand the exchange so I didn’t jump into the contest, but rather setup on a clear spot and calledCQ and worked a couple of stations there, one station of which is DX!!!

At this point. I got a phone call from KG4WBI about a completely unrelated matter, which we discussed and I told him to fire up his HF rig and we would see if ground wave would make the trip to his house from the park so I could get Georgia in the log as well as a 2nd mode on 15 meters! Well, it worked just fine and we had a great QSO on 15 meters SSB before he had to sign off and go run errands. So having confidence that SSB was going to work better I started calling CQ on SSB.

I got literally zero replies to my calls… so I went hunting instead. I found two more ops that could hear me and I was able to bag a couple more QSOs and these were Park to Park contacts to boot!

These three photo show me installing the band module into the Ten Tec Scout 555 transceiver. It really is that simple to change bands on this radio. Now to be fair, this is not as simple as just turning a knob, but it really isn’t that bad.

To remove the module you pull out the bottom of the little lever on the front of the module and it will pop out enough to be able to slip it out, then you grab a different one off of the pile and stick it back in the slot in the radio. Push it to seat it and then your ready to go.

At this point, I switched over to 20 meters and since the PTO (notice it is not a VFO) was still up in the SSB area, I decided to see if I could hunt some contacts with that mode. I landed one contact in SSB on 20 meters and was happy to get them in the log.

After working the one lonely SSB contact on 20 meters, I decided to give CW a try as well… Remember me mentioning a contest? Well, it was here too. It was going strong as well and because of this I was only able to work one contact on CW as well. If nothing else it netted me another band!

My last stop of the day was 17 meters. The 17 meter band is a WARC band and therefore it is off limits to contesting. Now, to be fair, POTA has been called contesting of sorts, but as of right now it is not considered a contest but is more in the spirit of something similar to a rare DX station activating on the same band and developing a huge pileup there. So I proceeded to hunt me an empty spot (18.078mhz) and started calling CQ.

This is when things literally took off! I netted a whole page of contacts in about a 1/2 hour span of time! What a day! 4th band in the books and I was stoked!!! 17 meters must have been where all the POTA ops had went due to the contest and I didn’t get the memo…lol. The Scout worked flawlessly after the cleaning and I cant be happier now with the old girl.

One of the great things about 17 meters is the propagation is really anybodys call. I worked Hungary at one point as well as California, Alaska and Idaho, then there is a ton of east coast stations too, it was everywhere today on 17 meters. Ham radio is so cool…

Before closing today I wanted to mention that Aaron KV9L and I have a youtube channel and we just hit 2000 subscribers (as of this writing)! If you are into ham radio and watching videos about it, then we would love for you to come over! I am currently doing a series of short form videos that are teaching CW one letter per day. There is no limit to the number ofd times you can watch them so I am hoping these become long term training aids for people.

Once I get the letters, numbers and punctuation done, I am going to start doing words next. After words will come sentences so we will see how it goes. Anyway, I just wanted to thank the 2000 people that made us as successful as we are!

73

David -WK4DS

POTA, HF Signals & Ten Tec

Today got so fast paced that I forgot how to do CW properly! Let me explain…

As you can see from the QSO map above, the bands were alive and well on this day from Raccoon Creek WMA (US-9875). The calls just kept coming in and I finally had to got QRT, with people still calling, and pack up as I had to meet the wife for a dinner date! I normally clear the little pile up I will draw in and then I can simply power down. I really felt bad about having to shut down with stations still calling me. I now know what those rare DX station feel like… It is surreal to experience that to be honest… especially from Alabama.

You see, today I activated US-9875 Raccoon Creek WMA and I went at a different time of day. I wanted to spend several hours here today as I dont get over to the area often and I wanted to try to get as many contacts as I could in the afternoon. My goal was 60 QSOs today which is way more than I usually worry about, but like I said, this park is a solid 40 minutes in one direction and I dont go this way too often so I wanted to “make it count” if you know what i mean.

I rolled into the WMA at about 14:30 local time and set out to build the radio down by the Tennessee river. This is a small parking lot at the end of one of the gravel roads into the WMA and a lot of locals come to this spot to fish in the river from the shore. Did I mention it started raining right after I got the antenna setup and I got into the truck? Well, let me tell you, the rain set in and it didnt stop the whole time. I even broke down the antenna and stuff out in the rain today!!!

The above photo shows what the station looks like right before I plug all that stuff in and turn the power on. Everything I need to get on the air is in that pile for both SSB as well as CW. To be fair, I can get by without the Picokeyer and the 1/8 mono to 1/8 mono cable and I could simple plug the paddle right into the radio and it would work just fine. So there is really excess here above what is absolutely needed.

I did get the nanoVNA out and look at the antenna to make sure the rain didn’t do something to it like push it out of band due to the conductivity of the wet earth or what ever. Turns out it was just fine… It was a little low with the water on everything but the null was really close to the bottom of the 15 meter band edge so I used it and didnt worry about it. If you will notice, it was only 1.233:1 SWR at the band edge so it was plenty happy and we never checked it again after this on 15 meters. When I switched to 20 meters, the SWR had moved up to about 1.8:1 or maybe it was at 2:1 but I ran it and as you can see in the log, that didn’t seem to matter at all.

I setup shop on 15 meters SSB for a change and wanted to see what I could get before 15 faded out, but I noticed almost immediately that the ALC light wasn’t coming on at all when I would transmit audio. This lamp should just blink on the energy peaks but it wasn’t coming on at all. I messed with the mic gain and nothing. I did notice that I could wiggle the cord at the plug and it would make all sorts of stuff happen. This pointed me to the connector being faulty somehow but workable to some degree.

I made a few contacts with it like this and basically yelling at the radio would barely get the ALC to illuminate. I asked the last person for a report and they told me the audio was muffled and weak. Armed with all this knowledge, I decided to open the connector and see if a wire was broken. Well, they were not broken…but they were also not connected either. There were two cold solder joints from the factory…figures. I got a pretty decent photo of the green wire in the below photo. Well…that shuts down the SSB portion of this activation so onto 15 meters CW!

The 15 meter band has two things that make it unique. It is quieter than the lower bands for me and the “skip distance” is MUCH further with mobile, or improvised antennas. I didn’t make many CW contacts on 15 but check out the locations! Washington state is pretty common for me on 15 meters, in fact the entirety of the western United States is easier for me to work on 15 meters during the day. But a lot of people apparently don’t know this so lots of times the band will be open but there wont be anyone on it. Case in point today, I worked 4 ops in 11 minutes of calling CQ.

After vacuuming the bands of CW ops, I moved to FT8 to see what was happening and this is my “yardstick” of how open the band is. Turns out it was REALLY open! 15 meters FT8 netted a nice dozen contacts and got me almost half way to my original goal of 60 QSOs! This didn’t take long either and just goes to show that the higher bands are great if you will just go check them.

I had switched radios to use the sBitx V4 (I did the software upgrade and got some new features like the color coding of FT8 information you see below) You can also see that I was transmitting with 17 watts too. This is not precise, but it is close and I use it for my logbook notes.

After finishing on 15 meters I moved down to 20 meters to finish out the day. The sun was starting to set at this point and I knew 20 meters would serve me better at this point. after hunting W9XT, I set up shop on a clear frequency and started calling CQ, this is when things got a little crazy. I had a nice little pileup form pretty quick that took me about half an hour to clear, then nothing… It must be band fading at this point as the stations just vanished. I figured I would switch over to FT8 for a few minutes at this point too and see what I could get in that mode. I didn’t mention that I used the sBitx for CW at this point and to be honest, I dont think I will use it for CW seriously till there is a software change to make it work better. If you try to use it much past 18WPM it is like it fights you and induces mistakes for some reason. The people smarter than me in the email reflector seem to think is has to do with how the software scans the inputs on the radio, but I do know it will induce mistakes in your sending if you are not very attentive and send a very specific way. It will work if your careful and go about 18WPM max. I can use a keyboard to send CW but that really isnt all that much fun to be honest.

I struggled with FT8 today on 20 meters for some reason. The QSO could not complete to the 73 so it would not autolog the QSO. It would also not finish the exchange many times either. I could get it to send the signal report then they would send back and they would just get stuck in that step and never move past it. It got really frustrating towards the end, but I did manage to net some contacts here too.

At this point I had made 48 contacts which is a phenomenal day out for me, but I wanted 12 more to have that 60 that I started out my goal with! So at this point I move back down into the CW portion but this time on the Ten Tec Scout 555 instead since the CW mode on it works beautifully.

It took me a couple minutes to switch the radios out and to find a clear frequency and get started, but once I did, it didn’t take long to get a HUGE PILE UP going!!!! I got so excited during this part that I was racing the clock to see how many I could work before having to shut down the radio and go home as I was almost out of time. I started rushing the closing and it was here that I got sloppy with my CW. You see, I normally use QRP power levels for the most part, just the sBitx and the Ten Tec Scout are QRO by the rules and the sBitx is borderline to me. I will usually turn the power down to 5 watts or so anyway to preserve the finals in the radio but today I had the Scout and it is set at 50 watts and is not easily field adjusted from that power level.

As the pile up raged on, I got so lost in the process that I started sending 72 intermingled with contacts getting 73 instead and finally someone stopped me and asked why I am signing both modes and this is when it hit me. I had been running on autopilot about half the time just logging contacts and the muscle memory would just send the 72 and I would not even think about it. I would then called QRZ and get another call, rinse and repeat… Here is the next problem for me. My ragchew copy and my POTA copy are very different levels. He sent me the message at the speed I was working POTA contacts. Your brain will learn how to copy the formatted style exchanges without even thinking about it. I know that I am going to get a two letter state after the signal report and I can copy callsigns really well for some reason up to about 25 WPM, but send me a full sentence question and it breaks my brain. I cant copy half of it without writing it down. I can ragchew comfortably at 18 WPM max at this time, but if you ask me a question about my radio or something that isn’t part of the usual banter in a POTA contact and I am lost… I apologized for this mistake I had been making for probably 20 minutes without even realizing it and it was really close to when I had to leave anyway so I went ahead and called QRT to get the radio put away as it was raining really well at this point too…

To everyone that reads this that I didn’t get to work on this day, I am sorry to have to had shut down the rig and leave, I was having a wonderful time and wanted to stay longer!!!

The moral of this part of the story is dont get in such a hurry that you cause confusion with the hunters, take your time and be sure in what you are actually sending out over the airwaves.

On a brighter note, I got my 60! and 14 more!!! I dont know how long it has been since I got that many calls in the log in one day. Until next time I hope to work you on the air!

73

WK4DS - David

Activating a park in bad conditions

POTA activating in bad conditions can be challenging… for some modes. Seems that FT8 doesn’t really care if the bands are crappy and will gleefully allow you to play radio anyway.

This was an experiment to see if the sBitx amateur radio could work in terrible space weather like you see below in the report.

The above and below photo show how fast the band conditions changed on me while activating on this day. The band noise could be seen on the display “walking” across the band segment at times. This band noise would pretty much wipe out what you could hear on the band, but the radio could still decode the FT8 signals so I was able to work almost 20 stations in about an hour even with this sort of noise coming into and out of the band.

I normally dont bother activating in bad band conditions, but this day I really wanted to see just how good FT8 was at extracting signals from the noise. I can honestly say that I was impressed by the performance of this mode. Digital modes have their place and this is very much one of those places. Weak signal in poor band conditions shows just how important it is to use all the modes available to the amateur radio operator. I was once a die hard CW only op and to be honest, it is still one of my favorite modes, but of late I have started getting into SSB more and now I am also dipping my toe into digital modes more and more and I am really liking it. Dont limit yourself out of a stigma like I did, you might just like what you find.

Below is a photo of the display that shows something that i had never seen before and that is the pulsing band noise. That was a new one for me. It too would come in waves and these would last much longer it seemed. The sBitx just kept chugging along though, I would sometimes pick calls and sometimes answer calls and this is what makes FT8 awesome. I couldn’t have possibly made SSB contacts in these conditions…

Here you can see the operating position for the day. Perfect blue skies and nice temps makes for a great day on the air at a park. I had to goto this spot as the lower lot was completely full since the weather was so nice, it was fall and a Saturday. Sometimes going to this spot has it’s perks like lower band noise from the campground inverters and such so I dont really mind it.

I had planned on using a battery and setting up on the picnic table I usually occupy down next to the canyon rim, but there just wasn’t anywhere to park and I didn’t want to wait on someone to leave just so I could park.

This radial is on it’s last leg. I have twisted this broken radial back together for months now and on top of that, the connector on the other end has also broken off. This is almost comical as this is the best performing radial in my set! That is why I am reluctant to repair it at this point…lol. It just works and if it ain’t broke (figuratively) then dont fix it.

I dont know when this bug (I think it is a jumping spider) decided to join in on the fun but here he was playing on the antenna mount. By the time I finished the activation, this little guy was long gone. Kinda fun to notice the little things like this when you are out in the park.

By the end of the activation, I had worked 18 unique contacts and garnered one dupe because I didn’t read my logbook very carefully and called one ham two times…then I decided to attempt to work some CW. This was pretty futile though as I only worked one contact out of about 15 minutes of calling as this is abysmal for me when I work CW. I can usually work about one call per minute even when I am running QRP power levels. So when I only work one…that is bad band conditions. The stations would literally just disappear then reappear and it was impossible to get complete calls so I called it quits after one.

All in all, it was still a great day in the field. Hope to work you soon!

WK4DS

POTA at US-0716 with only Digital Modes...

My plan was to only use FT8 and get the activation as quickly as possible, but in the end it ended up a little differently. This POTA park is a national military park commemorating the battle of Chickamauga during the US Civil War in the late 19th century. It is now a recreational hot spot along with a tourist destination so they have all the trappings of a outdoor recreational area along with the monuments about the war. This trail head was across the street from where I parked the truck today.

Another reason for the singular mode was that it would only require the use on a single radio this time since I am currently using the Ten Tec Scout along with the sBitx SDR for my POTA activities. I only deployed the sBitx SDR today so I could stay inside the truck and run the air conditioner as well since I didnt need to hear as well as I would if I were doing CW or SSB.

By only using one radio the rain that moved through the area didn’t hamper my activity at all. I was able to get setup quickly enough that I beat the rain and was on the air before it started to fall.

I like this location on sunny days since it is in the shade, but today it was overcast and rainy and this didn’t really matter. I did like it because it is “RF Quiet” and there is little to no RF noise except when certain cars drive by that are a little noisy.

I used my field expedient sealing technique again today, I know this is not a real seal but it will shed water long enough for me to complete my activation and then I can break it down and stow it in the truck easily enough. Long term would be a different story, but it works well for me like this for simple short time frames like a POTA activation.

This time since I was going to stay on 20 meters, I simply deployed the two radials that are associated with that band and I ran them at right angles just to see how it would perform, it looked great on the nanoVNA with the SWR running about 1.2:1 across the digital portion of the band.

I meant to get a photo of the waterfall and somehow ended up with an image of the transmit signal and associated info…figures. The point was going to be that the 20 meter section of FT8 was filled with ops today. So much so that is was hard to find a spot to send from. The band segment was that tightly packed today. I had to frequently move as I would go for several cycles with out so much as a hint of someone answering me only to find out that a strong station had setup on top of me and was sending over my little 20 watts. Get a contact, listen to the band a minute and find a new clear spot and repeat. This was the modus operandi for the day. I considered at one point of moving up to 15 meters just so I could more easily find band space, but talked myself out of it since I didn’t want to get out in the rain and change antennas and then worry with the tune of the radials and such.

After getting my minimum of ten contacts on FT8 thereby securing the activation, I wanted one more just so I could have a little buffer if one was a pirate of some such. Turned out that this was a huge problem as it took me a long time to get that last QSO in the log, I even went over to CW thinking it would be easier (which it was not) and finally got that last station… The QSB (signal fading in an out) was so bad that I could almost not hear strong stations that were normally easy to copy, but I did get the one I wanted before going QRT for the day. I grabbed the photo below to show the band fade and it is kinda visible in the signal on the lower edge of the waterfall, but it would go from what I would call S9 to nothing in literally 5 seconds, they signal would just vanish. I was blown away with how bad it was and then I checked the space weather and saw why, there is a ton of solar activity causing terrible band conditions right now…figures.

11 contacts in almost an hour is not a great hit rate for me where I am used to working about 1 CW contact a minute and a FT8 every two to 3 minutes, but it was fun none the less. Any day I can get out and do POTA is a good day. Until next week…

73

WK4DS

sBitx V2 Amateur Transceiver Mods for POTA Use

My newest radio, the HF Signals sBitx v2, is an HF radio with so many features that you just need to follow this link over to their website to see what it is capable of. Once you have finished going down that rabbit hole, come back over here to see what mods I have done to mine already for my POTA stuff. I want to say this right away, this radio is kind of in Beta, so if you don’t like to open up the radio and tinker with it in both the software as well as the hardware, this probably isn’t the radio for you. With that out of the way, let’s get started!

I decided to get in on this idea of a open source architecture radio design. I found this company, HFSignals, and they make several radios with their latest model being the sBitx V2. This is a touch screen radio with a huge screen and it is powered by a Raspberry Pi SBC. Using a legit computer to power the radio unlocks so many things that this radio can do that other radios can not do. For one, it will do FT8 IN THE RADIO! Yeah, no external computer needed at all, that in itself is a gamechanger and don’t be surprised if you see the big names in the industry doing this in the near future because of it. By using a Raspberry Pi SBC for the brains, this little machine is just chocked full of goodies that those other radios cant do.

FT8 is native inside the radio on the sBitx v2.

You heard that right, as I type this blog post it is on the bench next to me finishing a QSO with I1RJP, and when it does it will automatically put the QSO in my log for me. How awesome is that? Did I mention it is open source? Yeah, both the software AND the hardware are open source and it is encouraged to take your radio apart and tinker with it. Shoot, it even shipped to me with a spare set of output transistors in the accessory bag. These are well thought out but they are still rather simple overall designs with features like they are passively cooled radios and use a crystal filter network. Another thing about being passively cooled is that it means there is no fan noise to deal with at all this way, since there are no fans.

Well… I used it on a POTA activation and the little radio got HOT. Really hot to be honest. So I decided that since this radio is designed from the outset to be open source and to be tinkered with by the end user and I wanted to use it for POTA activations, that I would add some fans to cool the little machine on activations.

The first thing I wanted to tackle was the power amplifier heat sink. It is a great heat sink and does a wonderful job as it became really hot during the activation. I first started looking for a suitable fan in my junk box and found the perfect fan in an old computer power supply that I have cut apart for another project. The fan was still mounted in the sheet metal case which also happened to fit perfectly over the outer edges of the heatsink. All I needed to do was trim the sides down so it sat next to the heatsink and add some screw holes to attach it with.

Once I had it mounted with a couple of self threading screws, it was time to get it running. I went inside the radio and started looking for a suitable place to tap power out to it and found the incoming source point was the best, but the fan would run all the time if I used this spot…

Enter a simple electronic circuit that could be used for any temperature of fan control and could even be adjustable with a potentiometer if you were so inclined. This circuit is a simple power transistor rated for 6 amps of max draw (I used this so it could handle the 300mA of draw from the fan and not need a heat sink and they are still really cheap too). You simple have a voltage divider network for the base of the transistor where you have a fixed resistance between the base and ground and you add a thermistor (a temperature reactive resistor) between the voltage source and the base.

I chose a 10K ohm negative coefficient thermistor for my needs as this design has the resistance go down as the temperature goes up. The fan is simply wired in series with the transistor’s collector and the voltage source and the transistor is basically used as a electronic relay in the simplest form. I know the MOSFETS are more efficient, but this works and I had all of these parts (except the thermistor) on hand so I used these instead.

What happens during operation is that when powered up, the thermistor has so much resistance at a lower temperature that the base voltage is less than the .7 VDC required to bias the transistor since the fixed resistance of the base to emitter side of the voltage divider is calculated for the desired temperature. The thermistors have a chart showing the resistance at different temperatures so you can make these calculations fairly easily. Mine worked out to 270 ohms for the fixed resistance between the base and emitter. So you can imagine that with 10,000 ohms (at 77 degrees) on the other half of the voltage divider you only get .363 VDC on the base of the transistor and the transistor stays “OFF”…

I found the above chart online for 10k ohm thermistors and grabbed it for reference only. This may not be the right chart for your thermistor as they have different resistance curves so check with your brand of device and make sure you have the right chart for your device. Back to the story in progress…

Well knowing this, as per the chart above when the temp rises to about 107 degrees then the resistance drops in the thermistor to about 4980 ohms and this now allows about .7 VDC to develop across the base of the transistor, forward biasing it. Notice how I made the contact with the hear sink. I soldered on a pair of wires and put heat shrink tubing on them to insulate the connections then I ran a 1/4-28 tap down into the space between two fins cutting something resembling threads into the fins. Then I took a piece of a zip tie and put it on the thermistor and ran a setscrew down against the ziptie/thermistor stack to hold the thermistor against the heat sink it works really well too. The ziptie is only there to do two things, to prevent me from cracking the outer shell of the thermistor and to insulate the thermistor from the setscrew so the setscrew wont bleed off heat, I am not certain it helps, but it sounded good in my head at the time. LOL. I located it near the transistors so it will pick up the heat faster. I also changed the mica thermal spacers out for aluminum ceramic instead as those are supposed to conduct heat significantly better between the power transistors and the heatsink.

The fan control circuit can be seen on the perf board next to it in this photo.

Then once things heat up the fan spools up and blows on the heat sink, in turn cooling it down to the point where the temperature on the thermistor drops enough to raise the resistance back up and shut off the transistor again. This worked like a charm at the last activation I used the sBitx v2 on. The fan doesn’t even come on to drain the battery until the heat sink warms up enough to need it, so if your simply listening around or hunting activators and doing search and pounce, then the fan will not cycle on very often at all… if ever. I do understand that there is some current flowing through the resistor network all the time though as well, but it is low and I am not really worried about 3mAs of current anyway. I have really large batteries in the grand scheme of things and if I am going to hike with a radio it will be the Penntek TR-35 anyway…

The next “upgrade” I did to mine was to add a heat sink and fan module to my Raspberry Pi 4 SBC (single board computer) that is the heart of the sBitx radio. The original configuration has the SBC mounted right on the RF board with very little space between the two for airflow and no room at all for a fan of any kind as you can see above in the photo with the side radio cover removed. I did some experiments with positioning fans above the heat vent opening above the Pi and could feel that the air I was pulling out was quite warm. This led me to the assumption that the SBC would run smoother (and probably last longer) if it had a proper heat sink installed on it. You see as my unit would run, with time, it would start to lag a little from information input, like adjusting the VFO would result in the numbers jumping on the display instead of a smooth change as the knob is turned. I noticed when the fan was pulling air across it that this would be reduced significantly if not eliminated. So the mission became “how do I get a fan on this thing to cool it properly”

So I start looking at adding an external fan but ultimately that didn’t look really feasible as I wanted to pull the heat off of the Pi properly and not with just slight air movement. A case fan would only help a little and I am pretty sure at this point that the SBC needs a little more help than that. Enter a riser kit from amazon and a heat sink assembly with fans from one of those Raspberry Pi bundles that comes with the pi, a small case and a heat sink with fans. I measure every thing and come up with a height to elevate the SBC and check the cabinet and sure enough, there is plenty of room to add it. So I get it all together and when you get the riser kit from amazon, you get the little screw-in standoffs as well as the header socket extender to extend the pins up to the Pi once installed. If you will look closely in the photo below, you can see one of the fans and the heat sink fins for the SBC heat sink that I added to the Raspberry Pi SBC to help cool it.

In the above photo you can see the pin extender i also had to add so the Pi would plug in once elevated for the heat sink module. This worked wonderfully for fitment and I was stoked to have the heat sink on the SBC finally, but I still had to connect the fans to power to get them to run. Instead of putting them on a heat sensor, like the power amp fan, I chose to wire these straight to the main power switch on the sBitx so that when you turn it on, they spool up and run the whole time. My reasoning is that the computer will be working the whole time as the radio is in constant “refresh” mode so to speak as the sBitx software package has to keep everything up to date in real time on the display, then there is the background applications that are also running like the telnet server and dx cluster stuff and the other applications should they be active. Logically the SBC will be running nonstop, so lets just run the fans all the time…

This is where I run into a problem…

Did you notice where I drew power to run the fans? That seems completely reasonable for someone who just successfully wired up a fan on a temperature sensing circuit for the power amplifier. Well, it turns out that the fans for a Raspberry Pi are not 13.8VDC fans but rather 5VDC fans as they are designed to be plugged into the SBC IO header bus and not driven from a 13.8 VDC source. Let me tell you something, when you run a 5 volt fan on nearly 14 volts it sounds like a jet engine preparing for take off! On top of that, I had a pair of them!

I had brought the radio to the house and connected it to power to play with it and hunt some POTA activators and the whole time I am thinking to myself… “Man, these tiny little 25mm fans sure are loud…” but after about 10 minutes (yes, they lasted that long and in fact lasted much longer) I started to smell hot plastic… I shut down the radio and took it apart to find the fans incredibly hot and seriously, “soft” to the touch.

Back to the workshop and I figured out what I had done. So I start working on a way to lower the voltage that would not cause RF hash inside the radio cabinet. Those little buck converters from amazon are notorious for causing RF hash so that was out. This simplest thing to do was to put a big resistor in series with the fans. I had put them on a 5 VDC supply in the shop and measured the current draw so it was a simple matter to do a little ohms law and come up with about 50 ohms of resistance to mitigate the excess voltage from the supply. Also these fans pull very little current and I am running them at 4 VDC instead of the full 5 volts as they are rated to run from 3.3 to 5 VDC. This way they will be a little quieter too or at least that is my thinking and it gives me a little leeway should the voltage go down lower or even a little higher.

Inside this heatshrink is (4) 200 ohm 5 watt resistors soldered in parallel so make a 50 ohm 20 watt resistor.

Well the brute force approach worked and the (4) 200 ohm, 5 watt resistors in parallel dropped the voltage down to a very workable level. The resistors dissipate about 1/2 of a watt of heat total so I put them out of the way from everything else and it works, it just works. Maybe later I will add some sort of active voltage regulator and do it so that I don’t have to just burn off the excess power nonstop to keep the fans at the right speed. Maybe something like a 555 timer biasing a mosfet for pulse width modulation to keep the power draw down and keep fan speed at a constant level. I don’t know, I will look at that later, for now, the brute force resistor idea is working just fine.

Once I got the fans turned back on, I found “I had a rod knocking” in one of the fans… figures that if you run them at almost 3 times the rated voltage that something like a bearing fails in the fan… well I just ordered another set from amazon and swapped them a couple of days later and now we are all set. So it would seem at least as when I went to reassemble the radio I found that my USB and Ethernet ports no longer lined up with the openings in the side plate for the radio.

Turns out that if you have a fully equipped machine shop, this isn’t a real problem though. It could have been solved with a good file and probably a half hour of work, but who wants to do that if you have a CNC milling machine at your disposal? So I put it in the milling machine and manually open up the exiting holes to allow use of the ports and by default I also created air vents for the Pi to get fresh air to it much easier now. This also seems to work really well for the time being. Better air flow, adding a heat sink and FANS!!! The Pi runs cool now.

I also did a couple capacitor mods that I found on YouTube as well. A fellow on the interwebs had done these same capacitor mods and they made sense to me so I went ahead and added them as well. The following video describing these mods and why.

To summarize his video, he added decoupling capacitors to the three jacks on the side of the radio to drain off stray RF. He also adds one across the incoming supply lines to kill transients and possible RF on the power. Please note that in the photo below I have the plus side marked on the capacitor, this is incorrect as the board is marked erroneously and I translated that error to my cap as you can see in the photo. Check the supply on your radio with a meter before installing caps that can not be reverse biased. We all know how that would end…

One last thing I also did was to add some heatsinks (also per the video above) to the power devices inside the radio to help them survive use during digital modes. They worked just fine when I ran them with CW but when operating digital, the duty cycle is a lot higher so I didn’t want to risk thermal failure because of something as simple as adding a heat sink. This also comes from the above video and just made sense to me too, so I added them. The radio operates great and I really like using it for FT8 when calling CQ (more on this later) as well as for CW. I have some other mods planned for the future so another blog post describing these mods will be in order at some point. Till then, thanks for your time and attention and I hope this helps someone out there with their radio.

WK4DS

72