Roger (KG4WBI) and I have basically been going to the Dalton GA hamfest for 2 decades at this point and this year was incredible as far as attendance and the sheer volume of dealers in the “boneyard”.
So as per our usual modus operandi, we decide to see how many Ten Tec radios are available and this trip we also have some other items we are in search of as well.
I personally was looking for the following:
A high power dummy load
A low pass filter
A ten meter radio that has the CW mode
A oscilloscope that works up to 150mhz
Of all these items, I found the first three! Score!!!
This blog post covers my ten meter “radio” part of the adventure.I found a President HR2600 10 meter radio for sale that worked perfectly and bought it for really reasonable money. The below photo is of the finished project.
The first thing I need to do is turn down the output power to 1 watt or maybe 2 as this radio is going to be a 10 meter beacon on 28.221mhz. So I search online and find out the control to adjust the output power in CW for this radio and then I open the radio to make the adjustment…
Once open, I am greeted with a cooked lowpass filter on the output. Seems one of the capacitors failed, caught fire and in the process, scorched the circuit board as well as burning the enamel off of two inductors… now what to do?
Simple, rebuild the filter section.
Another rudimentary search on google turned up a schematic of this radio. The three parts shown are what needed replacing in the filter but since the board what charred, I opted to build a separate filter module on perf-board and wire it down to the edges where the old parts connected. This way I could cut out all the carbon so it wouldn’t leak current.
It literally burned a hole in the circuit board! The above photo shows how bad it was once I had pulled all the components from the board in the area of the damage. That carbon is a problem too as carbon is conductive so if you dont remove the carbon or remove enough parts that it isnt in play then it will cause lots of problems. In the below photo you can see where I cut out the burnt board so it would not conduct. The ground trace broke due to this damage so I simply added a jumper wire across this area to tie it back together. I only did this in case there was a need for ampacity since the ground plane went all around this part of the board. This actually worked so now to get a working lowpass filter.
In the above photo I have marked all the spots where connections are required for the various detectors to work properly as well as the input and the output to the filter section. I ended up not connecting the detectors though since it is not receiving at all and is only transmitting a beacon signal on CW. I checked and it worked fine without these two connections so I simply left them out. Someone will find this radio in the future and think they have scored a treasure until half the detectors don’t function for some reason…lol.
Below shows a coil that had the enamel burned off of it that was sitting next to the capacitor…This part of the radio was technically functioning too (well, passing the RF at the least), since this part was simply a low pass filter after the final transistor, when it burned it simply shorted through and just kept transmitting, albeit without the filtering it was supposed to have, but it was functioning…
Once I had decided to replace the filter, I started simply remaking the bad parts from scratch and using existing parts that were good, trying to emulate the original design. This is where I ran into a problem. In my desire to simply repair it like I did, I ended up creating a lower frequency null that was completely shutting out the 10 meter band… It was a wide enough null that I couldn’t bend the coils and fix it either… (With coils this small, you can shift the null around a little by spreading the coils out on the inductors, but this was not enough to get it to work for me.) So I simply opted for what I should have done to start with, make a filter from scratch and tune it out of the radio with the nanoVNA before installing it…
Above is my attempt at replicating the old parts, I measured the core diameter and it was real close to 1/4” so I simply made new coils based on this and hoped for the best. (The lighter colored inductor on the left is original from the radio, I reused it since it wasn’t damaged.) To test my theory, I simply soldered a wire across the the input and output terminals of where the filter would go (see below photo), powered the unit on and it worked fine (the transmitter was still working as it should). This is what told me the tune on the filter was off frequency. When I put the nanoVNA on the filter it had a DEEEEEEP null right at 28mhz all the way down to 24mhz and up to 33mhz! Seems my replacement parts and their new configuration had lowered the low end of the filter to the point that I was choking off my own signal. Haha…
So here is the filter I scratch built. Side note here, the capacitors in my filter are all high voltage capacitors as the output voltage of the RF signal can be very high at times. It can easily exceed 100volts in some circumstances so I chose capacitors that were at a minimum of 500 volt rated and some are even more. Better safe than sorry in this regard. It has larger inductors which give me more room to adjust them. I used an online calculator to figure the beginning values. This worked great because I had a good starting point to work from. I then connected it to my nanoVNA and checked the loss on the S21 input to see the filters shape. Air core inductors are a funny thing, they generate a nonlinear pattern of suppression that has deep nulls and at some points almost no effect on signal. The photo above shows where I adjusted one coil a little to get the null to land on the 6 meter band and also this is where this 2nd order harmonic would also appear, if there was one (knowing the quality of the factory low pass filter, it probably has a harmonic…lol). This would suppress any harmonics that might show up and also it has decent suppression across the region as well. You can see in the photo below that I added some capacitors to the bottom as well to help with the cutoff and to improve performance of the filter overall.
I ran it through the nanoVNA to tune it so it would let me run the beacon. It was losing about 3.5dB of signal at 28.221mhz but that isn’t a problem for me since I only want 1 watt forward power anyway. I simply connected the radio to a dummy load through a good wattmeter and tuned the pot inside the radio till I had 1 watt on the wattmeter. That 2nd deep null is at the first harmonic which is the 6meter band so it is working like it should. The original filter worked about as good as this one from what the math says. Air core inductors are not great compared to toroidal inductors but they do work. The first null appeared when I added one of the capacitors on the bottom, I will take what ever I can get and this added some suppression to the filter so I left it. The calculator showed the starting point of my filter with the image below captured from the online calculator. The calculator I used took the size of the coil and the number of turns and the over all length into account to come up with these numbers.
I ended up changing the first cap to 33pF by adding another 100pF in series with the other two and I also added a 33pF across the center coil (as this was what was on the original coil and the VNA showed an improvement in filtering when I soldered it on so I left it), I am pretty sure this is where the second null showed up at and also generally improved the whole filter.
So all in all, this was actually a fun little project and I learned alot about filter construction and tuning at the same time. I have said it before and I will say it again, if you don’t have a nanoVNA, get one for this sort of stuff. It works really well and it is super cost effective compared to professional gear that does the same job.