It’s been a few days with no updates on the blog but I’ve been VERY active in my downstairs lab. Unfortunately, my Bitx20 is still having some problems.
I have gotten a wealth of information from the bitx20 Yahoo group however, some from the group itself and some from emails exchanged privately. I’d certainly recommend it to others who are up against some problems with their builds. Through the group, I discovered a bitx builder, Leonard KC0WOX, who recorded very detailed videos on his build of the Hendricks kit and posted on his website. It helped me confirm that at least my voltages, currents and powers were approximately the same as his and it was just interesting to see him pull things up on the spectrum analyzer as he made adjustments.
I’ve also been speaking with Ron, WD4GWK, and Arv, K7HKL, and some of their observations and comments have led me to go back and really make sure my toroids were wound properly and to double check some things. I believe all of the analysis is leading me toward a few conclusions.
One, I have oscillations with higher drive levels, possibly (probably?) because of some trouble not reducing the carrier enough in the balanced modulator circuit. Whether that’s because of trouble with the diodes or a leaky capacitor or what, I’m not sure. I re-wound the trifilar quite tightly and that didn’t do anything to help. The oscillations will only allow me about 9 W output as I’ve said previously, and really much above 7 W I get pretty strong carrier levels. I’m not sure how many bitx builders have the same sort of problem and just keep their drive levels low, it’s hard to say. My impression from these guys is a lot of builders run into these problems and either chuck the rig up on a shelf and forget about it or sell it on e-bay! But that’s not my goal here!
Two, I don’t have enough audio from my un-amplified mic to drive the radio to more than a fraction of a watt. Apparently there’s a fair number of users of this rig who are running amplified mics of various types. Others have made modifications to the amplifier circuit in the bitx20 for more gain — this is the approach I’m going to try. The typical mics I have laying around and I suspect many others do will only put out maybe 10-15 mW peak to peak. That’s simply not enough to drive the rig, you need more like 40-50 mW with the amplifier as it appears in the Hendricks schematic.
There are also differences between the original Farhan circuit and the Hendricks one, and (I guess) there are advantages of one vs. the other in certain instances. I’m not in a position to try to compare or contrast the two but it might be useful to see whether the Farhan one might help with my carrier balance for instance.
So I think I can solve the audio problem fairly soon, either tonight or tomorrow probably when I get more shack time. The balanced modulator/oscillation issue is a bit tougher. I hope that getting more audio will allow me to at least put the rig on the air though (at reduced power) and make one or two contacts even if the signal-to-carrier ratio isn’t optimum.
The guys have given me a few things to read and so I’m going to try to digest some more of that info as well in the next few days. In about a week or so, I’ll be diving into my move to a new QTH, which means my shack will get boxed up. It’s a bummer for the hobby but good news for me in general.
There are lots of things I’ve learned from the investigations of the transmit problems in the last few days though. For instance, I didn’t realize the electrolytic between the mic amp and the balanced modulator is fairly “leaky”, in that DC values are kind of bleeding through to the modulator. Also, the audio frequency energy from the amp gets attenuated fairly substantially across all frequencies as they pass through this capacitor! I’m thinking I may want to use a different capacitor or maybe combination of capacitors for this coupling element to the modulator.
Also, I think I posted about low frequency audio being attenuated more than higher frequency audio. This is a problem that can come about when your modulated signal is too far removed from the IF filter skirt. For USB, that means it’s too far from the lower frequency end of the skirt. So your higher audio frequencies are still able to make it into the IF filter passband but your lower ones are attenuated a lot. In my case, I still have non-negligible carrier energy too, so if I tune my BFO to move the lower frequency audio signals into the IF passband, I start getting some of the carrier energy passing through as well, which is certainly not desirable. The solution is to either have a steeper filter skirt or better yet no carrier present! I think it is impossible to have “no” carrier at all with this design. But it seems with my board for some reason I’ve got too much carrier. And to get the carrier to be attenuated by the 40-50 dB below signal level that is common, I end up shifting the entire audio modulated waveform too far below the edge of the IF filter skirt.
I don’t know for sure if getting better carrier balance will solve the oscillation problem but I’m going to be looking into that. I’ll post updates on here and also to the Yahoo group as things progress.
A couple of other items of note. Doug Hendricks from qrpkits sent me the parts I needed, free of charge, for the main tuning knob polyvaricon! That was very nice since I don’t know if they were even supposed to be included. It was just a screw and a teflon style sleeve but it was great to have the proper thread and size (and saved me a trip to the hardware store, bumbling around for my own version of it).
A less happy situation lies with the digital display for the kit, the KD1JV digital dial. I assembled this last week but it won’t work. The digits light up but things seem scrambled and are not responding properly to the button pushes. Of course I re-re-checked my solder connections, which is hard with surface mount stuff like this particular kit, and couldn’t find anything amiss. I feel that I’m just out the $40 for this part of the kit as there’s nothing really “fixable” to malfunctioning chips. The kit is also about 6 years old so way beyond any reasonable request for a refund or swap for another one. I may make some inquiries to KD1JV to see if there’s some kind of special reset of something but I’m not holding up much hope.
All for now. Hopefully some happy “bitx-20 on air!” updates soon to come.