开云体育

ctrl + shift + ? for shortcuts
© 2025 Groups.io
  1. QRPLabs
  2. Topics

QCX T1 installation: A better way?

 

I began assembly of my 20 M version of the QCX nine days ago, and all went well thanks to the excellent manual provided by Hans. ?Until I got to the most difficult component - the transformer T1. ?I had no difficulty winding T1, but when it came to installing it on the populated board - fitting eight thin wires into the proper eight holes in the pcb, essentially simultaneously - that was a bit much for my 76 year old eyes and fingers! ?So I removed the three small windings from the toroid, and installed the toroid, with only the large winding, on the board, soldering both leads. ?I straightened the wires of the three removed windings and soldered one end of each length in the three outer holes for those windings. ?Then one by one I wound those three small windings on the toroid, starting over the top of the core, passing the wire under the core (with the aid of a tweezers) to the outside, and repeating the process until the three turns were completed. ?Then I inserted the free end of the wire through the inner hole for that winding. ?Verifying that the sense of each winding was correct, I tightened the wires and soldered the last three connections. ?This process went very smoothly, and I am pleased with the result. ?If you have difficulty with the installation procedure ?as described in the manual, you might want to try my method.

Of course, when it came time to align the QCX, the peaking capacitor was fully meshed and I had to add turns to the large winding! ?But that went well, and the alignment process proceeded smoothly. ?R24 ended up fully counterclockwise, but the image rejection is excellent, as is the sensitiviy. ?The rf power output was low, as many have experienced, about 1.0 - 1.2 W with a 12 V supply (increasing as the key is held down for a few seconds). ?My first contact (from Connecticut) was with OZ0JX in Denmark, with my 1 W and dipole up 20 ft! ?I subsequently raised the power output to 1.6 - 2.0 W by removing one turn from each of the LPF inductors. ?And I solved the turn-on issue with a diode (1N5711) between pins 7 and 20 of the ATmega328P. ?

Hans has done an incredible job with this project, great hardware design, software design, board layout, and the best construction ?manual ever done. ?And all at such a low cost! ?Plus he spends a lot of time in support of those of us encountering problems and having suggestions or questions about this remarkable little rig. ?What a fine contribution to the QRP world and the Ham Radio builder/maker community!

Ben - K0IKR
 

Ben nice to meet you and so many other great folks here, and Hans is amazing. Your idea of the T1 method is a very nice one. I made my leads not 3cm but much longer, so as to be able to pull it high in the sky above the board to see where to poke the next one, then pulled them all tight bit by bit, not easy. And it took me two goes to get it right, so your method sounds like a nice option.

I have been having such fun with the QCX, building it and now operating it. It has come at a time where it is a bit of a life saver too, as I needed distraction from some really awful events occurring here. Building it has kept my mind off things, and using it is just WOW. The RX is much quieter than my big rigs in terms of local noise. It is such a pleasure, even with the varying volume with signal strength, it is much more pleasurable, local noise just doesn't seem to impact. I think the choice of direct conversion RX with fantastic rejection of other side band, is just fantastic.

I had no problems to work Timor and USA shortly after going on air, and then I left it on WSPR and got reports from all around Europe through the night. Now I just had more than an hour rag chew with 599 both ways to my good friend Tim VK3IM on 30m, even though our distance would often be too short for 30m. Solid copy for over an hour, and great fun. The features and performance of this little rig are clearly what would be found in something at least 10 times the price. The fact it also allows for lots of additional construction and modifications fun, while already performing so well just all mounted as is on the PCB, is another big plus.

The QSK on CW is wonderful. Really wonderful. Some more CW characters need to be added in a future firmware update, and I'll volunteer to give a full list when the time comes. It's also fun that there is still small additional things that can be added in future updates, e.g. allowing menu editing input to be done from straight key not only paddle. Built in straight key works just great. Tonight rag chew with the old post office bug, great fun.

I drove around to the shops and back with it on the back seat with a 12V battery and a squid pole poking up through the side window, bottom 3 thick sections, helical wound, and got WSPR reports from around VK even like that without the high dipole. I'm looking forward to using the QCX out and about and on future travels, just as I enjoyed using an HW-8 back in the 80's from around France, England, Sudan etc. The size of this is amazing.

I had indeed forgotten, not in theory but in practice, how fun QRP is, and how a couple of watts gets you out so well. Local noise is the big killer of amateur radio these days, even more than contests, robots, unlimited/unspecified splits, and the thing that prevents many contacts as you cannot work what you cannot hear: but the QCX makes listening a lot more pleasurable in a fairly high local noise level as I just did not notice the noise AT ALL on the QCX. That's got to be worth 20dB.

Indeed what a fine contribution to the Ham Radio community, to reviving construction, and for many like myself, giving a first real taste and enjoyment inducting us into the world of construction, and also what a fine contribution to the Morse Code Revolution, with a lovely built in key and a high performance CW rig with WSPR built in.

I think soon we will have thousands of fun signals all having fun with QCX, I know several ham friends around Australia already building QCX and we will soon no doubt have much fun QCX to QCX. I don't even resent the fact that it has no chirp nor drift being a great fan of both, can't praise this rig highly enough. Thanks to all involved in its design, to the wonderful people in Japan who are doing such a fantastic job to sort, package and dispatch it with such high quality.? Please QSP our thanks!
w7aqk
 

开云体育

Ben and all,
?
It sounds like you and I are similarly afflicted!? My eyes are not very good either, one having macular degeneration to the point that everything is a blur with that eye.? Thus my vision tends to be 2 dimensional rather than 3 dimensional.? It’s weird!? I’m even close to your 76 year mark—two more weeks!
?
I agree that T1 is a bit daunting, but mostly it is just cumbersome with those 8 legs.? Your method makes some sense, but seemingly has some complications as well.? Admittedly I fumbled around a bit installing T1, but eventually I think I came around to a process that worked pretty well.? It’s a bit difficult to describe, but I’ll take a stab at it.
?
First, you need to make sure that you have the 4 connections on each side of the toroid nicely aligned as they come off the toroid.? I used WA4MZS’s drawing as a guide.? Each connection on the left has an opposite connection on the right, i.e. 7 and 2, 3 and 8, etc., but these leads are only “opposing” in how they are inserted into the board from top to bottom.? They are not the opposing ends of the same winding.? Then I make sure that each successive pair is somewhat shorter than the pair just above it.? 3 and 8 are shorter than 7 and 2, etc.? I work from the “top” inserting 7 and 2 first.? Then I go to 3 and 8, and since they are shorter, they aren’t crumpled too much due to the insertion of 7 and 2.? I “coax” 3 and 8 into their respective holes with needle nose pliers.? Then I go on to 5 and 4, and finally 6 and 1.? Working from “top to bottom” seemed easier than doing the reverse.? Once all 8 leads are inserted in their respective holes you can pull each lead from the bottom of the board to tighten everything up nicely.? Before soldering anything, I checked for proper continuity on each pair that actually opposes each other—1 and 2, 3 and 4, etc.? With 8 leads it isn’t difficult to get a couple of leads crossed, but the procedure I followed seemed to prevent that, or at least minimize the probability.?
?
I’m not suggesting that my method is necessarily better, but once I settled on this process, things seemed to go relatively smoothly.? I fumbled around a bit before coming to this method.? Clipping each successive pair shorter seemed to help a lot—you just don’t want to have the last pair be too short!? This also helps you properly identify each pair.? If all your leads are approximately the same length, it’s easy to get them confused.? I think my lack of “3D” vision made things a lot more difficult in my case, but when I went to dealing with just 2 wires at a time, it seemed much easier.
?
I hope this makes some sense to somebody!!!
?
Cheers,
?
Dave W7AQK
?
?
?
?

From: Ben Bangerter, K0IKR via Groups.Io
Sent: Sunday, November 5, 2017 5:43 AM
Subject: [QRPLabs] QCX T1 installation: A better way?
?
I began assembly of my 20 M version of the QCX nine days ago, and all went well thanks to the excellent manual provided by Hans.? Until I got to the most difficult component - the transformer T1.? I had no difficulty winding T1, but when it came to installing it on the populated board - fitting eight thin wires into the proper eight holes in the pcb, essentially simultaneously - that was a bit much for my 76 year old eyes and fingers!? So I removed the three small windings from the toroid, and installed the toroid, with only the large winding, on the board, soldering both leads.? I straightened the wires of the three removed windings and soldered one end of each length in the three outer holes for those windings.? Then one by one I wound those three small windings on the toroid, starting over the top of the core, passing the wire under the core (with the aid of a tweezers) to the outside, and repeating the process until the three turns were completed.? Then I inserted the free end of the wire through the inner hole for that winding.? Verifying that the sense of each winding was correct, I tightened the wires and soldered the last three connections.? This process went very smoothly, and I am pleased with the result.? If you have difficulty with the installation procedure? as described in the manual, you might want to try my method.

Of course, when it came time to align the QCX, the peaking capacitor was fully meshed and I had to add turns to the large winding!? But that went well, and the alignment process proceeded smoothly.? R24 ended up fully counterclockwise, but the image rejection is excellent, as is the sensitiviy.? The rf power output was low, as many have experienced, about 1.0 - 1.2 W with a 12 V supply (increasing as the key is held down for a few seconds).? My first contact (from Connecticut) was with OZ0JX in Denmark, with my 1 W and dipole up 20 ft!? I subsequently raised the power output to 1.6 - 2.0 W by removing one turn from each of the LPF inductors.? And I solved the turn-on issue with a diode (1N5711) between pins 7 and 20 of the ATmega328P.?

Hans has done an incredible job with this project, great hardware design, software design, board layout, and the best construction? manual ever done.? And all at such a low cost!? Plus he spends a lot of time in support of those of us encountering problems and having suggestions or questions about this remarkable little rig.? What a fine contribution to the QRP world and the Ham Radio builder/maker community!

Ben - K0IKR
_._,_._,_
?
 

Hi Ben, Lou, Dave

Thanks for the kind comments and support <blush />

I do wonder though... about T1, did any of you try to follow the manual steps and think that they need to be changed or improved somehow? The manual steps were designed to make winding T1 quite reliable. Doing everything in one long winding means you automatically get the "sense" (anticlockwise vs clockwise) correct for all windings. And my idea with loops that are cut one by one, should mean that it is a lot more manageable to fit the wires in the right holes. You only have to do 2 wires at a time, not 8. But I am very interested to hear if you think that the method can be improved - or maybe if your alternative methods are something that you feel worked best for you, but the manual steps should be left alone. Maybe winding and installing T1 is a very personal thing :-)

73 Hans G0UPL

On Sun, Nov 5, 2017 at 6:15 PM, w7aqk <w7aqk@...> wrote:
Ben and all,
?
It sounds like you and I are similarly afflicted!? My eyes are not very good either, one having macular degeneration to the point that everything is a blur with that eye.? Thus my vision tends to be 2 dimensional rather than 3 dimensional.? It’s weird!? I’m even close to your 76 year mark—two more weeks!
?
I agree that T1 is a bit daunting, but mostly it is just cumbersome with those 8 legs.? Your method makes some sense, but seemingly has some complications as well.? Admittedly I fumbled around a bit installing T1, but eventually I think I came around to a process that worked pretty well.? It’s a bit difficult to describe, but I’ll take a stab at it.
?
First, you need to make sure that you have the 4 connections on each side of the toroid nicely aligned as they come off the toroid.? I used WA4MZS’s drawing as a guide.? Each connection on the left has an opposite connection on the right, i.e. 7 and 2, 3 and 8, etc., but these leads are only “opposing” in how they are inserted into the board from top to bottom.? They are not the opposing ends of the same winding.? Then I make sure that each successive pair is somewhat shorter than the pair just above it.? 3 and 8 are shorter than 7 and 2, etc.? I work from the “top” inserting 7 and 2 first.? Then I go to 3 and 8, and since they are shorter, they aren’t crumpled too much due to the insertion of 7 and 2.? I “coax” 3 and 8 into their respective holes with needle nose pliers.? Then I go on to 5 and 4, and finally 6 and 1.? Working from “top to bottom” seemed easier than doing the reverse.? Once all 8 leads are inserted in their respective holes you can pull each lead from the bottom of the board to tighten everything up nicely.? Before soldering anything, I checked for proper continuity on each pair that actually opposes each other—1 and 2, 3 and 4, etc.? With 8 leads it isn’t difficult to get a couple of leads crossed, but the procedure I followed seemed to prevent that, or at least minimize the probability.?
?
I’m not suggesting that my method is necessarily better, but once I settled on this process, things seemed to go relatively smoothly.? I fumbled around a bit before coming to this method.? Clipping each successive pair shorter seemed to help a lot—you just don’t want to have the last pair be too short!? This also helps you properly identify each pair.? If all your leads are approximately the same length, it’s easy to get them confused.? I think my lack of “3D” vision made things a lot more difficult in my case, but when I went to dealing with just 2 wires at a time, it seemed much easier.
?
I hope this makes some sense to somebody!!!
?
Cheers,
?
Dave W7AQK
?
?
?
?
From: Ben Bangerter, K0IKR via Groups.Io
Sent: Sunday, November 5, 2017 5:43 AM
Subject: [QRPLabs] QCX T1 installation: A better way?
?
I began assembly of my 20 M version of the QCX nine days ago, and all went well thanks to the excellent manual provided by Hans.? Until I got to the most difficult component - the transformer T1.? I had no difficulty winding T1, but when it came to installing it on the populated board - fitting eight thin wires into the proper eight holes in the pcb, essentially simultaneously - that was a bit much for my 76 year old eyes and fingers!? So I removed the three small windings from the toroid, and installed the toroid, with only the large winding, on the board, soldering both leads.? I straightened the wires of the three removed windings and soldered one end of each length in the three outer holes for those windings.? Then one by one I wound those three small windings on the toroid, starting over the top of the core, passing the wire under the core (with the aid of a tweezers) to the outside, and repeating the process until the three turns were completed.? Then I inserted the free end of the wire through the inner hole for that winding.? Verifying that the sense of each winding was correct, I tightened the wires and soldered the last three connections.? This process went very smoothly, and I am pleased with the result.? If you have difficulty with the installation procedure? as described in the manual, you might want to try my method.

Of course, when it came time to align the QCX, the peaking capacitor was fully meshed and I had to add turns to the large winding!? But that went well, and the alignment process proceeded smoothly.? R24 ended up fully counterclockwise, but the image rejection is excellent, as is the sensitiviy.? The rf power output was low, as many have experienced, about 1.0 - 1.2 W with a 12 V supply (increasing as the key is held down for a few seconds).? My first contact (from Connecticut) was with OZ0JX in Denmark, with my 1 W and dipole up 20 ft!? I subsequently raised the power output to 1.6 - 2.0 W by removing one turn from each of the LPF inductors.? And I solved the turn-on issue with a diode (1N5711) between pins 7 and 20 of the ATmega328P.?

Hans has done an incredible job with this project, great hardware design, software design, board layout, and the best construction? manual ever done.? And all at such a low cost!? Plus he spends a lot of time in support of those of us encountering problems and having suggestions or questions about this remarkable little rig.? What a fine contribution to the QRP world and the Ham Radio builder/maker community!

Ben - K0IKR


 

73 Hans,

Well I found the instructions good along with the hand drawing, but, somehow I ended up winding the opposite direction or opposite way around, then when I tried to fit it to the board, I found that my small windings were on the wrong side, if that makes sense, and so to make it go in I ended up swapping 5&6, 3&4, 7&8 holes, which I guess would have changed the sense. So it did not work right. Perhaps a clearer way may be to instruct specifically regards clockwise or anti-clockwise... or perhaps I just did not follow it 100% the first time around. The second time around, I took great care to follow your instructions along with the images, and it was FINE.

So it seems the instructions aren't lacking and not sure how to improve it. I did first and second time make use of the loops, and that worked well in both cases. Second time around I just made everything long perhaps 10cm LOL since I wanted room to play with, plus, I wound 31 turns and 3x3 turns, hoping that may make it work on both 20m and 30m, but it seems to work fine on 30m but there is a horrible loud rapid fire sound as soon as I go above 12.505 MHz. Maybe not related to T1.
 

Hi Lou
?
So it seems the instructions aren't lacking and not sure how to improve it. I did first and second time make use of the loops, and that worked well in both cases. Second time around I just made everything long perhaps 10cm LOL since I wanted room to play with, plus, I wound 31 turns and 3x3 turns, hoping that may make it work on both 20m and 30m, but it seems to work fine on 30m but there is a horrible loud rapid fire sound as soon as I go above 12.505 MHz. Maybe not related to T1.

It is not T1-related. I think this is something else completely, and is related to the firmware. It was a short-cut I took, while under time-pressure getting it all ready in time for the YOTA summercamp buildathon. We cannot really call it a "bug" since QCX is intended as a single-band rig and everything works fine. But you are taking it a little out of normal usage, and tuning a QCX built for one band, onto another band. Well anyway, I can easily change it in the next firmware version I think!

73 Hans G0UPL
 

I followed the instructions in the manual and got T1 installed correctly I think. That T1 .... be afraid, be VERY afraid! Get it right first time!

Well as you all know I wanted to improve mine by redoing it - DON'T EVER REDO T1!!! I say that tongue in cheek but it's true really. As suggested on this thread, I will be rewinding my T1 and mounting it on veroboard. From there I'll take the wires to their various places, sometimes direct but where there is PCB damage? I'll have to wire to the nearest components and check continuity.

Still waiting for some positive energy but the "heart surgery" needs to be done! T1, ironically IS the heart :)

I even tried the TX as a WSPR beacon. It's no understatement, those finals get HOT!!! Heatsinks need to be fitted for WSPR. I set my frequency to 7.040.100 but no station heard my effort .... Certainly an overkill for a CW beacon ..... so I've got to get that RX operational again.

T1 ..... we'll meet each other half way, OK?

73 de Eddie ZS6BNE




On Mon, Nov 6, 2017 at 10:46 AM Hans Summers <hans.summers@...> wrote:
Hi Lou
?
So it seems the instructions aren't lacking and not sure how to improve it. I did first and second time make use of the loops, and that worked well in both cases. Second time around I just made everything long perhaps 10cm LOL since I wanted room to play with, plus, I wound 31 turns and 3x3 turns, hoping that may make it work on both 20m and 30m, but it seems to work fine on 30m but there is a horrible loud rapid fire sound as soon as I go above 12.505 MHz. Maybe not related to T1.

It is not T1-related. I think this is something else completely, and is related to the firmware. It was a short-cut I took, while under time-pressure getting it all ready in time for the YOTA summercamp buildathon. We cannot really call it a "bug" since QCX is intended as a single-band rig and everything works fine. But you are taking it a little out of normal usage, and tuning a QCX built for one band, onto another band. Well anyway, I can easily change it in the next firmware version I think!

73 Hans G0UPL
 

I followed the instructions and made loops and still got it wrong! I noticed before I soldered it in place so it was easy to correct. We are really lucky you know because there are not many people with the superb design skills Hans has that can put it all in words so well. He is a genius!
73
Roger.

On 5 Nov 2017 15:15, "w7aqk" <w7aqk@...> wrote:
Ben and all,
?
It sounds like you and I are similarly afflicted!? My eyes are not very good either, one having macular degeneration to the point that everything is a blur with that eye.? Thus my vision tends to be 2 dimensional rather than 3 dimensional.? It’s weird!? I’m even close to your 76 year mark—two more weeks!
?
I agree that T1 is a bit daunting, but mostly it is just cumbersome with those 8 legs.? Your method makes some sense, but seemingly has some complications as well.? Admittedly I fumbled around a bit installing T1, but eventually I think I came around to a process that worked pretty well.? It’s a bit difficult to describe, but I’ll take a stab at it.
?
First, you need to make sure that you have the 4 connections on each side of the toroid nicely aligned as they come off the toroid.? I used WA4MZS’s drawing as a guide.? Each connection on the left has an opposite connection on the right, i.e. 7 and 2, 3 and 8, etc., but these leads are only “opposing” in how they are inserted into the board from top to bottom.? They are not the opposing ends of the same winding.? Then I make sure that each successive pair is somewhat shorter than the pair just above it.? 3 and 8 are shorter than 7 and 2, etc.? I work from the “top” inserting 7 and 2 first.? Then I go to 3 and 8, and since they are shorter, they aren’t crumpled too much due to the insertion of 7 and 2.? I “coax” 3 and 8 into their respective holes with needle nose pliers.? Then I go on to 5 and 4, and finally 6 and 1.? Working from “top to bottom” seemed easier than doing the reverse.? Once all 8 leads are inserted in their respective holes you can pull each lead from the bottom of the board to tighten everything up nicely.? Before soldering anything, I checked for proper continuity on each pair that actually opposes each other—1 and 2, 3 and 4, etc.? With 8 leads it isn’t difficult to get a couple of leads crossed, but the procedure I followed seemed to prevent that, or at least minimize the probability.?
?
I’m not suggesting that my method is necessarily better, but once I settled on this process, things seemed to go relatively smoothly.? I fumbled around a bit before coming to this method.? Clipping each successive pair shorter seemed to help a lot—you just don’t want to have the last pair be too short!? This also helps you properly identify each pair.? If all your leads are approximately the same length, it’s easy to get them confused.? I think my lack of “3D” vision made things a lot more difficult in my case, but when I went to dealing with just 2 wires at a time, it seemed much easier.
?
I hope this makes some sense to somebody!!!
?
Cheers,
?
Dave W7AQK
?
?
?
?
From: Ben Bangerter, K0IKR via Groups.Io
Sent: Sunday, November 5, 2017 5:43 AM
Subject: [QRPLabs] QCX T1 installation: A better way?
?
I began assembly of my 20 M version of the QCX nine days ago, and all went well thanks to the excellent manual provided by Hans.? Until I got to the most difficult component - the transformer T1.? I had no difficulty winding T1, but when it came to installing it on the populated board - fitting eight thin wires into the proper eight holes in the pcb, essentially simultaneously - that was a bit much for my 76 year old eyes and fingers!? So I removed the three small windings from the toroid, and installed the toroid, with only the large winding, on the board, soldering both leads.? I straightened the wires of the three removed windings and soldered one end of each length in the three outer holes for those windings.? Then one by one I wound those three small windings on the toroid, starting over the top of the core, passing the wire under the core (with the aid of a tweezers) to the outside, and repeating the process until the three turns were completed.? Then I inserted the free end of the wire through the inner hole for that winding.? Verifying that the sense of each winding was correct, I tightened the wires and soldered the last three connections.? This process went very smoothly, and I am pleased with the result.? If you have difficulty with the installation procedure? as described in the manual, you might want to try my method.

Of course, when it came time to align the QCX, the peaking capacitor was fully meshed and I had to add turns to the large winding!? But that went well, and the alignment process proceeded smoothly.? R24 ended up fully counterclockwise, but the image rejection is excellent, as is the sensitiviy.? The rf power output was low, as many have experienced, about 1.0 - 1.2 W with a 12 V supply (increasing as the key is held down for a few seconds).? My first contact (from Connecticut) was with OZ0JX in Denmark, with my 1 W and dipole up 20 ft!? I subsequently raised the power output to 1.6 - 2.0 W by removing one turn from each of the LPF inductors.? And I solved the turn-on issue with a diode (1N5711) between pins 7 and 20 of the ATmega328P.?

Hans has done an incredible job with this project, great hardware design, software design, board layout, and the best construction? manual ever done.? And all at such a low cost!? Plus he spends a lot of time in support of those of us encountering problems and having suggestions or questions about this remarkable little rig.? What a fine contribution to the QRP world and the Ham Radio builder/maker community!

Ben - K0IKR

 

I think the T1 winding instructions are very good. I made the mistake of winding my 80m kit first. To my way of thinking, the 80m version of T1 is the most difficult, but not difficult. Had I wound the 20m version first I would have had more experience with winding this kind of transformer. As it is, I consider the entire process as a wonderful classroom for things I have always wanted to know how to do

73
NE5U

Mike

PS- my T11 tirade of a few weeks ago was, for the most part, tongue in cheek. I thought the added "dramatic twists" would be humorous. They were to me. They weren't to others.?

On Mon, 6 Nov 2017 11:15:14 +0300, "Hans Summers" wrote:
?
Hi Ben, Lou, Dave
?
Thanks for the kind comments and support <blush />
?
I do wonder though... about T1, did any of you try to follow the manual steps and think that they need to be changed or improved somehow? The manual steps were designed to make winding T1 quite reliable. Doing everything in one long winding means you automatically get the "sense" (anticlockwise vs clockwise) correct for all windings. And my idea with loops that are cut one by one, should mean that it is a lot more manageable to fit the wires in the right holes. You only have to do 2 wires at a time, not 8. But I am very interested to hear if you think that the method can be improved - or maybe if your alternative methods are something that you feel worked best for you, but the manual steps should be left alone. Maybe winding and installing T1 is a very personal thing :-)
?
73 Hans G0UPL
?
On Sun, Nov 5, 2017 at 6:15 PM, w7aqk <w7aqk@...> wrote:

Ben and all,
?
It sounds like you and I are similarly afflicted! My eyes are not very good either, one having macular degeneration to the point that everything is a blur with that eye. Thus my vision tends to be 2 dimensional rather than 3 dimensional. It’s weird! I’m even close to your 76 year mark—two more weeks!
?
I agree that T1 is a bit daunting, but mostly it is just cumbersome with those 8 legs. Your method makes some sense, but seemingly has some complications as well. Admittedly I fumbled around a bit installing T1, but eventually I think I came around to a process that worked pretty well. It’s a bit difficult to describe, but I’ll take a stab at it.
?
First, you need to make sure that you have the 4 connections on each side of the toroid nicely aligned as they come off the toroid. I used WA4MZS’s drawing as a guide. Each connection on the left has an opposite connection on the right, i.e. 7 and 2, 3 and 8, etc., but these leads are only “opposing” in how they are inserted into the board from top to bottom. They are not the opposing ends of the same winding. Then I make sure that each successive pair is somewhat shorter than the pair just above it. 3 and 8 are shorter than 7 and 2, etc. I work from the “top” inserting 7 and 2 first. Then I go to 3 and 8, and since they are shorter, they aren’t crumpled too much due to the insertion of 7 and 2. I “coax” 3 and 8 into their respective holes with needle nose pliers. Then I go on to 5 and 4, and finally 6 and 1. Working from “top to bottom” seemed easier than doing the reverse. Once all 8 leads are inserted in their respective holes you can pull each lead from the bottom of the board to tighten everything up nicely. Before soldering anything, I checked for proper continuity on each pair that actually opposes each other—1 and 2, 3 and 4, etc. With 8 leads it isn’t difficult to get a couple of leads crossed, but the procedure I followed seemed to prevent that, or at least minimize the probability.
?
I’m not suggesting that my method is necessarily better, but once I settled on this process, things seemed to go relatively smoothly. I fumbled around a bit before coming to this method. Clipping each successive pair shorter seemed to help a lot—you just don’t want to have the last pair be too short! This also helps you properly identify each pair. If all your leads are approximately the same length, it’s easy to get them confused. I think my lack of “3D” vision made things a lot more difficult in my case, but when I went to dealing with just 2 wires at a time, it seemed much easier.
?
I hope this makes some sense to somebody!!!
?
Cheers,
?
Dave W7AQK
?
?
?
?
From: Ben Bangerter, K0IKR via Groups.Io
Sent: Sunday, November 5, 2017 5:43 AM
Subject: [QRPLabs] QCX T1 installation: A better way?
?
I began assembly of my 20 M version of the QCX nine days ago, and all went well thanks to the excellent manual provided by Hans. Until I got to the most difficult component - the transformer T1. I had no difficulty winding T1, but when it came to installing it on the populated board - fitting eight thin wires into the proper eight holes in the pcb, essentially simultaneously - that was a bit much for my 76 year old eyes and fingers! So I removed the three small windings from the toroid, and installed the toroid, with only the large winding, on the board, soldering both leads. I straightened the wires of the three removed windings and soldered one end of each length in the three outer holes for those windings. Then one by one I wound those three small windings on the toroid, starting over the top of the core, passing the wire under the core (with the aid of a tweezers) to the outside, and repeating the process until the three turns were completed. Then I inserted the free end of the wire through the inner hole for that winding. Verifying that the sense of each winding was correct, I tightened the wires and soldered the last three connections. This process went very smoothly, and I am pleased with the result. If you have difficulty with the installation procedure as described in the manual, you might want to try my method.

Of course, when it came time to align the QCX, the peaking capacitor was fully meshed and I had to add turns to the large winding! But that went well, and the alignment process proceeded smoothly. R24 ended up fully counterclockwise, but the image rejection is excellent, as is the sensitiviy. The rf power output was low, as many have experienced, about 1.0 - 1.2 W with a 12 V supply (increasing as the key is held down for a few seconds). My first contact (from Connecticut) was with OZ0JX in Denmark, with my 1 W and dipole up 20 ft! I subsequently raised the power output to 1.6 - 2.0 W by removing one turn from each of the LPF inductors. And I solved the turn-on issue with a diode (1N5711) between pins 7 and 20 of the ATmega328P.

Hans has done an incredible job with this project, great hardware design, software design, board layout, and the best construction manual ever done. And all at such a low cost! Plus he spends a lot of time in support of those of us encountering problems and having suggestions or questions about this remarkable little rig. What a fine contribution to the QRP world and the Ham Radio builder/maker community!

Ben - K0IKR

?

?
 

I just wound and installed my T1 transformer yesterday evening for my 60m QCX.? I didn't find it too bad, but if I can't peak the C1 trimmer properly, I don't look forward to adding or removing windings from T1.? I will likely tack a small fixed cap across C1 trimmer to peak the bandpass filter...

I recall that the SW line of QRP rigs came with a few extra caps so you could get the VFO in the right part of the band.?? Hans might want to consider this option when he makes his next revision to the QCX circuit board.
w7aqk
 

开云体育

Hans and all,
?
No, I don’t think you need to change the instructions.? They are fine, and very possibly the best method.? I read them, and pondered them for a good while.? In the end I concluded, and probably erroneously, that they were intended mainly to insure that the primary and secondary windings were all wound in the same sense.? I didn’t think I would have a problem with “sense” so long as I was careful, and I really wanted to be dealing with shorter overall lengths of wire.? So,? I decided to wind each secondary, and the primary, individually.? That may not be very sound reasoning in the minds of most.? That said, I had? already concluded that I would try your method on my 2nd kit, which will be one for 20 meters.? It’s highly probable that I will be a complete convert after doing that!!? Hi.
?
Cheers,
?
Dave W7AQK
?
?

From: Hans Summers
Sent: Monday, November 6, 2017 1:15 AM
Subject: Re: [QRPLabs] QCX T1 installation: A better way?
?
Hi Ben, Lou, Dave
?
Thanks for the kind comments and support <blush />
?
I do wonder though... about T1, did any of you try to follow the manual steps and think that they need to be changed or improved somehow? The manual steps were designed to make winding T1 quite reliable. Doing everything in one long winding means you automatically get the "sense" (anticlockwise vs clockwise) correct for all windings. And my idea with loops that are cut one by one, should mean that it is a lot more manageable to fit the wires in the right holes. You only have to do 2 wires at a time, not 8. But I am very interested to hear if you think that the method can be improved - or maybe if your alternative methods are something that you feel worked best for you, but the manual steps should be left alone. Maybe winding and installing T1 is a very personal thing :-)
?
73 Hans G0UPL
?
?
 

开云体育

On 7/11/2017 12:41 AM, Roger G4IUP wrote:
I followed the instructions and made loops and still got it wrong! I noticed before I soldered it in place so it was easy to correct.

The hand drawn diagram by Ed WA4MZS that is included in the manual version that I downloaded the other day was a great help in ensuring the turns went onto the toroid the right way and the wires went into the correct holes.? Between that and Hans' instructions, winding T1 and fitting it to the board was fairly straight forward (just a few 'deep breaths' and mutterings of "patience, patience"!)? However, I did manage to break the wire at the ground/earth end of the BPF winding while moving T1 about to get it sitting neatly on the PCB after it was soldered into place.? Fortunately that could be easily fixed without removing T1 by soldering the very short broken wire end directly to the solder pad on the top side of the through-hole.

The rest of the assembly went well, and after doing the recommended modifications the QCX powered up and aligned without a problem.? Updating the .hex file from 1.00a to 1.00c with eXteme Burner software and a "$2" e-Bay USBASP program interface was easy and made a good difference to the performance (in particular, reducing the back ground 'digital'/'popping' noises) and the QCX's overall operation.? There is plenty of audio and a 0.1uV signal from my sig-gen is readily detected.

A cursory listen about 40m in the late afternoon and last night with the QCX connected to a base loaded 20ft vertical netted stations from NA, AS, OC and EU.? Best 'DX' heard was an EA1 near my antipode about 2 hours before my sunset.

73, Bob? ZL1RS
Karl Schwab
 

While I await my QCX kit to arrive, where can I find " WA4MZS’s drawing" for the T1 windings?? Tnx, Karl


On Monday, November 6, 2017 4:19 PM, Bob Sutton via Groups.Io <zl1rs@...> wrote:


On 7/11/2017 12:41 AM, Roger G4IUP wrote:
I followed the instructions and made loops and still got it wrong! I noticed before I soldered it in place so it was easy to correct.

The hand drawn diagram by Ed WA4MZS that is included in the manual version that I downloaded the other day was a great help in ensuring the turns went onto the toroid the right way and the wires went into the correct holes.? Between that and Hans' instructions, winding T1 and fitting it to the board was fairly straight forward (just a few 'deep breaths' and mutterings of "patience, patience"!)? However, I did manage to break the wire at the ground/earth end of the BPF winding while moving T1 about to get it sitting neatly on the PCB after it was soldered into place.? Fortunately that could be easily fixed without removing T1 by soldering the very short broken wire end directly to the solder pad on the top side of the through-hole.

The rest of the assembly went well, and after doing the recommended modifications the QCX powered up and aligned without a problem.? Updating the .hex file from 1.00a to 1.00c with eXteme Burner software and a "$2" e-Bay USBASP program interface was easy and made a good difference to the performance (in particular, reducing the back ground 'digital'/'popping' noises) and the QCX's overall operation.? There is plenty of audio and a 0.1uV signal from my sig-gen is readily detected.

A cursory listen about 40m in the late afternoon and last night with the QCX connected to a base loaded 20ft vertical netted stations from NA, AS, OC and EU.? Best 'DX' heard was an EA1 near my antipode about 2 hours before my sunset.

73, Bob? ZL1RS
 

Hi Karl,

All you have to do is Read The Fine Manual.
[Page 50 in the A4 version of v1.08]

A good read, while you wait.

Best Wishes
David, GM8XBZ

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Karl Schwab
via Groups.Io
Sent: 07 November 2017 13:51
To: [email protected]
Subject: Re: [QRPLabs] QCX T1 installation: A better way?

While I await my QCX kit to arrive, where can I find " WA4MZS's drawing" for
the T1 windings? Tnx, Karl
 

Just an update on my 60 metre QCX.? So, I did have to remove 5 turns from T1 to be able to peak the C1 trimmer.

Luckily, I have a solder sucker, because trying to poke a 30 AWG wire through a solder-filled hole is quite tricky.? I don't have an antenna for 60 metres, but a 25 foot length of wire run along the floor does cause the noise floor to increase when I jam it in the BNC connector and I can hear a few RTTY signals around 5340 kHz.

I also upgraded my firmware to version C.? I couldn't figure out how to do it with my linux laptop, so I had to spark up my Windows desktop (and wait for all those updates to load)

Now I guess I better build an antenna while I'm waiting for the feds to update the regs to allow operation from 5351.5 to 5366.5 kHz.? I believe I can already operate on ~5358.5?kHz
 

Hi all

Just a quick observation. It is indeed VERY difficult to replace components, or to take turns off toroids and get the wire back into the hole when it is filled with solder. Here I do have solder braid (which does not work in this case) but I don't have a solder sucker. I know, I ought to get organised. Separate topic. But anyway - what I wanted to say (observe) is that there isn't any harm just soldering the component (or wire) on the top side of the PCB, not going through the hole at all. I do this all the time :-)

73 Hans G0UPL

On Wed, Nov 8, 2017 at 5:02 PM, va3rr via Groups.Io <va3rr@...> wrote:
Just an update on my 60 metre QCX.? So, I did have to remove 5 turns from T1 to be able to peak the C1 trimmer.

Luckily, I have a solder sucker, because trying to poke a 30 AWG wire through a solder-filled hole is quite tricky.? I don't have an antenna for 60 metres, but a 25 foot length of wire run along the floor does cause the noise floor to increase when I jam it in the BNC connector and I can hear a few RTTY signals around 5340 kHz.

I also upgraded my firmware to version C.? I couldn't figure out how to do it with my linux laptop, so I had to spark up my Windows desktop (and wait for all those updates to load)

Now I guess I better build an antenna while I'm waiting for the feds to update the regs to allow operation from 5351.5 to 5366.5 kHz.? I believe I can already operate on ~5358.5?kHz


 

I'm waiting for the feds to update the regs to allow operation from 5351.5 to 5366.5 kHz.
I hope that's not too long.
You could test the RX on Dial 5.364700 (TX 5.366200+/-) as used in EU.
Conditions not been too good X-pond the last few days but usually signals make it OK.

73 Alan G4ZFQ
 

Use a round wooden toothpick and put it through the molten solder, reheat and jam it through.? Bamboo toothpicks work great.

73
Larry
KB3CUF

On Wed, Nov 8, 2017 at 9:28 AM, Alan G4ZFQ <alan4alan@...> wrote:
?I'm waiting for the feds to update the regs to allow operation from 5351.5 to 5366.5 kHz.?

I hope that's not too long.
You could test the RX on Dial 5.364700 (TX 5.366200+/-) as used in EU.
Conditions not been too good X-pond the last few days but usually signals make it OK.

73 Alan G4ZFQ
 

开云体育

Hi Hans,

Thanks for that and info. That is exactly how I removed turns from my LPF, I just resoldered the lead to the top pad instead of trying to get it back in the hole, a lot less headache.

Joel?
N6ALT

On Nov 8, 2017, at 6:14 AM, Hans Summers <hans.summers@...> wrote:

Hi all

Just a quick observation. It is indeed VERY difficult to replace components, or to take turns off toroids and get the wire back into the hole when it is filled with solder. Here I do have solder braid (which does not work in this case) but I don't have a solder sucker. I know, I ought to get organised. Separate topic. But anyway - what I wanted to say (observe) is that there isn't any harm just soldering the component (or wire) on the top side of the PCB, not going through the hole at all. I do this all the time :-)

73 Hans G0UPL

On Wed, Nov 8, 2017 at 5:02 PM, va3rr via Groups.Io <va3rr@...> wrote:
Just an update on my 60 metre QCX.? So, I did have to remove 5 turns from T1 to be able to peak the C1 trimmer.

Luckily, I have a solder sucker, because trying to poke a 30 AWG wire through a solder-filled hole is quite tricky.? I don't have an antenna for 60 metres, but a 25 foot length of wire run along the floor does cause the noise floor to increase when I jam it in the BNC connector and I can hear a few RTTY signals around 5340 kHz.

I also upgraded my firmware to version C.? I couldn't figure out how to do it with my linux laptop, so I had to spark up my Windows desktop (and wait for all those updates to load)

Now I guess I better build an antenna while I'm waiting for the feds to update the regs to allow operation from 5351.5 to 5366.5 kHz.? I believe I can already operate on ~5358.5?kHz


1 - 19 of 19