Topics

QCX-mini update, 24-Sep-2020


Hans Summers
 

Hi all

On Monday the SMD parts arrived and I have been busy soldering together two QCX-mini prototypes, debugging etc. I now have them both working properly. I built one for 30m and one for 20m. The transmitters produce 5.0W and 4.5W respectively (at 12V supply). Receiver alignment and performance is the same as QCX+. 

This afternoon I connected the 20m version to the antenna and had the first QCX-mini QSO, with Feri HA2EOU who reported 559. 

I have added some photographs of the completed boards here http://qrp-labs.com/qcxmini/qcxminidev2.html

The enclosure prototype should be here in the next day or two. 

73 Hans G0UPL
http://qrp-labs.com


Julian Rolfe
 

Hi Hans,

Can we assume that every component NON surface mount will ours to assemble, so NP0 capacitors, wind the Toroids, switches, sockets etc.

Thanks,
Jules - G4UET




On Thursday, September 24, 2020, 2:47 pm, Hans Summers <hans.summers@...> wrote:

Hi all

On Monday the SMD parts arrived and I have been busy soldering together two QCX-mini prototypes, debugging etc. I now have them both working properly. I built one for 30m and one for 20m. The transmitters produce 5.0W and 4.5W respectively (at 12V supply). Receiver alignment and performance is the same as QCX+. 

This afternoon I connected the 20m version to the antenna and had the first QCX-mini QSO, with Feri HA2EOU who reported 559. 

I have added some photographs of the completed boards here http://qrp-labs.com/qcxmini/qcxminidev2.html

The enclosure prototype should be here in the next day or two. 

73 Hans G0UPL
http://qrp-labs.com


Hans Summers
 

Hi Jules

Yes! All SMD is factory pre-assembled. Everything else will be yours to assemble. We will also have a limited number of assembled kits like we do for QCX+.

73 Hans G0UPL

On Thu, Sep 24, 2020 at 5:07 PM Julian Rolfe via groups.io <g4uet=yahoo.co.uk@groups.io> wrote:
Hi Hans,

Can we assume that every component NON surface mount will ours to assemble, so NP0 capacitors, wind the Toroids, switches, sockets etc.

Thanks,
Jules - G4UET




On Thursday, September 24, 2020, 2:47 pm, Hans Summers <hans.summers@...> wrote:

Hi all

On Monday the SMD parts arrived and I have been busy soldering together two QCX-mini prototypes, debugging etc. I now have them both working properly. I built one for 30m and one for 20m. The transmitters produce 5.0W and 4.5W respectively (at 12V supply). Receiver alignment and performance is the same as QCX+. 

This afternoon I connected the 20m version to the antenna and had the first QCX-mini QSO, with Feri HA2EOU who reported 559. 

I have added some photographs of the completed boards here http://qrp-labs.com/qcxmini/qcxminidev2.html

The enclosure prototype should be here in the next day or two. 

73 Hans G0UPL
http://qrp-labs.com


Al Holt
 

Nice photos! The LCD brightness resistor (?) in one of the photos caught my eye. In the QCX+20m I just completed I used the exact same value (560R) carbon film resistor in my build after reading through the thread on the 'hot' 240R in the QCX+. BTW, mine came from a Radio Shack resistor pack I got when they were going out of business.

That's a great looking little rig, you'll sell a bunch!!



--Al
WD4AH


Brian George
 

Could not help noticing the extra turn on L2 !!

73
Brian
G3ZOH


Hans Summers
 

Hi Al
 
Nice photos! The LCD brightness resistor (?) in one of the photos caught my eye. In the QCX+20m I just completed I used the exact same value (560R) carbon film resistor in my build after reading through the thread on the 'hot' 240R in the QCX+. BTW, mine came from a Radio Shack resistor pack I got when they were going out of business.

What sharp eyes you have...

It's 270-ohms in the QCX/QCX+ not 240-ohms. After all the comments about reducing the brightness slightly and it still being perfectly OK I decided to try the higher value, particularly in view of reducing the current consumption. Indeed it is bright enough, very fine!

73 Hans G0UPL
http://qrp-labs.com


Hans Summers
 

Hi Brian
 
Could not help noticing the extra turn on L2 !!

What sharp eyes YOU have too... evidently the 3 hours of sleep allowed at night by the most junior lab tech hasn't improved my ability to count.

73 Hans G0UPL
http://qrp-labs.com


Michael N6MST
 

I'm no designer and have never created my own PCB, so take my comment with that grain of salt. Why the separate board for the controls?


Brian George
 

Hi Hans, I don't know how you manage to run a company, produce all these innovative designs and raise a family at the same time. 
I suffer from acute toroiditis - every time I see a toroid I have to count the turns!!

73
Brian
G3ZOH 


Christiaan PA3FUN
 

As I stated before in another newsgroup;
money burning in my pocket, can't wait for the QCX-mini ordering hotline to open up :).


Steve Horton
 


Hans,

I enjoyed watching the QCX-mini Video! It looks like a hit.

I was wondering what the typical receive mode current at 12v and 15v would be on the  QCX-mini? I remember you talked about using different OP AMPS and it would be lower current than the Classic QCX. But mainly 12v will be my portable voltage.

Thanks,
Steve WA4YGG
Georgia USA


Hans Summers
 

Hi Steve
 
I enjoyed watching the QCX-mini Video! It looks like a hit.

I was wondering what the typical receive mode current at 12v and 15v would be on the  QCX-mini? I remember you talked about using different OP AMPS and it would be lower current than the Classic QCX. But mainly 12v will be my portable voltage.

I just measured the QCX-20 prototype here, reading the mA from my bench PSU meter, in Receive. I will measure again more perfectly in the end, but for now these are preliminary measurements and I think they will not change much, the PSU is pretty accurate. 

12V supply, LCD backlight OFF: 58mA
12V supply, LCD backlight ON: 72mA
13.8V supply, LCD backlight OFF: 58mA
13.8V supply, LCD backlight ON: 76mA

For comparison, a standard QCX measured 112mA at 12V, 118mA at 13.8V. I also measured a QCX+ and the results are identical to the QCX. 

FYI the op-amp I am using for IC6/7/8/9/10 is OPA2272. It's still a very high performance op-amp, but not quite as low noise as the LM4562. IC5 is the op-amp which is critical to performance (and largely determines the sensitivity of the receiver), and this is still the good old LM4562. OPA2272 sacrifices a bit on noise performance but with a much lower power consumption; the overall performance of the receiver has already been determined by IC5 and does not get impacted by IC6/7/8/9/10. 

73 Hans G0UPL


Kelly Jack
 

Re photo on the web page. There appears to be a solder bridge between two pins on the back of the main board immediately to the left of C22. Or is it just the lighting?


I'm really liking the low current ratings you've achieved with this Hans.


73



Simon
VK3ELH


Hans Summers
 

Hi Simon
 
Re photo on the web page. There appears to be a solder bridge between two pins on the back of the main board immediately to the left of C22. Or is it just the lighting?

Some of you guys have eyes like a hawk!   :-)    GOOD QUESTION! 

There isn't a solder bridge. Those two pins are part of the 2x4-pin header connector which connects to the Controls PCB. What you see there is not a solder bridge, it is a 10K resistor. It is R46. I removed R46 from the Controls PCB and I soldered it instead, temporarily, between these two pins of the 2x4-pin header on the main PCB. If you look at the photo of the Controls PCB, you will see that R46 is absent on the Controls PCB. Look for the three resistors between the two buttons. These are the resistors which set the analog multiplexing voltages for the buttons reading signal fed to the processor. R46 (10K) is what pulls the signal low to 0V when no buttons are pressed. 

So why did I move that resistor to the main PCB, not on the Controls PCB sub-assembly? 

Because there are times when you want to debug an issue. Or, perhaps you want to squeeze/compress the turns on the LPF toroids, to optimize the power output... and you notice that the L1 toroid is situated right under the rotary encoder when the Controls PCB is installed in place. And maybe you want to debug something else that needs touching a 'scope probe to other things that are on the main PCB, under the installed Controls PCB. So you want to unplug the Controls PCB, and unplug the LCD PCB. Unplugging the LCD PCB is OK, just like on a QCX-Classic, you can take the LCD off and the radio functions just as normal. You want to unplug the Controls PCB to get at stuff underneath it and debug etc. BUT, you cannot, because then the analog signal that measures the button voltages, is floating. It is no longer tied to ground because the 10K resistor R46 is on the Controls PCB. The processor spontaneously and sporadically interprets the floating voltage as button presses and who knows what is happening, you haven't got the LCD plugged in so you have no idea, and no further useful debugging can be done.

Silly me, I didn't think of this while designing the PCB layout. But it became immediately apparent when working with the prototypes. Therefore on both my prototypes I moved R46 to the main PCB, tacked temporarily between the appropriate two pins of the 2x4-pin header. See? And now I can unplug the LCD board and unplug the Controls PCB and debug on the main board by itself, and the buttons signal is still grounded (via R46), so the processor does not implement any button presses. 

It is #10 on my list of 15 minor modifications needed to the PCB before production. 

73 Hans G0UPL
http://qrp-labs.com
 


Hans Summers
 

Hi Michael
 
I'm no designer and have never created my own PCB, so take my comment with that grain of salt. Why the separate board for the controls?

It is all to do with how the board assembly slots together, in minimum volume, and fits into the enclosure. In the design, the top PCB (the LCD PCB) slots into the PCB guide-rails in the top half of the enclosure extrusion. It slides in. It could not do so, if the controls were bolted to it, since the shafts of the controls and the buttons would absolutely prevent that.

So the controls must be fixed to the BOTTOM (main) PCB, where they can be plugged into the LCD PCB and bolted to it via 11mm spacers, AFTER the LCD PCB has been slid into the PCB guide rails. 

Yet, the controls cannot be installed on the main PCB itself, for two reasons:

1) They would take up too much space and I would not be able to make the QCX-mini as small as I wanted

2) The shafts of the controls and buttons would not reach through the enclosure. We would therefore have a situation the same as the QCX-classic, where Markus at BaMaTech had to machine shaft extensions and button extensions with chassis mount button hardware, to be able to fit the controls through the front panel. I don't want to do all that which would all escalate the costs and then you customers wouldn't so much like the higher price. 

So I worked hard on it, and after a LOT of thought, I came up with a quite neat design solution (I think neat, anyway), using the additional little Controls PCB. The added cost is low since I am able to create the little Controls PCB in the cut-out area of the LCD PCB, into which the LCD module body fits. The LCD module, when installed in the LCD PCB, and the LCD PCB slid into the guide rails, is at exactly the right height to fit perfectly behind the rectangular cut-out in the enclosure. The Controls PCB fits in this cut-out so I do not have to manufacture one more PCB, so there is no increased cost for PCB making. I just add the 2x4-pin connectors and another 11mm spacer. The Controls PCB is at a height of 12.6mm above the main PCB, which itself is 11mm behind the LCD PCB (mounted in the extrusion guide rails). It is designed at this 12.6mm height so that the shafts of the gain control and the rotary encoder prodtrude exactly a perfect amount through the holes in the enclosure, to be able to fit the knobs on perfectly. 

It's VERY neat even if I do say so myself ;-)   

The sample QCX-mini enclosure arrived here yesterday evening. 7 boxes dumped on my doorstep unceremoniously. Contents: 300pcs QCX+ enclosure, and 1pcs QCX-mini prototype enclosure! 

Photos to come later today...

73 Hans G0UPL
http://qrp-labs.com


Shirley Dulcey KE1L
 

Potentially half the power consumption -- that's a big change and a boon to the SOTA operators.

It's also big for anybody who wants to use a QCX Mini in the Spartan Sprint. I suspect it will become a popular rig (sans case and perhaps with the knobs removed) in that competition.

On Fri, Sep 25, 2020 at 3:41 AM Hans Summers <hans.summers@...> wrote:
Hi Simon
 
Re photo on the web page. There appears to be a solder bridge between two pins on the back of the main board immediately to the left of C22. Or is it just the lighting?

Some of you guys have eyes like a hawk!   :-)    GOOD QUESTION! 

There isn't a solder bridge. Those two pins are part of the 2x4-pin header connector which connects to the Controls PCB. What you see there is not a solder bridge, it is a 10K resistor. It is R46. I removed R46 from the Controls PCB and I soldered it instead, temporarily, between these two pins of the 2x4-pin header on the main PCB. If you look at the photo of the Controls PCB, you will see that R46 is absent on the Controls PCB. Look for the three resistors between the two buttons. These are the resistors which set the analog multiplexing voltages for the buttons reading signal fed to the processor. R46 (10K) is what pulls the signal low to 0V when no buttons are pressed. 

So why did I move that resistor to the main PCB, not on the Controls PCB sub-assembly? 

Because there are times when you want to debug an issue. Or, perhaps you want to squeeze/compress the turns on the LPF toroids, to optimize the power output... and you notice that the L1 toroid is situated right under the rotary encoder when the Controls PCB is installed in place. And maybe you want to debug something else that needs touching a 'scope probe to other things that are on the main PCB, under the installed Controls PCB. So you want to unplug the Controls PCB, and unplug the LCD PCB. Unplugging the LCD PCB is OK, just like on a QCX-Classic, you can take the LCD off and the radio functions just as normal. You want to unplug the Controls PCB to get at stuff underneath it and debug etc. BUT, you cannot, because then the analog signal that measures the button voltages, is floating. It is no longer tied to ground because the 10K resistor R46 is on the Controls PCB. The processor spontaneously and sporadically interprets the floating voltage as button presses and who knows what is happening, you haven't got the LCD plugged in so you have no idea, and no further useful debugging can be done.

Silly me, I didn't think of this while designing the PCB layout. But it became immediately apparent when working with the prototypes. Therefore on both my prototypes I moved R46 to the main PCB, tacked temporarily between the appropriate two pins of the 2x4-pin header. See? And now I can unplug the LCD board and unplug the Controls PCB and debug on the main board by itself, and the buttons signal is still grounded (via R46), so the processor does not implement any button presses. 

It is #10 on my list of 15 minor modifications needed to the PCB before production. 

73 Hans G0UPL
http://qrp-labs.com
 


David Wilcox
 

Hans, 

You are a genius!!!!  I think is it very important that you were self taught so you couldn’t be told by the university boffins that it “can’t be done”.  It’s engineers like you that keep the true spirit of ham radio alive for us “button pushers”.  Thank you for all you are doing for us.  With Covid we have cancelled all our normal swaps that I would have spent my radio money at.  I can’t wait until the mini’s are ready for shipment.  (also the QSX)....

Dave K8WPE since 1960 and still learning.  It’s “remembering” that I now have trouble with.

David J. Wilcox’s iPad

On Sep 25, 2020, at 4:19 AM, Hans Summers <hans.summers@...> wrote:


Hi Michael
 
I'm no designer and have never created my own PCB, so take my comment with that grain of salt. Why the separate board for the controls?

It is all to do with how the board assembly slots together, in minimum volume, and fits into the enclosure. In the design, the top PCB (the LCD PCB) slots into the PCB guide-rails in the top half of the enclosure extrusion. It slides in. It could not do so, if the controls were bolted to it, since the shafts of the controls and the buttons would absolutely prevent that.

So the controls must be fixed to the BOTTOM (main) PCB, where they can be plugged into the LCD PCB and bolted to it via 11mm spacers, AFTER the LCD PCB has been slid into the PCB guide rails. 

Yet, the controls cannot be installed on the main PCB itself, for two reasons:

1) They would take up too much space and I would not be able to make the QCX-mini as small as I wanted

2) The shafts of the controls and buttons would not reach through the enclosure. We would therefore have a situation the same as the QCX-classic, where Markus at BaMaTech had to machine shaft extensions and button extensions with chassis mount button hardware, to be able to fit the controls through the front panel. I don't want to do all that which would all escalate the costs and then you customers wouldn't so much like the higher price. 

So I worked hard on it, and after a LOT of thought, I came up with a quite neat design solution (I think neat, anyway), using the additional little Controls PCB. The added cost is low since I am able to create the little Controls PCB in the cut-out area of the LCD PCB, into which the LCD module body fits. The LCD module, when installed in the LCD PCB, and the LCD PCB slid into the guide rails, is at exactly the right height to fit perfectly behind the rectangular cut-out in the enclosure. The Controls PCB fits in this cut-out so I do not have to manufacture one more PCB, so there is no increased cost for PCB making. I just add the 2x4-pin connectors and another 11mm spacer. The Controls PCB is at a height of 12.6mm above the main PCB, which itself is 11mm behind the LCD PCB (mounted in the extrusion guide rails). It is designed at this 12.6mm height so that the shafts of the gain control and the rotary encoder prodtrude exactly a perfect amount through the holes in the enclosure, to be able to fit the knobs on perfectly. 

It's VERY neat even if I do say so myself ;-)   

The sample QCX-mini enclosure arrived here yesterday evening. 7 boxes dumped on my doorstep unceremoniously. Contents: 300pcs QCX+ enclosure, and 1pcs QCX-mini prototype enclosure! 

Photos to come later today...

73 Hans G0UPL
http://qrp-labs.com


Hans Summers
 

Hi Shirley

FYI I just popped my QCX-mini in the enclosure on the scales and it weighs in at 202g (QCX-30). Of course there is a little variation, depending on the amount of wire in the toroids (dependent on band). Also the final version may be a gram or two lighter, I am sure the SMD machines will do a much more precise job of the SMD soldering than my heavy-handed attempts which you see in the photos ;-) 

73 Hans G0UPL

On Fri, Sep 25, 2020 at 11:35 AM Shirley Dulcey KE1L <mark@...> wrote:
Potentially half the power consumption -- that's a big change and a boon to the SOTA operators.

It's also big for anybody who wants to use a QCX Mini in the Spartan Sprint. I suspect it will become a popular rig (sans case and perhaps with the knobs removed) in that competition.

On Fri, Sep 25, 2020 at 3:41 AM Hans Summers <hans.summers@...> wrote:
Hi Simon
 
Re photo on the web page. There appears to be a solder bridge between two pins on the back of the main board immediately to the left of C22. Or is it just the lighting?

Some of you guys have eyes like a hawk!   :-)    GOOD QUESTION! 

There isn't a solder bridge. Those two pins are part of the 2x4-pin header connector which connects to the Controls PCB. What you see there is not a solder bridge, it is a 10K resistor. It is R46. I removed R46 from the Controls PCB and I soldered it instead, temporarily, between these two pins of the 2x4-pin header on the main PCB. If you look at the photo of the Controls PCB, you will see that R46 is absent on the Controls PCB. Look for the three resistors between the two buttons. These are the resistors which set the analog multiplexing voltages for the buttons reading signal fed to the processor. R46 (10K) is what pulls the signal low to 0V when no buttons are pressed. 

So why did I move that resistor to the main PCB, not on the Controls PCB sub-assembly? 

Because there are times when you want to debug an issue. Or, perhaps you want to squeeze/compress the turns on the LPF toroids, to optimize the power output... and you notice that the L1 toroid is situated right under the rotary encoder when the Controls PCB is installed in place. And maybe you want to debug something else that needs touching a 'scope probe to other things that are on the main PCB, under the installed Controls PCB. So you want to unplug the Controls PCB, and unplug the LCD PCB. Unplugging the LCD PCB is OK, just like on a QCX-Classic, you can take the LCD off and the radio functions just as normal. You want to unplug the Controls PCB to get at stuff underneath it and debug etc. BUT, you cannot, because then the analog signal that measures the button voltages, is floating. It is no longer tied to ground because the 10K resistor R46 is on the Controls PCB. The processor spontaneously and sporadically interprets the floating voltage as button presses and who knows what is happening, you haven't got the LCD plugged in so you have no idea, and no further useful debugging can be done.

Silly me, I didn't think of this while designing the PCB layout. But it became immediately apparent when working with the prototypes. Therefore on both my prototypes I moved R46 to the main PCB, tacked temporarily between the appropriate two pins of the 2x4-pin header. See? And now I can unplug the LCD board and unplug the Controls PCB and debug on the main board by itself, and the buttons signal is still grounded (via R46), so the processor does not implement any button presses. 

It is #10 on my list of 15 minor modifications needed to the PCB before production. 

73 Hans G0UPL
http://qrp-labs.com
 


Hans Summers
 

Hi David

Hi hi, not sure about that genius allegation... but yes I think it helps to "not know what can't be done". Because a lot of the time, it actually can be. I noticed this during my former 22 years in IT in banks... a lot of the time I came out with a different solution than the "proper" software guys with their posh Computer Science degrees - because nobody had told me what couldn't be done ;-)   My education was in physics. So I am not even a proper engineer at all in any discipline. But I'm happy :-) 

73 Hans G0UPL
http://qrp-labs.com

On Fri, Sep 25, 2020 at 11:44 AM David Wilcox via groups.io <Djwilcox01=yahoo.com@groups.io> wrote:
Hans, 

You are a genius!!!!  I think is it very important that you were self taught so you couldn’t be told by the university boffins that it “can’t be done”.  It’s engineers like you that keep the true spirit of ham radio alive for us “button pushers”.  Thank you for all you are doing for us.  With Covid we have cancelled all our normal swaps that I would have spent my radio money at.  I can’t wait until the mini’s are ready for shipment.  (also the QSX)....

Dave K8WPE since 1960 and still learning.  It’s “remembering” that I now have trouble with.

David J. Wilcox’s iPad

On Sep 25, 2020, at 4:19 AM, Hans Summers <hans.summers@...> wrote:


Hi Michael
 
I'm no designer and have never created my own PCB, so take my comment with that grain of salt. Why the separate board for the controls?

It is all to do with how the board assembly slots together, in minimum volume, and fits into the enclosure. In the design, the top PCB (the LCD PCB) slots into the PCB guide-rails in the top half of the enclosure extrusion. It slides in. It could not do so, if the controls were bolted to it, since the shafts of the controls and the buttons would absolutely prevent that.

So the controls must be fixed to the BOTTOM (main) PCB, where they can be plugged into the LCD PCB and bolted to it via 11mm spacers, AFTER the LCD PCB has been slid into the PCB guide rails. 

Yet, the controls cannot be installed on the main PCB itself, for two reasons:

1) They would take up too much space and I would not be able to make the QCX-mini as small as I wanted

2) The shafts of the controls and buttons would not reach through the enclosure. We would therefore have a situation the same as the QCX-classic, where Markus at BaMaTech had to machine shaft extensions and button extensions with chassis mount button hardware, to be able to fit the controls through the front panel. I don't want to do all that which would all escalate the costs and then you customers wouldn't so much like the higher price. 

So I worked hard on it, and after a LOT of thought, I came up with a quite neat design solution (I think neat, anyway), using the additional little Controls PCB. The added cost is low since I am able to create the little Controls PCB in the cut-out area of the LCD PCB, into which the LCD module body fits. The LCD module, when installed in the LCD PCB, and the LCD PCB slid into the guide rails, is at exactly the right height to fit perfectly behind the rectangular cut-out in the enclosure. The Controls PCB fits in this cut-out so I do not have to manufacture one more PCB, so there is no increased cost for PCB making. I just add the 2x4-pin connectors and another 11mm spacer. The Controls PCB is at a height of 12.6mm above the main PCB, which itself is 11mm behind the LCD PCB (mounted in the extrusion guide rails). It is designed at this 12.6mm height so that the shafts of the gain control and the rotary encoder prodtrude exactly a perfect amount through the holes in the enclosure, to be able to fit the knobs on perfectly. 

It's VERY neat even if I do say so myself ;-)   

The sample QCX-mini enclosure arrived here yesterday evening. 7 boxes dumped on my doorstep unceremoniously. Contents: 300pcs QCX+ enclosure, and 1pcs QCX-mini prototype enclosure! 

Photos to come later today...

73 Hans G0UPL
http://qrp-labs.com


David Wilcox
 

 Hans,

You are “happy”!  Isn’t that the goal in life?  I was an old fashioned family doctor for 44 years and can honestly say “I never worked a day in my life”..... they paid me for having fun.  And ham radio surely helped on those days when life didn’t make sense.

Dave K8WPE

David J. Wilcox’s iPad

On Sep 25, 2020, at 4:52 AM, Hans Summers <hans.summers@...> wrote:


Hi David

Hi hi, not sure about that genius allegation... but yes I think it helps to "not know what can't be done". Because a lot of the time, it actually can be. I noticed this during my former 22 years in IT in banks... a lot of the time I came out with a different solution than the "proper" software guys with their posh Computer Science degrees - because nobody had told me what couldn't be done ;-)   My education was in physics. So I am not even a proper engineer at all in any discipline. But I'm happy :-) 

73 Hans G0UPL
http://qrp-labs.com

On Fri, Sep 25, 2020 at 11:44 AM David Wilcox via groups.io <Djwilcox01=yahoo.com@groups.io> wrote:
Hans, 

You are a genius!!!!  I think is it very important that you were self taught so you couldn’t be told by the university boffins that it “can’t be done”.  It’s engineers like you that keep the true spirit of ham radio alive for us “button pushers”.  Thank you for all you are doing for us.  With Covid we have cancelled all our normal swaps that I would have spent my radio money at.  I can’t wait until the mini’s are ready for shipment.  (also the QSX)....

Dave K8WPE since 1960 and still learning.  It’s “remembering” that I now have trouble with.

David J. Wilcox’s iPad

On Sep 25, 2020, at 4:19 AM, Hans Summers <hans.summers@...> wrote:


Hi Michael
 
I'm no designer and have never created my own PCB, so take my comment with that grain of salt. Why the separate board for the controls?

It is all to do with how the board assembly slots together, in minimum volume, and fits into the enclosure. In the design, the top PCB (the LCD PCB) slots into the PCB guide-rails in the top half of the enclosure extrusion. It slides in. It could not do so, if the controls were bolted to it, since the shafts of the controls and the buttons would absolutely prevent that.

So the controls must be fixed to the BOTTOM (main) PCB, where they can be plugged into the LCD PCB and bolted to it via 11mm spacers, AFTER the LCD PCB has been slid into the PCB guide rails. 

Yet, the controls cannot be installed on the main PCB itself, for two reasons:

1) They would take up too much space and I would not be able to make the QCX-mini as small as I wanted

2) The shafts of the controls and buttons would not reach through the enclosure. We would therefore have a situation the same as the QCX-classic, where Markus at BaMaTech had to machine shaft extensions and button extensions with chassis mount button hardware, to be able to fit the controls through the front panel. I don't want to do all that which would all escalate the costs and then you customers wouldn't so much like the higher price. 

So I worked hard on it, and after a LOT of thought, I came up with a quite neat design solution (I think neat, anyway), using the additional little Controls PCB. The added cost is low since I am able to create the little Controls PCB in the cut-out area of the LCD PCB, into which the LCD module body fits. The LCD module, when installed in the LCD PCB, and the LCD PCB slid into the guide rails, is at exactly the right height to fit perfectly behind the rectangular cut-out in the enclosure. The Controls PCB fits in this cut-out so I do not have to manufacture one more PCB, so there is no increased cost for PCB making. I just add the 2x4-pin connectors and another 11mm spacer. The Controls PCB is at a height of 12.6mm above the main PCB, which itself is 11mm behind the LCD PCB (mounted in the extrusion guide rails). It is designed at this 12.6mm height so that the shafts of the gain control and the rotary encoder prodtrude exactly a perfect amount through the holes in the enclosure, to be able to fit the knobs on perfectly. 

It's VERY neat even if I do say so myself ;-)   

The sample QCX-mini enclosure arrived here yesterday evening. 7 boxes dumped on my doorstep unceremoniously. Contents: 300pcs QCX+ enclosure, and 1pcs QCX-mini prototype enclosure! 

Photos to come later today...

73 Hans G0UPL
http://qrp-labs.com