QCX-mini IC9 3rd time not a charm #problem #40m #qcxmini #opamp


RT Takezo
 

Sadly IC9 in my QCX-mini failed for a 2nd time.

Even more sadly, while replacing the opamp a 3rd time, the pcb pad for pin 7 came off.

I cannot see any visible trace to even attempt connection (jewelers loupe & scope didn't help).  Is it time for her for the solder pot (to recover what components I can) then into the bin of lost hopes?

73
Scott
JS6UEY


Mont Pierce KM6WT
 

On Tue, Mar 8, 2022 at 06:46 PM, RT Takezo wrote:
Even more sadly, while replacing the opamp a 3rd time, the pcb pad for pin 7 came off.

I cannot see any visible trace to even attempt connection (jewelers loupe & scope didn't help).  Is it time for her for the solder pot (to recover what components I can) then into the bin of lost hopes?
Not at all...

Look at the Schematics (black and white Hi-Res Schematics Click Here)  and find the component that pin connects to.  Then solder a small wire from IC9 pin to the component that the missing pad used to connect to.

NOTE:  The schematic in the link above is for the QCX+, however ALL QCXs share the same schematics.  On the mini though, all (most) the parts are SMDs and some have been exchanged with compatible parts with lower power consumption.

Which pin's pad on IC9 PCB came off?

73
km6wt


RT Takezo
 

Thanks for the quick response.  I had thought this possible, albeit challenging, as C20 (to which IC9's pin 7 connects by the manual) is on the underside of the board.  I had simply felt it odd that I could not see any trace contact where the pad was.


RT Takezo
 

Well, we can close this thread.  This mini is going in the bin.

After connecting IC9 pin 7 to C20 underneath the IC voltage measured off on all pins except 8.  Then when replacing the IC with yet another, more pads came off.....


Andy
 

If I send you the shipping costs will you send it to me?

73
Andy


From: QRPLabs@groups.io <QRPLabs@groups.io> on behalf of RT Takezo via groups.io <rttakezo2000@...>
Sent: Tuesday, March 8, 2022 10:17 PM
To: QRPLabs@groups.io <QRPLabs@groups.io>
Subject: Re: [QRPLabs] QCX-mini IC9 3rd time not a charm #problem #qcxmini #40m #opamp
 
Thanks for the quick response.  I had thought this possible, albeit challenging, as C20 (to which IC9's pin 7 connects by the manual) is on the underside of the board.  I had simply felt it odd that I could not see any trace contact where the pad was.


Mont Pierce KM6WT
 

On Tue, Mar 8, 2022 at 10:17 PM, RT Takezo wrote:
I had thought this possible, albeit challenging, as C20 (to which IC9's pin 7 connects by the manual) is on the underside of the board. 
But there's a much closer component it also connects to.  IC9 pin 7 AND pin 6.
So just attach pin 6 & 7 together...

     
( the bottom line in this pic goes off to the left to C20 )


km6wt



Mont Pierce KM6WT
 

On Tue, Mar 8, 2022 at 11:33 PM, RT Takezo wrote:
Well, we can close this thread.  This mini is going in the bin.

After connecting IC9 pin 7 to C20 underneath the IC voltage measured off on all pins except 8.  Then when replacing the IC with yet another, more pads came off.....
I just sent you a private email. 

73
km6wt


RT Takezo
 

I would like to publicly thank Mont (KM6WT) for talking me off the proverbial ledge.  His encouragement and advice over the last couple days allowed me to get my mini back from the grave.

For those curious; it appears there are no true 'in board' traces for IC9's pads 1 & 7 (Hans would need to weigh in to confirm), but rather fine 'surface traces' from pads 1->2 and 6->7.  With all the mucking about with opamp removal/install/removal/repeat not only did I loose pads 1 & 7 but also those small fine surface traces.  In the end, repair only required me to install yet another opamp, but with jumpers installed across pins 1&2 and 6&7.  All appears fully functional at this point.

BZ and a big TU to Mont
73
Scott
JS6UEY


Hans Summers
 

Hello Scott
 
For those curious; it appears there are no true 'in board' traces for IC9's pads 1 & 7 (Hans would need to weigh in to confirm), but rather fine 'surface traces' from pads 1->2 and 6->7.  With all the mucking about with opamp removal/install/removal/repeat not only did I loose pads 1 & 7 but also those small fine surface traces.  In the end, repair only required me to install yet another opamp, but with jumpers installed across pins 1&2 and 6&7.  All appears fully functional at this point.

Glad to hear it is all working!

I'm not sure what you mean by "true in board traces". 

The QCX-mini (and other QRP Labs boards) are all two-sided boards having a top copper layer and a bottom copper layer. There are no additional hidden layers in the middle (such as would be present on a four or six etc layer board). If what you mean by "surface traces" is traces on the top surface of the PCB, yes, these exist in multiple places. If you look in the early pages of the QCX-mini manual you'll see a trace diagram, with red traces (top layer) and blue traces (bottom layer) shown. 

There are no circumstances where a PCB cannot be made to work, even if traces get removed; the only question is how patient you are, and how much beauty it loses in the process!

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


RT Takezo
 

Hans

Thank you for taking to the time to jump in.  Sorry if I created confusion with my terminology; I was uncertain of the correct (hence the quotes).  When the IC9 pads 1 & 7 came off, I had expected, and feared, there to be traces that flowed within the board (bottom copper layer?) as the trace diagram had both red and blue traces at both pad positions; however, that did not appear to be the case.  Instead it seemed as though there was only the hairline surface traces jumpering pads 1&2 and 6&7 (visible in red on the trace diagram).  I suspect I am simply misunderstanding the trace diagram.

73
Scott
JS6UEY


Hans Summers
 

Hi Scott

It can be quite tricky, understanding what goes where on the diagram. My PCB layouts are all only 2-layer - however quite compact and complex, all the traces are placed by hand by me - auto-route does a terrible job on compact boards and knows nothing about RF or AF - and I spend a lot of time optimizing the PCB layout by hand to get the best possible groundplane coverage on each side. 

I've attached a zoomed in image of the traces around IC9. The groundplanes are NOT shown in this diagram. Yes it's confusing because it's like an X-ray image, it shows both the top layer (red) and the bottom layer (blue) and the silkscreens, component outlines, etc. 

Anyway you can see that all the pads EXCEPT pin 1 and 7, do  have through-hole vias linking them to traces on the bottom side of the board. If these pads come off, they are more difficult to repair due to the fact that you also have these vias between the sides. HOWEVER, on the plus side, the vias also anchor the pads more securely so those pads are a lot less likely to come off. As you found :-)   

Pins 1 and 7 don't have any vias under them, but do have the top layer traces connecting them to pin 2 and 6 respectively. If pads 1 or 7 come off, it's easy... because all you need is a blob of solder bridging pins 1-2 or 6-7. Don't ask me how I know. Actually bridges of solder are normally the enemy and happen to me all the time anyway, hence the little reel of desoldering braid is never far from my reach here on the workbench, however messy it gets. 

Finally... note that officially it is a little naughty of me to put vias under component pads. Officially, I broke the rules. BUT, I do it anyway because it helps me achieve a more compact layout and more groundplane coverage (good for RF performance) and because the component dimensions (0603, SOIC etc) are fairly huge anyway as far as SMD technology is concerned. I wouldn't do it on the truly microscopic components in a cellphone for example. I've never had any problems with the SMD assembly due to my breaking this vias-under-pads rule. 

On Thu, Mar 10, 2022 at 11:16 AM RT Takezo via groups.io <rttakezo2000=yahoo.com@groups.io> wrote:
Hans

Thank you for taking to the time to jump in.  Sorry if I created confusion with my terminology; I was uncertain of the correct (hence the quotes).  When the IC9 pads 1 & 7 came off, I had expected, and feared, there to be traces that flowed within the board (bottom copper layer?) as the trace diagram had both red and blue traces at both pad positions; however, that did not appear to be the case.  Instead it seemed as though there was only the hairline surface traces jumpering pads 1&2 and 6&7 (visible in red on the trace diagram).  I suspect I am simply misunderstanding the trace diagram.

73
Scott
JS6UEY


HA3HZ
 

Hi Scott,
just a small comment about soldering.
If you use a temperature-controlled soldering iron, it's a good idea to experiment with a small panel that is not in use to see what the highest temperature the panel can withstand without peeling off the copper foil (pad). The temperature at which you soldered and soldered may be too high. The melting point of the solder can also be decisive. The smaller the parts you solder, the more care must be taken to handle them.
I just wanted to draw your attention to the details of the solder.
--
Gyula HA3HZ


Hans Summers
 

The trouble I have here, is that the most junior lab tech has this habit, when visiting QRP Labs HQ R&D division laboratory, of reaching up and turning the knob on my soldering iron to the max. Which is 480C (900F).  He finds it very funny if I catch him. But I don't always notice immediately until everything starts getting charred when I try to soldering anything.


On Thu, Mar 10, 2022 at 12:39 PM HA3HZ <gyula@...> wrote:
Hi Scott,
just a small comment about soldering.
If you use a temperature-controlled soldering iron, it's a good idea to experiment with a small panel that is not in use to see what the highest temperature the panel can withstand without peeling off the copper foil (pad). The temperature at which you soldered and soldered may be too high. The melting point of the solder can also be decisive. The smaller the parts you solder, the more care must be taken to handle them.
I just wanted to draw your attention to the details of the solder.
--
Gyula HA3HZ


Donald S Brant Jr
 

There are also low melting point solders which are used to lower the melting point of existing solder to ease part removal, one is ChipQuik:  https://www.chipquik.com/store/index.php?cPath=200
Another is to apply background heat with a hot plate or hot air gun (or hair dryer) so that you do not have to apply so much heat to flow the solder.  I use a hot plate set to 150-160C.
72, Don N2VGU


George Korper
 

Donald, thanks for the link. Neat stuff!


Mont Pierce KM6WT
 

On Thu, Mar 10, 2022 at 01:39 AM, HA3HZ wrote:
just a small comment about soldering.
If you use a temperature-controlled soldering iron, it's a good idea to experiment with a small panel that is not in use to see what the highest temperature the panel can withstand without peeling off the copper foil (pad). The temperature at which you soldered and soldered may be too high. The melting point of the solder can also be decisive.
To add to HA3HZ's comment, the methods "some" use to remove a part seem a bit scary to me...  I'm not saying they don't work, but certainly not the methods I would use with these tiny parts.

I'm sure I'd be in deep trouble If "I" attempted to "cut" the leads with an X-Acto knife.

If I tried to use "hot-air" gun to remove a part, I'd probably get too much heat, and parts would fall off the opposite side. 

On the other hand, the "ChipQuiK" method is probably the best option I've heard discussed in the group.

On Thu, Mar 10, 2022 at 04:50 AM, Donald S Brant Jr wrote:
There are also low melting point solders which are used to lower the melting point of existing solder to ease part removal, one is ChipQuik:  https://www.chipquik.com/store/index.php?cPath=200
Thanks Donald for the link !!



You can probably tell, SMD rework kind of scares me, even though I do enjoy "building" kits with SMDs.  Watching the solder melt under my hot-air gun, and seeing the parts snap into place, it's pretty cool.

So... my hats off to all of you with the courage to do your own SMD Rework !!


Sincerely,
Mont


Giuseppe Marullo
 

"Officially, I broke the rules. BUT, I do it anyway because it helps"

Some of them can be bent. Others can be broken.

Hans "Morpheus" Summers