Re: QCX PC board spots and mini holes
I presume that everybody knows this already - or not? Never fill up vias with solder! For RF to travel from one layer to the other it needs to travel via the surface. Filling vias with solder moves the surface to the wrong side and the way RF has to travel becomes far longer. On Freitag, 29. Oktober 2021 05:18:20 -03 David Wilcox K8WPE via groups.io wrote: My Thanks to you Hans and the others who helped me understand what
those dots and mini holes are for. A well designed kit and fun to
build and more fun to use. Thanks again.
Dave K8WPE
David J. Wilcox¡¯s iPad
On Oct 29, 2021, at 12:06 AM, Hans Summers <hans.summers@...>
wrote:
?
Hi Dave
Yes, they are vias.
QRP Labs kits use 2 layer boards. The vias are necessary to connect
the bottom and top layers of a signal trace as it is routed around
the board and sometimes it has to change from top to bottom in
order to avoid obstacles (other traces).
A common PCB layout technique is to run horizontal traces along one
side of the board and vertical traces along the other. I also start
my board layouts like this, then as the design progresses commit
may violations to make everything fit well. By "horizontal" and
"vertical" I mean, as you hold the board in front of you. X and Y
axis if you like... Not Z axis which is a hole drilled through the
board itself.
In some QRP Labs boards I do indeed use two sizes of vias, one for
signal connections between the two sides and one for ground plane
stitching. Sometimes I even use a larger size to via for power
lines which will be carrying relatively more current.
The minimum size via has to be chosen carefully because PCB
manufacturers have constraints like minimum drill bit size, and
minimum spacing between drill holes and nearby traces etc etc. As
you go smaller and smaller with feature sizes there comes a point
where the manufacture is still possible but the cost jumps
alarmingly and suddenly.
Some folk say you can't mix analog and digital domains on one PCB
such as in this case, digital circuits and sensitive RF circuits.
Others say you have to use 4 (at least) layer boards so you can
have signal layers on the top and bottom and supply planes in the
middle so have a good nearly continuous ground plane. Again more
layers more cost.
QRP Labs boards are 2 layer and successfully house digital and
analog domains on one board. This achieves compact low cost designs
that work very well. The vias are a crucial part of this.
I go to enormous trouble to ensure that the two sides of the board
both have as much ground plane as possible. Everywhere there isn't
a component pad or a signal trace. The vias link the two sides
together and placing them at intervals of not more than 0.1 inches
ensures that even up to VHF frequencies or higher, the PCB looks to
RF, like a near-perfect uninterrupted ground plane. I try to make
sure that if there are signal traces on one side of the board,
there is ground plane on the other side. Signal traces on one side
should cross, if they must, signal traces on the other side at
right angles, so as to minimize inductive or capacitative coupling.
I move components and traces around to try to be able to imagine
the dual-layer ground the way RF will see it and totally minimize
the size of any RF holes or gaps.
The result is good but it doesn't just happen by chance, it takes a
lot of hard work to layout the PCB.
73 Hans G0UPL
-------- Original message --------
From: GM7TFP <dbrankin@...>
Date: Fri, Oct 29, 2021, 1:22 AM
To: [email protected]
Subject: Re: [QRPLabs] QCX PC board spots and mini holes
Hi Dave
I suspect what you're seeing are indeed the vias (as described by
others), some of which may have been filled or semi-filled by the
soldermask ink due to viscosity/meniscus effect. It's overprinted
once the vias have been drilled.
Derek -- Eike Lantzsch ZP6CGE Ag. Molas Lopez Casilla de Correo 13005 01726 Asuncion / Paraguay SIPgate: +49-4131-9279632 Land-line: +595-21-553984 Cell-phone: +595-971-696909 Skype: eikelan Signal: zp6cge Eike Lantzsch WIRE: @eikelantzsch
|
Re: QCX PC board spots and mini holes
My Thanks to you Hans and the others who helped me understand what those dots and mini holes are for. ?A well designed kit and fun to build and more fun to use. ?Thanks again.
Dave K8WPE David J. Wilcox¡¯s iPad
toggle quoted message
Show quoted text
On Oct 29, 2021, at 12:06 AM, Hans Summers <hans.summers@...> wrote:
? Hi Dave
Yes, they are vias.?
QRP Labs kits use 2 layer boards. The vias are necessary to connect the bottom and top layers of a signal trace as it is routed around the board and sometimes it has to change from top to bottom in order to avoid obstacles (other traces).?
A common PCB layout technique is to run horizontal traces along one side of the board and vertical traces along the other. I also start my board layouts like this, then as the design progresses commit may violations to make everything fit well. By "horizontal" and "vertical" I mean, as you hold the board in front of you. X and Y axis if you like... Not Z axis which is a hole drilled through the board itself.?
In some QRP Labs boards I do indeed use two sizes of vias, one for signal connections between the two sides and one for ground plane stitching. Sometimes I even use a larger size to via for power lines which will be carrying relatively more current.?
The minimum size via has to be chosen carefully because PCB manufacturers have constraints like minimum drill bit size, and minimum spacing between drill holes and nearby traces etc etc. As you go smaller and smaller with feature sizes there comes a point where the manufacture is still possible but the cost jumps alarmingly and suddenly.?
Some folk say you can't mix analog and digital domains on one PCB such as in this case, digital circuits and sensitive RF circuits. Others say you have to use 4 (at least) layer boards so you can have signal layers on the top and bottom and supply planes in the middle so have a good nearly continuous ground plane. Again more layers more cost.?
QRP Labs boards are 2 layer and successfully house digital and analog domains on one board. This achieves compact low cost designs that work very well. The vias are a crucial part of this.?
I go to enormous trouble to ensure that the two sides of the board both have as much ground plane as possible. Everywhere there isn't a component pad or a signal trace. The vias link the two sides together and placing them at intervals of not more than 0.1 inches ensures that even up to VHF frequencies or higher, the PCB looks to RF, like a near-perfect uninterrupted ground plane. I try to make sure that if there are signal traces on one side of the board, there is ground plane on the other side. Signal traces on one side should cross, if they must, signal traces on the other side at right angles, so as to minimize inductive or capacitative coupling. I move components and traces around to try to be able to imagine the dual-layer ground the way RF will see it and totally minimize the size of any RF holes or gaps.?
The result is good but it doesn't just happen by chance, it takes a lot of hard work to layout the PCB.?
73 Hans G0UPL http://qrp-labs.com -------- Original message -------- From: GM7TFP <dbrankin@...> Date: Fri, Oct 29, 2021, 1:22 AM To: [email protected]Subject: Re: [QRPLabs] QCX PC board spots and mini holes Hi Dave
I suspect what you're seeing are indeed the vias (as described by others), some of which may have been filled or semi-filled by the soldermask ink due to viscosity/meniscus effect. It's overprinted once the vias have been drilled.
Derek
|
Mismatched antenna on Class-D & Class-E amplifier
Hi all,
For portable QRP operation I am planning to use a small semi-automatic antenna tuner
at the feedpoint* of my vertical. Manual tuning takes some time - set carrier, walk over to
the antenna and press a button ;-)
Wonder how Class-E in the QCX series and the Class-D in QDX behave in such
'mis-matched antenna' situation? Are there any limits ?
73's de OE1MWW
Wolfgang
*an antenna tuner belongs to the feedpoint of the antenna and not inside or right behind a rig.?
Otherwise the [cable+antenna] is 'tuned' (matched) = losses could be remarkable high!.
|
My new lab is ready for the QDX - :-)
?
73 - Knut
?
|
I received an email from QRP Labs on 26/10 that my? order?49973 was shipped - and an email from TNT that they had my order and had the status of "collecting"
The order is still in "Collecting" status at 29/10 - so it will take some time to reach me I'm afraid? ??
But one day - it will be here - hehe
73 - Knut?
|
Re: QDX new firmware 1_01
I'm very excited about the IQ output! Question: What is the frequency range on receive? Is it also limited to 3.4-14.5 MHz?
The Si5351A 90-degree quadrature mode will only work down to 2.93 MHz. That is therefore the lower limit.?
If the QDX receives a CAT command to operate outside the range 3.4 to 14.5 MHz, its response is: - Try to set the VFO to that frequency (which works OK as long as it is more than 2.93 MHz and less than something near 300 MHz)
- Disable transmit?until a within-range frequency setting command is received.
- Do nothing about what band pass and low pass filters are switched in.
- Continue to receive as normal
An important point is #3... so for example, if you want to receive the 17m band, make sure that you are on the 20m band already, until you send the CAT command to change the VFO to a frequency in the 17m band. This way, you are at least operating on the 20m LPF and 20m BPF so the attenuation through those won't be too drastic. It would not be so good for example, if you were on 3.5 MHz and you suddenly changed to 17m band... because then the 80m BPF and LPF would be left in, and severely attenuate everything on 17m.?
73 Hans G0UPL
|
Re: British electrical plugs
The breaker in the consumer unit protects the wiring to the household sockets, the fuse in the plug protects the item being powered. If the wiring was overloaded by a ring main for example there could be a fire in the partition wall.
230 V is an RMS value, so peak voltage achieved is about 330 V briefly, enough to ruin your day.
I know little about the argument regarding 110/230, but presumably for a given power consumption wiring at 110 V carried twice the current so has to have a larger cross section area, copper is not cheap these days.
|
Re: QCX Challenge - October 2021
On Thu, Oct 28, 2021 at 08:45 AM, KEN G4APB wrote:
Also my qcx mini rx decoder does not seem to work, I just get random letters even though I can read the signals by ear, are there any special settings needed to make this work?
Ken, there are some parameters in menu 5 that you can adjust (see Firmware manual ), but my decoder works fine with the default settings. Try tuning a good strong signal first and check if it works with that. A very weak or noisy signal will often not decode. 73, Luc ON7DQ??
|
Re: British electrical plugs
Oh, I like the landscaping circuit option for outbuilding, small loads...
For small IT loads, you could also consider PoE , particularly if we're taking a RaspPi node (with PoE hat)...
toggle quoted message
Show quoted text
-------- Original message -------- From: ajparent1/KB1GMX <kb1gmx@...> Date: 2021-10-28 15:36 (GMT-05:00) Subject: Re: [QRPLabs] British electrical plugs
Here in the USA (conus) 24V AC is typically used for landscape? lighting and uses direct bury cable.? The requirements to do that are trivial no electrician required.? Good care (fusing) is easy as low voltages so not require special hardware. I've used that to deliver power to the shed for LED lighting and also rectified regulated to run gear out there.? Its poer limite to about 100W as the transformer is small (100VA). Since I can choke the leads at both ends noise is not an issue. Allison ------------------------------- Please reply on list so we can share. No private email, it goes to a bit bucket due to address harvesting
|
Re: QCX PC board spots and mini holes
Hi Dave
Yes, they are vias.?
QRP Labs kits use 2 layer boards. The vias are necessary to connect the bottom and top layers of a signal trace as it is routed around the board and sometimes it has to change from top to bottom in order to avoid obstacles (other traces).?
A common PCB layout technique is to run horizontal traces along one side of the board and vertical traces along the other. I also start my board layouts like this, then as the design progresses commit may violations to make everything fit well. By "horizontal" and "vertical" I mean, as you hold the board in front of you. X and Y axis if you like... Not Z axis which is a hole drilled through the board itself.?
In some QRP Labs boards I do indeed use two sizes of vias, one for signal connections between the two sides and one for ground plane stitching. Sometimes I even use a larger size to via for power lines which will be carrying relatively more current.?
The minimum size via has to be chosen carefully because PCB manufacturers have constraints like minimum drill bit size, and minimum spacing between drill holes and nearby traces etc etc. As you go smaller and smaller with feature sizes there comes a point where the manufacture is still possible but the cost jumps alarmingly and suddenly.?
Some folk say you can't mix analog and digital domains on one PCB such as in this case, digital circuits and sensitive RF circuits. Others say you have to use 4 (at least) layer boards so you can have signal layers on the top and bottom and supply planes in the middle so have a good nearly continuous ground plane. Again more layers more cost.?
QRP Labs boards are 2 layer and successfully house digital and analog domains on one board. This achieves compact low cost designs that work very well. The vias are a crucial part of this.?
I go to enormous trouble to ensure that the two sides of the board both have as much ground plane as possible. Everywhere there isn't a component pad or a signal trace. The vias link the two sides together and placing them at intervals of not more than 0.1 inches ensures that even up to VHF frequencies or higher, the PCB looks to RF, like a near-perfect uninterrupted ground plane. I try to make sure that if there are signal traces on one side of the board, there is ground plane on the other side. Signal traces on one side should cross, if they must, signal traces on the other side at right angles, so as to minimize inductive or capacitative coupling. I move components and traces around to try to be able to imagine the dual-layer ground the way RF will see it and totally minimize the size of any RF holes or gaps.?
The result is good but it doesn't just happen by chance, it takes a lot of hard work to layout the PCB.?
73 Hans G0UPL http://qrp-labs.com
-------- Original message -------- From: GM7TFP <dbrankin@...> Date: Fri, Oct 29, 2021, 1:22 AM To: [email protected]Subject: Re: [QRPLabs] QCX PC board spots and mini holes Hi Dave
I suspect what you're seeing are indeed the vias (as described by others), some of which may have been filled or semi-filled by the soldermask ink due to viscosity/meniscus effect. It's overprinted once the vias have been drilled.
Derek
|
Re: How do the big companies make radios?
in the niche market of point to point microwave transport, filters are hand tuned as each filter has different band pass/band reject characteristics depending upon the type of radio being produced. The filters are tuned with a spectrum analyzer connected so adjustment of each section can be measured instantaneously. After final adjustments are made, tne filter response is recorded in the spec-an as part of the filter pedigree.
73, kevin kc6pob
Hi,
I know we have many very experienced? people in this group.
I see each kit usually needs the LPF filters tweaked for maximum output, etc.
I look under the hood of ICOM, Yaesu, Harris radios which are made in the thousands, and there are one helluva lot of coils, etc.?
How do they build radios in a manner that makes money? This must be a challenge.?
I have never seen a radio factory, and have no experience in the radio industry,
please if you would explain to me, how they do it, if you survived, (hi hi).?
Is there a Hans at the end of the line?
.
How did you do it??
|
Re: How do the big companies make radios?
HF is particularly difficult because it covers many octaves.? And in many cases can only be done by hand. Many physically big part filters need to be used to cover 2-30 MHz because filters work by fractions of octaves covered and the lower the frequency the smaller the bandwidth in an octave.?
?Though machine produced parts will be more repeatable than hand wound parts.?
Most higher end commercial designs are designed to be trimmed electronically and automated test equipment actually tweaks the circuits iteratively, which then has those adjustments saved to non volatile memory.? ?This can give you very flattened out response if done correctly.?
A good RF test station to do this is often in between $500k and $1M.? Plus guys like me have to write the secret sauce software to make them do their job.? ? At the same time you have to look for failures as well.?
?I did that sort?of thing as a test engineer most of the last 12 years or so when I was at Collins (left 4 years ago), but in general that is how anything that needs to be tuned works in industry.? Not just RF.
That approach actually allows sloppier designs to work as well or better than fancy designs and usually is more reproducible.? ?You have very expensive equipment cancel out the non linearities in the circuits over several points.?
Fred W0FMS
This question occurred to me because I've been watching the Antique Wireless, History of Radio
series on YouTube. Fascinating...but the Ham Radio industry is littered with failed companies.?
My beloved TenTec went under. It's gotta a be a tough business.?
|
Re: How do the big companies make radios?
First Japan, and then China, crushed the US ham equipment makers. The speed with which old-line companies like Hallicrafters, National, E F Johnson, Clegg, etc crashed in the late 1960/early 1970s was breathtaking.
What amazes me is that the Chinese saw the amateur radio market big enough to develop impressive, low-cost products for it. Now we see Russian companies getting in, too.
toggle quoted message
Show quoted text
This question occurred to me because I've been watching the Antique Wireless, History of Radio
series on YouTube. Fascinating...but the Ham Radio industry is littered with failed companies.?
My beloved TenTec went under. It's gotta a be a tough business.?
|
Re: How do the big companies make radios?
This question occurred to me because I've been watching the Antique Wireless, History of Radio
series on YouTube. Fascinating...but the Ham Radio industry is littered with failed companies.?
My beloved TenTec went under. It's gotta a be a tough business.?
|
Re: How do the big companies make radios?
I don't remember toroids in my once owned low end Kenwood. Far east has firms that specialize in low cost inductors. My k2 has toroids, who knows how many I have wound, likely between 100 and 200. Ubitx has a team of winders, glad our community gave them some livelihood. Well I imagine our mid and low price factory rigs use machine wound coils. Radio technology through the decades is quite cool. You should see the millimeter wavelength stuff, much of this is wafer fab. Antique radios a different kind of cool.?
73 Curt
|
How do the big companies make radios?
Hi,
I know we have many very experienced? people in this group.
I see each kit usually needs the LPF filters tweaked for maximum output, etc.
I look under the hood of ICOM, Yaesu, Harris radios which are made in the thousands, and there are one helluva lot of coils, etc.?
How do they build radios in a manner that makes money? This must be a challenge.?
I have never seen a radio factory, and have no experience in the radio industry,
please if you would explain to me, how they do it, if you survived, (hi hi).?
Is there a Hans at the end of the line?
.
How did you do it??
|
Re: AA Battery Power for QRP Labs gear
The old tube rigs that are considered desirable have high price tags. But less popular ones can be affordable, at least if you find them from places other than eBay. At NEAR-fest recently, I saw a physically clean FTDX-101EE that was tagged $50; all the knobs worked, the tubes were all present, and the vernier dial worked. I don't know how close to working order that radio was; if it's been on a shelf for a while it's likely to require replacement of the electrolytic capacitors at a minimum,?and possibly some tubes, transistors in the VFO, and semiconductor diodes in the power supply. I didn't buy it because I'm not looking to take on a tube project right now, but somebody did.
The premium brands, Collins and Drake, still fetch a pretty penny. And Heathkits are expensive because of nostalgia, except for the ubiquitous linear?amplifiers. (The SB-200 and SB-220 are usually reasonably priced when you encounter one. Linear nostalgia seems to be reserved for Alpha, which will cost an arm and a leg if you see one for sale.) But Japanese tube transceivers from the 70s can be quite affordable,?at least if you buy?them at a hamfest. Besides that Yaesu?and a couple more that I saw at the same?event,?I have seen other rigs such as the Tempo One with affordable price tags. Radios from that era are generally maintainable; you don't need any impossible to get parts. (The tubes aren't made any more but there are plenty of them out there.) The older rigs that you really have to approach with caution are the somewhat newer solid state ones with frequency synthesizers; make sure those are in working order before you buy because the ICs in them (and perhaps the final output transistors as well) are likely to be unobtainable.
Or if you don't want to deal with any of that hassle, buy something from Hans. :) It's true that if it breaks 50 years from now you probably won't be able to fix it because the?ICs will have long since become history, but it's not 50 years from now yet! (Request to Hans: please release the firmware as open source when you retire, so at least we'll have a chance of being able to fix those 50 year old radios. Not that I'll live that long, but future hams might care.)
On Thu, Oct 28, 2021 at 12:17 PM Mike Perry, WA4MP < editor@...> wrote: Agreed. Just look at what 1960-era Heathkit tube transceivers sell for. I¡¯m looking at several on eBay in the $450 range. Even the ¡°parts only¡± ones are around $150. But for the collector¡¯s urge, new hams could be purchasing them for under $100. Even an ancient Heathkit GR-91 receiver is $99.99. I owned one. It was my first SW radio¡ªa $40 Christmas present. I know for a fact that no one is buying it for the quality of reception. Heathkit simply took the design for a cheap table radio and extended its coverage into SW.
That¡¯s why what Hans is doing is wonderful. He¡¯s a one-man Heathkit, creating inexpensive kits that both new and experienced hams can enjoy.
¡ªMike Perry, WA4MP
> On Oct 28, 2021, at 10:56 am, Julian N4JO <n4jo@...> wrote:
>
> Yes, that would be a great idea, but many hams are collectors who get very emotionally attached to their stuff, which keeps resale value absurdly high, keeping such equipment out of reach of newcomers. You only have to look at the preposterous resale value of Morse keys to see this.
|
I'd be interested to hear your calibration routine for "gently" ;-)
--
Julian, N4JO.
|
Re: QCX PC board spots and mini holes
Hi Dave
I suspect what you're seeing are indeed the vias (as described by others), some of which may have been filled or semi-filled by the soldermask ink due to viscosity/meniscus effect. It's overprinted once the vias have been drilled.
Derek
|
My QDX arrived from Sedalia, MO today.? I have been to Sedalia many times.? They used to be famous for a hamburger topped with peanut butter called, the "Gooberburger."??
|
David Wilcox K8WPE
Knut Steinar Fremme OE4KSF
ON7DQ Luc
Wayde VA3NCA/AD2GX
Fred Spinner
Shirley Dulcey KE1L
Julian N4JO