|
Hi has anyone found some SM 47uH inductors than can replace the original ones that self resonate in the HF band? or could I put a cap on top to move the self resonance down to MW? I would rather replace the SM ones than fit the leaded ones supplied?
73
Andrew
|
I don't know the direct answer, but if you're wanting to fit SMD parts rather than the leaded parts because it looks hard- it's not. Not at all. I'll be posting a video about it today or tomorrow. I documented my complete build.? -- Ryan Flowers - W7RLF
|
I don't know the answer either.?
I think the only answer actually is trial and error... If you want to substitute components other than those supplied, be sure to carefully check the harmonic output on all four bands using a good spectrum analyzer.
73 Hans G0UPL http://qrp-labs.com
-------- Original message -------- From: Ryan Flowers <geocrasher@...> Date: Mon, Nov 1, 2021, 7:43 PM To: [email protected]Subject: Re: [QRPLabs] QDX Surface Mount inductors I don't know the direct answer, but if you're wanting to fit SMD parts rather than the leaded parts because it looks hard- it's not. Not at all. I'll be posting a video about it today or tomorrow. I documented my complete build.?
--
Ryan Flowers - W7RLF
|
The self resonant frequency is often on the datasheet for these inductors. However, but as Hans mentions, you have to also consider the harmonics as well. Anyway, If I were to experiment I would start with the datasheet (example attached) and then really prove it by testing the actual result.
Good Luck & 73,
I don't know the answer either.?
I think the only answer actually is trial and error... If you want to substitute components other than those supplied, be sure to carefully check the harmonic output on all four bands using a good spectrum analyzer. -------- Original message -------- From: Ryan Flowers < geocrasher@...> Date: Mon, Nov 1, 2021, 7:43 PM To: [email protected]Subject: Re: [QRPLabs] QDX Surface Mount inductors I don't know the direct answer, but if you're wanting to fit SMD parts rather than the leaded parts because it looks hard- it's not. Not at all. I'll be posting a video about it today or tomorrow. I documented my complete build.?
--
Ryan Flowers - W7RLF
|
I was thinking that it might be more trouble than it is worth when you mentioned digging for new inductors.? If you are worried about the leaded ones physical security to the board, once you get it working, simply epoxy them down.? ?That's an industry fix on prototypes and for many "white wire" changes.?
Fred W0FMS?
The self resonant frequency is often on the datasheet for these inductors. However, but as Hans mentions, you have to also consider the harmonics as well. Anyway, If I were to experiment I would start with the datasheet (example attached) and then really prove it by testing the actual result.
Good Luck & 73, I don't know the answer either.?
I think the only answer actually is trial and error... If you want to substitute components other than those supplied, be sure to carefully check the harmonic output on all four bands using a good spectrum analyzer. -------- Original message -------- From: Ryan Flowers < geocrasher@...> Date: Mon, Nov 1, 2021, 7:43 PM To: [email protected]Subject: Re: [QRPLabs] QDX Surface Mount inductors I don't know the direct answer, but if you're wanting to fit SMD parts rather than the leaded parts because it looks hard- it's not. Not at all. I'll be posting a video about it today or tomorrow. I documented my complete build.?
--
Ryan Flowers - W7RLF
|
It would be interesting to know the context of how and where part is used to determine if it's a problem. Presumably the qdx works already is what I am thinking. In a few apps designers I observed in the past wanted to operate near self resonance, but that may not be intended here. Don't readily jump to conclusion there is a problem yet, in particular if part isn't a sole bias choke.?
Curt
|
Hi Curt
This all happened in January 2020 or so which seems like a whole other era ago. So my memory is a bit hazy. But when the PCBs arrived with the SMD inductors on, I noticed that the harmonic content of the output?was worse than it was supposed to be. Somehow the SMD inductors used on the board manufacture were not the same as the ones I had used in prototyping. The problem was simply solved by changing to axial leaded (through-hole) 47uH inductors of the type used in other QRP Labs kits.?
The QDX manual states that the 47uH inductors should be removed and replaced with the through-hole ones supplied. Then the harmonic levels will be correct (low).?
If someone wants to NOT use the through-hole inductors and instead change the SMD inductors to other SMD inductors, that's fine, but now you are in experimental territory so you had better experiment - that is to say, look at the harmonic content on a decent spectrum analyzer and make sure it is OK...?
73 Hans G0UPL
It would be interesting to know the context of how and where part is used to determine if it's a problem. Presumably the qdx works already is what I am thinking. In a few apps designers I observed in the past wanted to operate near self resonance, but that may not be intended here. Don't readily jump to conclusion there is a problem yet, in particular if part isn't a sole bias choke.?
Curt
|
Hi all,
Thanks for your responses, I do have have a number of spectrum analysers, and I have a surface mount work station, I can remove the inductors with a hot air pencil, and fit different ones with ease. After I have built the kit with originals parts, I will measure the harmonic content, let's assume the power is 37dBm, then 60dBc would be -23dBm (25mW), I would like to see harmonic content -36dBm (250nW), or -54dBm (3.98nW), this may be a tall order?
I have HF coupler and some high pass filters and HF notches, for HF so can make the measurements. What spec are we supposed to meet?
Yaesu FTDX 10 is quoted at -50dBc for 100 Watts! on HF!
Hans, what was the original manufacture part number of the inductors you were using? I would like to look up the spec sheet.
RF design, does have a tendency to bite you!. all part of the fun
73 G8UUG
|
Hello Andrew
You asked:
"Hans, what was the original manufacture part number of the inductors you were using?"
I am attaching to this post, a photograph of the SMD reel label. So that should answer your question...?
Before any smart person notices that the date code is 2020-09-08 which is AFTER I manufactured the QDX kits (around January 2020)... yes, that is because this is the current reel of 3000pcs of this inductor that we are using. We are also using the reel for other kit manufacturing too such as the QCX-mini. But it's always the same supplier and the same part number as shown on this photo.?
73 Hans G0UPL
On Thu, Nov 4, 2021 at 4:25 PM Andrew Lenton < a@...> wrote: Hi all,
Thanks for your responses, I do have have a number of spectrum analysers, and I have a surface mount work station, I can remove the inductors with a hot air pencil, and fit different ones with ease. After I have built the kit with originals parts, I will measure the harmonic content, let's assume the power is 37dBm, then 60dBc would be -23dBm (25mW), I would like to see harmonic content -36dBm (250nW), or -54dBm (3.98nW), this may be a tall order?
I have HF coupler and some high pass filters and HF notches, for HF so can make the measurements. What spec are we supposed to meet?
Yaesu FTDX 10 is quoted at -50dBc for 100 Watts! on HF!
Hans, what was the original manufacture part number of the inductors you were using? I would like to look up the spec sheet.
RF design, does have a tendency to bite you!. all part of the fun
73 G8UUG
|
Hans,
Boa, you have still 3000 SMD inductors on stock... :o)
For the Rev 2 boards, do you plan to use the same SMD inductors ?
And during assembly they have to be removed and replaced by the
extra delivered through-hole ?
73's de OE1MWW
Wolfgang
|
In Rev 2, I used through-hole inductors for those 4 and deleted the SMDs. So the SMDs don't need to be removed. Just install the wire-ended ones.?
Given the high demand for QDX I don't want to go through samples testing to check self-resonance. We have a known solution that works: use through-hole 47uH inductors. For now, that works...?
73 Hans G0UPL
On Mon, Nov 8, 2021 at 12:51 PM Wolfgang OE1MWW < oe1mww@...> wrote: Hans,
Boa, you have still 3000 SMD inductors on stock... :o)
For the Rev 2 boards, do you plan to use the same SMD inductors ?
And during assembly they have to be removed and replaced by the
extra delivered through-hole ?
73's de OE1MWW
Wolfgang
|
As an interim solution to get past the high demand stress I would agree.? You do need to note that this will add a complexity that some intermediate level builders with have issues in the installation.? I consider myself a somewhat advanced builder and it was a challenge for me.? The board is very small and cramped, so this type of modification is a challenge.?
I apologize if I was unclear.?
Rev 2 PCB uses through-hole axial molded 47uH inductors, as were supplied with the Rev 1 boards. However on the Rev 2 PCB, there are through-hole pads on the board for these, and they take the place of the SMD parts which are deleted from the design. In Rev 2, there is therefore no need to remove SMD components and replace them with tacked-on through-hole inductors. Installation?of the four through-hole inductors is as easy as any other?through-hole components.?
73 Hans G0UPL
|
Hans,
This is a totally acceptable solution for me.? Glad that you cleared up my understanding.? Now to wit for the availability of the v2 kits.
73
Evan
AC9TU
|
Thank Hans,
Kit now built worked first time and software updated, however, I need to learn how to fly WSJT-X , I get a full 5 Watts on TX with 2:2 as suggested for 12V operation, I will plot the harmonics, and send you, I will try adding some caps across the inductors to bring them self resonance to 1MHz and see what happens first then replace them with your wired inductors, if unsuccessful.
PS windows 10 was tiresome with the com port, first it work then it lost it, and took awhile to get back.
What is the duty cycle at 5 Watts?
73
Andrew
|
What is the duty cycle at 5 Watts?
It's hard to say.?
I think running WSPR on a 100% cycle - for example setting up WSJT-X to cycle through 80-40-30-20, may be pushing things a bit too hard. The thing is not 100% efficient and it's a tiny enclosure, so the temperature starts rising.?
FT8 has a duty cycle something under 50%, each transmission being in alternate slots and lasting 12.6 seconds, so 42%.?
On the other hand - I have been having fun on JS8Call without any trouble, and here, the underlying protocol is the same as FT8, but unlike FT8 where you always transmit in alternate 15 second slots, in JS8Call you can transmit in multiple consecutive slots, if you type a long message it will keep transmitting chunks of it sequentially until it gets through to the end of your message. During that time the duty cycle is 84%. Of course if you go to slow mode the slots are twice as long (30 seconds).?
Other factors will influence what duty mode will be acceptable without damage - such as any differences arising from running a 12V QDX vs a 9V QDX; different bands have different efficiencies, attenuation in the LPF, SWR of the antenna, ambient temperature, and and and... And of course a sustained very high duty cycle may not cause failure immediately but may hasten early failure.?
Sorry I don't have a more accurate answer to the question.?
73 Hans G0UPL
|
Andrew Lenton
Fred Spinner