Fw: [QRPLabs] QMX new build showed up as USB but unresponsive after firmware transfer
First mistake Copy/paste error Q105 should have been D108!
?
73
?
Ian G4GIR
?
toggle quoted message
Show quoted text
----- Original Message -----
Sent: 24/02/2025 10:21:24
Subject: Re: [QRPLabs] QMX new build showed up as USB but unresponsive after firmware transfer
I cant find any reference to the SMPS 5.6V PCB and it's different configurations, I see in the latter type boards C104 has been removed. However all the Assembly manuals show the original board configuration, including listing C104.
?
Having a few problems with the boards I have created a 5.6v PCB searchable document attached. Hope this helps let me know if there are any mistakes please.
?
73
?
Ian G4GIR
?
----- Original Message -----
Sent: 23/02/2025 22:53:38
Subject: Re: [QRPLabs] QMX new build showed up as USB but unresponsive after firmware transfer
On Sun, Feb 23, 2025 at 09:41 PM, Florin wrote:
If I'm not mistaken the only SMPS available for purchase separately were for REV 1 boards, Florin, no, the available spare SMPS boards are good for all QMX revisions. And these boards are with some changed elements compared to REV 1 to enhance the robustness.
?
And congratulations to your new build!!!
?
73 Ludwig
|
Re: QMX new build showed up as USB but unresponsive after firmware transfer
I cant find any reference to the SMPS 5.6V PCB and it's different configurations, I see in the latter type boards C104 has been removed. However all the Assembly manuals show the original board configuration, including listing C104.
?
Having a few problems with the boards I have created a 5.6v PCB searchable document attached. Hope this helps let me know if there are any mistakes please.
?
73
?
Ian G4GIR
?
toggle quoted message
Show quoted text
----- Original Message -----
Sent: 23/02/2025 22:53:38
Subject: Re: [QRPLabs] QMX new build showed up as USB but unresponsive after firmware transfer
On Sun, Feb 23, 2025 at 09:41 PM, Florin wrote:
If I'm not mistaken the only SMPS available for purchase separately were for REV 1 boards, Florin, no, the available spare SMPS boards are good for all QMX revisions. And these boards are with some changed elements compared to REV 1 to enhance the robustness.
?
And congratulations to your new build!!!
?
73 Ludwig
|
Re: QCX 50W PA multibanding?
#50wpa
?As for the interactions, I would not expect that "just appending" a lower-frequency section would "just work" but it sort of does. I.e. leaving the original LPF as-is and just plugging in an extra Pi-network designed for the lower frequency (input capacitance of the "plugin" adds to the output capacitance of the "original") does not destroy the general frequency performance (at least not too much, see below).
?
My simulation of an extra LPF stage for 3.6 MHz appended to a 7 MHz LPF shows that it works fine if it is designed as elliptic, i.e. with parallel resonant circuit for 7.2 MHz in signal path. However, if I omit the capacitor from the parallel resonant circuit, the combination sort of works as LPF for 3.6 MHz (it goes through) but 7.2 MHz suppression is not great (less than or around 20 dB) and there is a peak (very low suppression, less than 10 dB) around the critical frequency of the original LPF (which seems to be at around 9.0 - 9.2 MHz, at least theoretically, based on available data).
?
As soon as I get some time and at least two BNC connectors I will test it :) Presumably, for 5.3 MHz with elliptic filter tuned to suppress 10.6 MHz it should be more or less the same story.
?
I don't have a blog, so I have to think where to post the simulation images and how.
?
73 Jindra
?
On Mon, Feb 24, 2025 at 12:24 AM, ok4rm wrote:
Yep, in RX path it has almost the same effect. Theoretical insertion loss of the standalone LPF (with ideal C and L) calculated by LTspice is 0.01 dB worse than the big combo. So in reality it will be the same in and out.
Jindra?
?
On Mon, Feb 24, 2025 at 12:11 AM, ok4rm wrote:
Hmm, it looks interesting:
- to make an "appendix LPF" that would work for both 3.6 and 5.3 MHz does not seem to be possible
- however, adding an elliptical PI-network C-L-C, sort of an "extra stage", seems to work for 3.6 MHz:?
?
first shunt capacitance: 727 pF (680+47), together with the 270 pF output capacitance in the PA it makes 997 pF
inductance: 1.77 uH (19 turns on T50-2), and 270 pF in parallel with it makes a parallel resonant circuit at 7.2 MHz
output capacitance: 727 pF (670+47)
?
This combination should suppress 7.2 MHz by 50-60 dB (resonance is critical and there might be tolerances in capacitor values as well as winding). Optimum values are slighly different by 2 to 3 pF but impact is insignificant (order of 0.001 dB according to simulation).
?
Now I have to check how will this appendix behave standalone in the "opposite direction", i.e. transceiver in RX, PA is bypassed and RX signal goes only through this LPF. But I think it should be fine. Tomorrow or on Tuesday I will try this. Unfortunately I think I have no BNC connectors in the drawer.
?
73 Jindra
?
On Sun, Feb 23, 2025 at 10:26 PM, ok4rm wrote:
I like this idea, Don. I built the PA for 7 MHz (for emcomm this would be the highest band).? For the other two bands, 60 meters and 80 meters, I could perhaps live fine with just one simple elliptical lowpass aimed at suppressing 7 MHz.? 10 and 14 MHz are out of the game in this case.
?
Two ideas:
- in LTspice I was unable to design a lowpass that would work fine for both 80 and 60 meters (in OK we have the WARC 15-15 segment, i.e. 5351.5 to 5366.5 15W PEP) - 7 MHz is too close to 5.3 so 2nd harmonic suppression was not great, despite the assumption that push-pull amp creates relatively weak 2nd harmonic. 3rd harmonic of both 3.6 (central spot of all digital mods on 80m) and 2nd harmonic of 5.3 should be well suppressed by the original lowpass. So I need to resolve just the 2nd harmonic from 80 meters (7.2) and whatever else above that is attenuated is a bonus.
- I cannot just take some 5.3 MHz lowpass and put it as-is after the original 7 MHz lowpass because the input shunt capacitor of the appended LPF would add to the output capacitor of the original LPF and that would result in a mess. So the "appendix" LPF must be designed as "continuation" of the original built-in LPF. But that should be easy with LTSpice.
?
I will try to simulate this, build it, and if it works I will post it here.
?
73 Jindra
?
.
?
On Fri, Jan 5, 2024 at 01:25 PM, Donald S Brant Jr wrote:
To keep it simple in an emergency situation you could just use outboard lowpass filters, swapping them manually as needed.? ?Put your highest-band filter in the box (always inline) and add lower-band ones when required.??
Instead of "...and a small SMPS giving ~ -300 volts to shut down 1N4007..."? ?you could just tap and rectify/voltage-double the RF output.? This would guarantee always having sufficient voltage to bias your diodes off.
73, Don N2VGU
|
Thanks Hans. It seems I am going to have to modify a couple of filters in either my QMX or QMX+ to become an SSB walkie-talkie on 6 and 4m ;-)?
73?
Tony
G3OVH
?
?
?
?
|
ATU 10 tuner with QMX/QMX+
Hi
Set the Protection Tune % to 75 (56.25% Power) (Tune % 70 is 50% Power)
Then Tune SWR triggers the ATU 10 nicely (It has problems triggering with Tune % 50 (25% Power) setting)
If you can not get a tune less than 2, don't repeat it.?
73 DE muhsin TA1MHS
|
Re: QMX new build showed up as USB but unresponsive after firmware transfer
Florin, no, the available spare SMPS boards are good for all QMX revisions. And these boards are with some changed elements compared to REV 1 to enhance the robustness.
?
Ludwig, got it that's good to know. I was already set on doing some things better from the previous build, so in the end it worked out but in case I'll ever need just the power supply boards I'll make sure to get those. Again, thanks for all the support!
?
I used to reverse-engineer cell phones and other devices and my best solution is to use multiple colors of AWG28 wire-wrap wire and a fine-tip iron to tap the circuit nodes I want to measure.
Those are some very good points, Don. The build has definitely sparked more interest in debugging hardware and relearning skills I haven't used in more than a decade at this point. I'll be applying your suggestions in the near future for sure, thanks for sharing.
|
Re: Using WSJT-X 27.7.0-rc8 "FYI"
The full release of WSJT-X 2.7.0 is now available, and is very similar to -rc8.
https://sourceforge.net/projects/wsjt/files/
?
WSJT-X WILL work fine with the newly added radio selection "QRPLabs QMX" (as shown in the previous post) - but you have to upgrade hamlib to the current version 4.6.2, otherwise it will crash - the QMX wasn't properly in hamlib v4.6.1.
?
Stan KC7XE
|
Can the 5351 clock generator operate at 70MHz and be phase modulated fast enough to generate SSB??
I see no reason why not, though I have not tested beyond 6m.?
Note also that the maximum update rate of the '5351 clock generator, when using 1MHz I2C bus, is around 16,000 times per second. This is not dependent on the operating frequency of the '5351. So there is no reason why it would be different on 70 MHz. And since 16,000 updates is plenty more than required for SSB modulation, it all should work.
|
Re: SMD chip rework equipment
You just need resonable tools.?
?
I have been using Weller, my neighbor uses? Hakko branded.
?
The air gun and nozzles to warm the board so a pair of mechanical tweezers can pick off the SMD chip.
For thu hole, I use warm the area with the air gun and with the 200W iron and solder wick remove the majority of solder.
The use the Hakko Desolder gun to clean it up more.?
?
The Soldering Tweezers are the best to desolder SMD chips. Occasionally I will solder with them.
?
For the most part I use the 70W iron and solder wick, Chipquick?
Second is the air gun,?
?
Listed below is a SMD practice kit. Better to learn on a practice kit than to overheat a kit board or an actual radio chassis.?
?
Jerry's advise is good. Youtube is a great refressher? on technique.
?
I use :
?
A REALLY GOOD LIGHTED MAGNIFIER.? I have a couple. one of my favorites is one with a extra high power lense that is in one spot of the?
overall magnifier lense. If you can't see it clearly, you can't fix it or build it better.
?
Flux solder, solder paste and liquid flux, 90% isopropyl alcohol and lint free rags are the consumables.??
Compressed air is a plus.?
?
An basic temerature controlled air gun with a variety of tips to focus the hot air stream.
Air Gun? - https://www.amazon.com/dp/B0752Z4165
Nozzles - https://www.amazon.com/gp/product/B07G352P13
?
Weller 200W Temperature Controlled Iron with an assortment of large tips, Chisel or round nose
?Iron -
Tips -
?
Weller Soldering Tweezers RTW3? ?T0054465399N
https://www.amazon.com/dp/B06XCMN8W2
These heated soldering tips have flat paddles for tips and are great to grip SMD chips with and solder/desolder them
?
Weller Temperature controlled 70W Iron
?
Weller Assorted Tips for 70W iron
?
Hakko Desoldering gun
?
SMD soldering practice kit
https://www.amazon.com/gp/product/B08KDLR6P6
|
Re: SMD chip rework equipment
I have had a hot air rework station for 20 years and couldn't live without it. Mine is a very old one with analog controls. I see many digital ones on amazon in the $35 price range which is a lot less than I paid for mine years ago. Don't know about the quality?
?
For a heat table, I use an electric griddle I picked up at a thrift store. It does help the hot air a lot to have a heated board especially on boards with lead free solder. I have even assembled whole boards with it using a solder paste syringe, placing all the components, then ramping up the heat on the griddle until the solder melts.
?
I would avoid the Harbor Freight/hardware store 1500 watt heat guns as they tend to blow multiple components off the board. However, I have used them in the past to harvest all the components off of old boards.
|
Re: QCX buzzing sound on key down
Thanks for the tip but I think the microcontroller is damaged. I guess I will order one. Very erratic. Voltage is ok. When booted it emits a tone constantly when keyed. Rebooting is the only way to stop it. Then it may work. It ?stops responding to the paddle input after a few characters. I am going to look a little more. I think I may have a spare microcontroller. It would be any easy test to swap it.
73s John N4HNO
|
Re: SMD chip rework equipment
Many people here have spent a fortune on rework tools.
That is not necessarily warranted when working on some $100 QRP kit.
?
These boards go through an auto assembly house using no-lead solder.
That makes them very hard to work on, as that solder doesn't melt till it gets extremely hot.
Easiest way to get a part off the board is to use some low temp solder alloy such as chipquik,
there are cheaper kinds of this stuff to consider if you use a lot of it.
Melt some of that into the nasty lead free high temp solder the boards come with.
Then you can pick the part up using a hot air gun.? A cheap embossing gun might work,
though one of the $20 1500W heat gun (such as from Harbor Freight) would also work
and might be useful for larger jobs.? I've never had a heat table.
?
Be sure to thoroughly clean the low temp solder alloy off with solder braid and flux.
You don't want parts falling off the board on a hot summer's day.
?
I then use isopropyl alcohol to clean flux and such off the board, the less water the better
since the water takes much longer to dry off and can affect how things work.
Most of the stuff at the drugstores is 70% alcohol, but some do have 90%.
If you see the 99%, get a couple bottles.
?
Here's a cheap soldering iron kit I am happy with (though I do have Hakko's and Wellers too):
Includes several tips, flux, and solder, good temperature control.
That one does not have a safety ground at the power plug,? not a bad idea to run a
ground wire from the soldering iron barrel to your workplace ground
in case there is a minor fault inside the iron.
?
You might want thinner solder for some of the smaller parts, but that's a good start.
You can wet the iron with most any size of solder, then bring the tip to the pin you
wish to solder, having put a bit of flux on the pin first.
?
Lots of light and good magnification is a must, need to see small defects like solder whiskers.
Here's a cheap magnifying headband that has been mentioned in the forum:
All three choices are the same except the mid-level item has a brighter light,
and the high end one also has a bag you can store everything neatly away in
if you happen to be that sort of person.
Many reviews do say that the nose rest on it could be more comfortable,
that will depend on your particular nose.
?
Beyond that, basic hand tools such as tweezers, wire stripper, xacto knife, and needle nose pliers.
?
Be careful about static, especially if it's freezing cold out and humidity is low.
A static mat on your work surface plus wrist strap is ideal but can be expensive.
I doubt most in the forum bother.
At least get in the habit of touching a ground point on the device first before
sticking your finger into some especially sensitive signal node.
?
?
|
Re: SMD chip rework equipment
First watch several YouTube videos about diy smd replacement.? Google will give you many options.? ?There are a few different methods,? and you will want to educate yourself on them to decide how you want to do it.? Then purchase the requisite tools and supplies (they don't need to be expensive, but don't get the cheapest ones, either).? Then practice several times using discarded circuit boards (from old phones or electronics) with components comparable to what you want to work on - practice both removal and replacement.?
?
Then go for it.? That's what I did, and have since successfully done several smd rework tasks.? But I have also sometimes lifted pads and had to repair them, even after some experience,? so you need to be very patient and careful.?
?
Stan
|
Re: QCX 50W PA multibanding?
#50wpa
Further, from AN006 ?
"The highest frequency
(10m band LPF) must go in slot 1, closest
to the RF output from the relay board. If you have at least Rev 5 of the relay PCB, the highest
frequency LPF is always in circuit. This provides improved attenuation of VHF harmonics."
?
One of the guys in my radio club is building a legal-limit LDMOS amp which uses the same scheme.
73, Don N2VGU?
|
Re: QCX 50W PA multibanding?
#50wpa
On Sun, Feb 23, 2025 at 04:26 PM, ok4rm wrote:
I cannot just take some 5.3 MHz lowpass and put it as-is after the original 7 MHz lowpass because the input shunt capacitor of the appended LPF would add to the output capacitor of the original LPF and that would result in a mess.
Actually, Hans' Ultimate Lowpass Filter kit does just that, puts the highest filter always inline and switches lower-frequency ones in series with it:? " This relay-switched LPF board kit can have up to five additional LPF's plugged in, permitting the??to sequence between 6 bands (including the LPF on the main U3 board)."
I have successfully cascaded filters and while the result may not have been textbook-perfect due to filter interactions it got the job done.??
It would not be difficult to do an experiment, you will need the filters in any case, just hook 'em up and see.
?
73, Don N2VGU
|
Re: QCX buzzing sound on key down
One thing to check is that your power supply voltage is not drooping at keydown below a level that supports proper operation.? My QMX will squeal if the battery voltage gets too low during transmit.? Think of it as a low battery alarm :-).
?
Brien - KE7WB
?
On Sun, Feb 23, 2025 at 01:23 PM, Tech Guy wrote:
ud buzzing on key down at times. It is not transmitting then. If I power off and on it usually goes away and sounds ok
|
Re: QMX new build showed up as USB but unresponsive after firmware transfer
On Sun, Feb 23, 2025 at 03:41 PM, Florin wrote:
- wielding the DMM probes while the radio is powered up is a bad idea most likely
I used to reverse-engineer cell phones and other devices and my best solution is to use multiple colors of AWG28 wire-wrap wire and a fine-tip iron to tap the circuit nodes I want to measure.
This has 8 colors and 250 meters will last a long time: ?
It is also great for PCB repairs and alterations.
I use hot melt or Krazy-type glue to tack down the wires someplace close so that it is not easy to snag one and rip up a trace, then tape down the free ends spaced so that they cannot short together as you connect to them.
It takes a little prep time and forethought, but I was working with irreplaceable prototypes and stealthily-obtained items so care was paramount.? Nice ham radios deserve care, too.
73, Don N2VGU
|
Re: SMD chip rework equipment
Unless you plan on doing a lot of it get friendly with the local laptop repair shop?
?
Otherwise?
Quality soldering iron
Background heat table?
Excellent (ideally stereo) magnification/light
Fume extraction, static protection?
Decent hot air gun (no not the cheapest one)?
Assorted pads and kapton tape to protect surrounding components?
picks, hooks pins and lots of cocktail sticks?
Solder paste. ?Rework flux, desolder braid
Cleaner, UV varnish, thin enamelled wire?
Drills, grinders, scalpels etc?
?
A ton of old boards to learn/practice on.
i do mean a lot before you work on anything valuable?
?
Ps. ?Different people will do the same task in quite different ways so it’s not a ‘one way only’ process. ? You’ll want to find techniques YOU can use confidently and successfully. ? ?
Like I say I use loads of toothpicks for small component installation, ‘back-flip’ removal and hole clearance to name just a few . . ?Others do it other ways. ??
|
SMD chip rework equipment
In anticipation of future needs. I've yet to attempt anything with more than three leads, so I was curious what SMD chip removal and replacement methods, techniques, and equipment those in the know would recommend for the home hobbyist.?
|
Re: QCX 50W PA multibanding?
#50wpa
Yep, in RX path it has almost the same effect. Theoretical insertion loss of the standalone LPF (with ideal C and L) calculated by LTspice is 0.01 dB worse than the big combo. So in reality it will be the same in and out.
Jindra?
?
On Mon, Feb 24, 2025 at 12:11 AM, ok4rm wrote:
Hmm, it looks interesting:
- to make an "appendix LPF" that would work for both 3.6 and 5.3 MHz does not seem to be possible
- however, adding an elliptical PI-network C-L-C, sort of an "extra stage", seems to work for 3.6 MHz:?
?
first shunt capacitance: 727 pF (680+47), together with the 270 pF output capacitance in the PA it makes 997 pF
inductance: 1.77 uH (19 turns on T50-2), and 270 pF in parallel with it makes a parallel resonant circuit at 7.2 MHz
output capacitance: 727 pF (670+47)
?
This combination should suppress 7.2 MHz by 50-60 dB (resonance is critical and there might be tolerances in capacitor values as well as winding). Optimum values are slighly different by 2 to 3 pF but impact is insignificant (order of 0.001 dB according to simulation).
?
Now I have to check how will this appendix behave standalone in the "opposite direction", i.e. transceiver in RX, PA is bypassed and RX signal goes only through this LPF. But I think it should be fine. Tomorrow or on Tuesday I will try this. Unfortunately I think I have no BNC connectors in the drawer.
?
73 Jindra
?
On Sun, Feb 23, 2025 at 10:26 PM, ok4rm wrote:
I like this idea, Don. I built the PA for 7 MHz (for emcomm this would be the highest band).? For the other two bands, 60 meters and 80 meters, I could perhaps live fine with just one simple elliptical lowpass aimed at suppressing 7 MHz.? 10 and 14 MHz are out of the game in this case.
?
Two ideas:
- in LTspice I was unable to design a lowpass that would work fine for both 80 and 60 meters (in OK we have the WARC 15-15 segment, i.e. 5351.5 to 5366.5 15W PEP) - 7 MHz is too close to 5.3 so 2nd harmonic suppression was not great, despite the assumption that push-pull amp creates relatively weak 2nd harmonic. 3rd harmonic of both 3.6 (central spot of all digital mods on 80m) and 2nd harmonic of 5.3 should be well suppressed by the original lowpass. So I need to resolve just the 2nd harmonic from 80 meters (7.2) and whatever else above that is attenuated is a bonus.
- I cannot just take some 5.3 MHz lowpass and put it as-is after the original 7 MHz lowpass because the input shunt capacitor of the appended LPF would add to the output capacitor of the original LPF and that would result in a mess. So the "appendix" LPF must be designed as "continuation" of the original built-in LPF. But that should be easy with LTSpice.
?
I will try to simulate this, build it, and if it works I will post it here.
?
73 Jindra
?
?
?
On Fri, Jan 5, 2024 at 01:25 PM, Donald S Brant Jr wrote:
To keep it simple in an emergency situation you could just use outboard lowpass filters, swapping them manually as needed.? ?Put your highest-band filter in the box (always inline) and add lower-band ones when required.??
Instead of "...and a small SMPS giving ~ -300 volts to shut down 1N4007..."? ?you could just tap and rectify/voltage-double the RF output.? This would guarantee always having sufficient voltage to bias your diodes off.
73, Don N2VGU
|
Donald S Brant Jr
Brien - KE7WB
Bruce Akhurst