Re: QDX - protection of BS17O PA
Hi Zach, Ted, all
Making a transmitter completely resilient to any SWR mismatch is quite a hard task. It considerably increases the complexity (and thereby costs). I think this is what Ted was referring to, when he said "Wild SWRs will do for every transmitter I own, this was drummed into me early on." - in other words, none of his transmitters could withstand every possible SWR mismatch. And that the best possible protection is to go through a series of mental checklists before transmitting. Which is what I do every time too. Before I switch on the power supply, I'm always mentally tracing the cables on my workbench from the unit I'm going to switch on, to which power supply, anything else connected to that power supply, what voltages any of these things are accepting, what voltage the power supply is currently set at, what rig is the antenna connected to, did I unplug it for a storm, blah blah.?
I'd also like to remind everyone that a couple of weeks' back I posted here, wish I had the post reference handy, where I did a quick experiment?- QDX with open load, and short-circuit load. In neither case was the voltage at the drains anywhere near the 60V limit of the BS170s. Which doesn't mean that there IS no type of mismatch which would cause a high voltage at the drain; but I would challenge anyone to identify such a load and experimentally verify that it does create a > 60V drain voltage on the QDX finals.?
If such a traumatic load is non-existent or very unlikely, then by adding the 47V Zener you are achieving nothing? or almost nothing against high voltages due to SWR mismatch.?
The other ways bad SWR could cause problems are high current, or high power dissipation causing overheating. I think this is much more likely (the short-circuit load causes high current consumption) but it is not addressed by the 47V zener.?
What the 47V zener DOES do, and also the diode across the PA choke, is to eliminate a very short (sub-microsecond) spike at key-down which occurs due to the sudden dI/dt in the choke (the RF envelope isn't shaped). One may speculate that this spike causes avalanche breakdown in the BS170s which is a bad thing and one may speculate that it can contribute to stress that makes the BS170s more likely to fail; however these are speculations without experimental evidence.?
So in the end I think protecting against high current versions of SWR that cause failure due to overheating is the more important thing - but it is not being addressed by the zener or 1N4148 diode solutions.?
On Mon, Sep 4, 2023 at 10:20?AM Ted 2E0THH < qrp@...> wrote: Well call it the pilot in me.
I have an old check list I perform prior to any transmission and have never blown any final in all my years.
On my vintage valve transmitters, cooking output tubes would be a somewhat more expensive exercise.
Ted
2E0THH
|
Re: QDX - protection of BS17O PA
Dave,
When you do the diodes across all your rigs lets us know if any issues you come up against. ?I have a built QMX, a QDX kit and another QMX kit in the boxes and want to be as protective as I can be.
Thank you.
Dave K8WPE since 1960 David J. Wilcox’s iPad
toggle quoted message
Show quoted text
On Sep 2, 2023, at 11:26 PM, David R. Hassall WA5DJJ <dhassall@...> wrote:
?Dear Paul,
Thank you for that clarification.?? 1N4148 across the choke? with the cathode towards the 12V lead.?? OK? I got it.
Take care and have fun.? Happy Labor Day weekend
73 Dave WA5DJJ
SUPER QRSS/WSPR GRABBER
|
Hello John, all
When I started out in QRSS weak signal modes in around 2004 sending FSKCW and such simple things... we quite often would notice doppler shifts, particularly around sunrise/sunset, due to ionsopheric?layer rapid movement. It could be a Hz or two and easily noticeable. You could also get reflections from nearby aircraft, with their own larger shift and quite distinctive.?
Both ends of the communication path can be fixed. But if the reflective thing(s) in the middle move then you can still get measurable doppler.?
More on Ionosphere Doppler between two fixed points
Yes Doppler-effect a fascinating subject
Seems like I've always know about the different between day/night propergation AND flutter (rapid QSB) but ...
...While tooling up for 2023 and 2024 solar eclips I discover there are four? 'Twlight Zones'? (Astronomical, Nautical, Civil? (first light), then SunRise) over each station for a total of eight that a common NE602 DC receiver and fldigi can detect and record (milli Hz) and amplitude (dB Volt) then plotted with EXCEL (see my page for more info)
Yet another facet on this marvoules ham radio hobby
72 73
John
N3AAZ
|
Re: #qmx Don’t Use USB-C PD to power your QMX?!
#qmx
Well Hans,
I made an LTSpice model and the simulation shows that the proposed solution will work.
The voltage step 6 to 12 Volt occurs at 30 milliseconds with zero risetime
Thank you, that is certainly very encouraging! Making it worthwhile to investigate further in hardware.?
Note that the actual PWM freq in QMX is around 110-120kHz (116kHz nominally). Not 166. But I don't think that will affect the results.?
If you could save that picture in a lossless compression e.g. png, because even the subsequent?one you attached was hard to read (fuzzy when enlarged), it would be very useful. I'd like to be able to zoom in on the details. I can't quite read off the top line which identifies what the traces are.?
|
?""I've not see a method too edit sent text to update it"" ... Paul
On far right click three bars 'More'
then 'Delete'
confirm click 'Yes'
then re input your comment?
72 73
John
N3AAZ
|
Re: #qmx Don’t Use USB-C PD to power your QMX?!
#qmx
Hello Don
so at least for the time being, I will inhibit the "handover procedure" from the linear to the 3,3V buck converter
Could this be a software/firmware option?
Yes it could be. But then I would need to test the long term thermal situation when running off linear regulated supply like that, at much lower efficiency;?
In any case bear in mind, inhibiting the linear regulator handover only applies to the 3.3V board (#2) which is the only one having a 3.3V linear regulator. There is no such equivalent on the 5V board #1. In the currently discussed failure scenarios?we are discussing?board #1, the 5V board. There hasn't been any equivalent case of the 3.3V board failing. So what would be the point of keeping?the 3.3V board on linear regulator, but not the 5V...?
|
Re: #qmx Don’t Use USB-C PD to power your QMX?!
#qmx
Dear Razvan
Your emails are always friendly and entertaining!?
So why all the nonsense boost up then convert down again... because requiring a user to provide multiple power supply voltages would?clearly be a recipe for disaster. So we need to accept a single supply?voltage, which is the highest voltage which will be required by the transceiver - typically during transmit operations.?
Addition of an internal boost converter to provide 12V during transmit only, from say a 5V supply, would be quite a complication and expense - particularly when you consider full break-in CW. What happens then... the boost converter stays on during the breaks too? Or you try and make it quick enough??
I am quite happy with the current supply voltage situation, starting with 12V and converting down to other needed supplies makes perfect sense for most users.?
Don't forget everyone has the option to build their QMX for 9V, it's a good option too.?
Starting with a 3.7V battery and boosting it up to 5V, then to 12V, doesn't make a lot of sense to me either, I fear conversion noise disrupting the precious RF; I like to start with a supply that is really 12V then let the radio convert it down; clean. In the lab I use linear regulated supplies, good old fashioned transformers followed by linear regulators; on the (very) rare occasion of a portable op such as? then I like a real battery able to deliver 12V. I think this is the best way.?
But a note again, for anyone wishing to linearly regulate for QMX; I don't recommend it; but, if you want to do it, then you should connect the ADC inputs from both the 3.3V and 5V regulators to Vdd (3.3V) such that the microcontroller will think that the SMPS are brought up correctly, then it will proceed with the initialization and ongoing operation of the radio.?
If you decided to use the onboard 3.3V regulator 78M33 on the SMPS board #2 then?bear in mind I have NOT tested this for thermal suitability for long term operation, from 12V down to 3.3V, linearly. The PCB is 6-layer and the drain tab has a big copper area, which is only a fraction of a mm from the ground plane right under it; but still, you will be in experimental territory.?
On Sun, Sep 3, 2023 at 10:10?PM DL2ARL < dl2arl@...> wrote: On Fri, Aug 11, 2023 at 05:38 AM, Shirley Dulcey KE1L wrote:
Cell phone chargers are often in the category of only delivering 5V or 9V
Most gentle Ladies and Gentlemen,
I humbly think that the PowerSupply Railway-Exchange we are all driving trough is for one reason unnecessarily complicated. I try never to raise oppinion against something if I do not have a better alternative. But in this case, I do. It is not my idea: I read about it in:
CQ March 2015
QRP Special
Page 18:?Behold Rockmitey, de ED Cobb?K4YFR es Ray Kauffman?AJ4YN.
Look what happens. Basically we need 5V to operate the radio (disclaimer: most of our qrp radios). In some cases we also need 3,3V, but that set apart, let us stick with the magic 5V level: because of USB and Cell-Phones 5V has become the standard of our age.
What we do in most cases is so wrong, in so many ways:
we probably start with the voltage of a single 3,7V LiPo Cell built in some Power Bank. The built-in converter (most probably of a some noisy, very bad efficiency) boosts this voltage to 5V. Than we use maybe another internal Boost-UP-Converter, maybe an external one to increase 5V to the 12V expected at the input of the (qrp) transceiver.
We feed the 12V into the rig of our choice, where, for the most of the time it gets down-converted to 5V (and probably to 3,3V too) with another efficiency loss. The 12V is only needed for the short TX Phase. I do not know what your mileage is, but for me, I am an SWL, most of the time, probably 90% of the time I spend on the HAM bands. I listen most of the time and transmit just now and then.
So why travelling trough this complicated UP-and-DOWN Railway-Exchange anyhow? What do I do instead?
Following the learning done in that CQ Magazine quoted above, I have modified some of my MTR-Transceivers (superb KD1JV designs) for the following path:
I start with 2 LiPo cells, yielding together 7,4V. This choice has another advantage: for a reason I do not know, it is much easyer to find a solution to charge a 2Cell LiPo battery than it is for a 3C or even worse for a 4C battery. A 2C pack not only can it be easily charged, but it also gets well balanced.
This 7,4 V supply is in MOST Cases enough for the MTR to offer a HF output at the qrp-qrpp border. In most cases enough for me. For the eventuality I would try to go up to "full legal qrp power", I use a Boost-Up converter to 11..12V that is only active during the TX phase. The transition could be made automatic (like described in the CQ article), driven by some PTT Level in the rig, but because I do not use this option often, it remains for me a manual switch.?
For the rest, using a LDO, I convert the 7,4V to 5V and feed the Rx Part of the MTR bypassing it's internal converter, whatever of a sort this one might be. Because of the relative low current consumption of the Receiver in the MTR, the conversion loss is minimal and not to be taken into account. The MTR5 even has an internal buck converter for this purpose.
Whatever I did, whatever I do suits best my needs. But in order to be able to do so, I had to modify the existing rigs to fit my needs. It would have been much easyer and straight forward if they would have been designed for a wiser and more modern Supply Voltage Railroad-Exchange. The QMX is a huge step ahead, but not in this direction. And because of it's adventurous behaviour, I will probably not feel at ease with it, so at least for the time being, I will inhibit the "handover procedure" from the linear to the 3,3V buck converter and at least feel safe, in the event I will not be.
If we like it or not (I do not!), we live in a USB dominated world, we must use LiPo cells and both of them are very badly suited for our old habitual 12V, voltage probably coming from the stone-age of electric development related to those lead-acid batteries formerly used in gasoline eating cars ...
just thinking loud.
yours friendly, Razvan dl2arl
|
More on Ionosphere Doppler between two fixed points
Yes Doppler-effect a fascinating subject
Seems like I've always know about the different between day/night propergation AND flutter (rapid QSB) but ...
...While tooling up for 2023 and 2024 solar eclips I discover there are four? 'Twlight Zones'? (Astronomical, Nautical, Civil? (first light), then SunRise) over each station for a total of eight that a common NE602 DC receiver and fldigi can detect and record (milli Hz) and amplitude (dB Volt) then plotted with EXCEL (see my qrz.com page for more info)
Yet another facet on this marvoules ham radio hobby
72 73
John
N3AAZ
|
Re: #qmx Don’t Use USB-C PD to power your QMX?!
#qmx
Hi Guido ?
Q=C*U? where Q is constant, but C decreases (e.g 10%) under the influence of increasing U -> making U even higher (e.g. 10 times higher).
I didn't get that bit...
What I?had in mind is the capacitance-loss that comes with X5R material when DC-voltage increases: e.g. a 100nF cap becomes 90nF at 5V and 10nF at 12V. Now when this capacitor is charged at 5V the charge Q = 5 * 90nF = 450n. Now when stepping the 5V to 12V the capacitance suddenly becomes 10nF, but since we already had 450n charge the new voltage becomes a voltage spike of 450n / 10nF = 45V. Does this make sense? I could be mistaken.
Doesn't make sense to me. I know X5R capacitors suffer capacitance loss at increased voltage but 90% capacitance loss by 12V seems higher than I would expect; the bit about the voltage spike is not anything I've heard about before. And those capacitors are on low impedance supply rails, it's hard to imagine a harmful voltage spike developing out of that. Any other comments from people here, on that topic?? ?
So my question is still... Wouldn't just making the zener diode bigger fix the sensitivity to this unusual supply scenario? In my tests, with the standard circuit (500mW zeners) I was able to do 6V to 9V, 10V, 11V steps without issue and the zener ate the brief over-voltage. It even did on 6 to 12V steps but not every time. So given that it almost works even with the 500mW zeners isn't it worth trying a 5W zener say, being optimistic that it may be able to swallow that extra current for a millisecond? Then if QMX has a specified supply voltage range of 7V to 12V say, you'd be able to do anything at all with the supply voltage in that range without any fear of damage.
True, that using bigger diodes might fix it eventually, the transients are?short. But on the other hand with L101=1.5R at 12V there is a theoretical peak current of more than 4A (12-5.6)/1.5, peaking to more than 25W 4*(12-5.6), so 5W might not be sufficient.
I'm not sure of your calculation; there are three P-ch MOSFETs in series with the 12V supply too; furthermore the 330uH I measured here at 2.0 ohms.?
The 5W zener diode I am considering is?SMBJ5339B, whose datasheet specifies a maximum surge current of 13.4 A; where? "The Surge Current (IZM) is Specified As The Maximum Peak of a Nonrecurring Sine Wave of 8.3 Milliseconds Duration." This makes it look well capable enough to withstand a couple of amps for 1 or 2 milliseconds.?
|
Re: QDX - protection of BS17O PA
Well call it the pilot in me.
I have an old check list I perform prior to any transmission and have never blown any final in all my years.
On my vintage valve transmitters, cooking output tubes would be a somewhat more expensive exercise.
Ted
2E0THH
|
Re: Successful QMX build - and some bug reports
#qmx
Hans said there was a volume issue, and he is working on AGC, SSB, and being able to manually adjust the time.
Thought there would be another update by now....patiently waiting!
73's Michael N2ZDB
|
Re: QDX Rev 4 Only Shows 80M Band After Working Near T2
Hello Evan
> Are you sure you downloaded the correct firmware??? > There are versions for a single-band QDX.?? > Try an older version and see if that helps.
No, firmware is the same... It's just a documentation error, the QDX-M web page should have been updated. The firmware for QDX and QDX-M is the same. I have corrected the QDX-M page now.?
On Mon, Sep 4, 2023 at 12:35?AM Evan Hand < elhandjr@...> wrote: Are you sure you downloaded the correct firmware?? There are versions for a single-band QDX.? Try an older version and see if that helps.
Correct multiband QDX is here:
Single band is here:
I did note that the latest single band is 1_08, and the screenshot you posted lists 1_10, so it's not likely.? Try the prior release:
73
Evan
AC9TU
73
Evan
AC9TU
|
Re: QDX Rev 4 Only Shows 80M Band After Working Near T2
I'm not quite ready to call it fixed but I got all my bands back!
I found the MOSFET that controls 20M(I think Q5) was faulty. The output from the microcontroller would only come up to 2.1V instead of 3.3V on the 20M band line. When powered off I found less than 100Ohms between the pins on Q5 where Q4 and Q3 would be in the megaohm range.
After removing the faulty MOSFET the voltage from the microcontroller reached 3.3V as expected. I then swapped the MOSFET from the 40M filter over to the 20M position. I put firmware 1_10 back on and all the bands stayed!??
I'm not sure what caused the issue and I'm not convinced there aren't still other problems but I have a clear path forward now. I'll be ordering some BSS123s. If I find anything else which might be helpful I'll post back here for anyone who might see something similar in the future.
|
Hi Darren,
One other point to check would be the power supply.? Verify that the voltage is not dropping.? Also, monitor the current.
Can you hear signals when connected to an antenna?? Check on 7.074MHz FT8 frequency.? It is busy most of the time with digital signals you should be able to hear.? It is not morse.??
73
Evan
AC9TU
|
|
|
Re: QMX build for [40m/30m] , [20m/17m] & [15m/12m/10m]
#qmx
Could you repost the RWTST graphic.? I have low 10m power on my QDX.
|
WSPR is a Weak Signal Propagation Reporter system, someone needs to be on the other end to RX. To get the most out of using WSPR one needs to REDUCE the TX power to somewhere between 9mW and 200 mW in order to gain any meaningful data that will translate into a modes potential (at that moment in time) I usually run 100-200mW when I setup for a session of QRP/CW. ?I have found that for the most part what I saw on WSPR is representative of the results that I will have but is of course dependent upon someone on the other end willing to participate.
My primary use of the mode is to confirm that my antenna (as I built it) is functioning as it should, there are no other guarantees with the mode. ?I set the parameters on my reporting program to Minimum SNR of -20dB +/- depending on where I wish to go with my signals. ?Anywhere near -10+ usually will get my 4w CW in the area.
I am now trying the QRP-Labs U4B @ 27mW and find the predictions are extremely close to QRP/CW reality....... running 1+ Watts really doesn’t make much sense to me, for the power that I am running.
John KK4ITX?
|
Hi Darren. I would definitely first look at the audio muting circuit R60, D5 and Q7. You should see 5v on the gate of Q7 when the radio is keyed. If not there, look for it on the cathode first, then the anode of D5, and finally on pin 11 of IC3. Also, you didn't indicate whether you were using a dummy load or an antenna, but the symptom could also be caused?by RF getting back into the audio circuits. Hope that gives you a starting point to look at.
73 .. .Ron
On Sun, Sep 3, 2023 at 2:43?PM Darren KN4WML < darrent78@...> wrote: Good evening all.? ?
? ? ? ?On my qcx+ build which went well this time, I have an audio issue that I'm trying to debunk.? Everything seems to be working as it should - about 4.97 watts in the dummy load- no smoke .? ? But when I key down I get the noise I'm hearing sounds like a blown speaker.? ?It doesn't matter which speaker or headphones I have attached to it.? ?I thinking that there's an problem in the audio part.? ??
? ? ?I'm definitely a rookie when it comes to electronics, but I can hold my own.? ?Any questions or suggestions? on where I should start to look at would be helpful .
? ? ?Thanks
|
Good evening all.? ?
? ? ? ?On my qcx+ build which went well this time, I have an audio issue that I'm trying to debunk.? Everything seems to be working as it should - about 4.97 watts in the dummy load- no smoke .? ? But when I key down I get the noise I'm hearing sounds like a blown speaker.? ?It doesn't matter which speaker or headphones I have attached to it.? ?I thinking that there's an problem in the audio part.? ??
? ? ?I'm definitely a rookie when it comes to electronics, but I can hold my own.? ?Any questions or suggestions? on where I should start to look at would be helpful .
? ? ?Thanks
|
Re: QDX Rev 4 Only Shows 80M Band After Working Near T2
Are you sure you downloaded the correct firmware?? There are versions for a single-band QDX.? Try an older version and see if that helps.
Correct multiband QDX is here:
Single band is here:
I did note that the latest single band is 1_08, and the screenshot you posted lists 1_10, so it's not likely.? Try the prior release:
73
Evan
AC9TU
73
Evan
AC9TU
|
David Wilcox K8WPE
Ted 2E0THH
KK4ITX John
Ronald Taylor
Darren KN4WML