Hi Evan,
Thanks for your help on this.
?
This morning, I took a look at those nylon insulators. There was a very small amount protruding from the bottoms of the IRF510's so I sanded them down. Good call. And yes, I used freeze spray to try and locate the thermal issue.
?
From a cold start, the amp turned on with idle current at 0.024A. Oddly. The amp initially keyed up, oddly, with the power down at 23 watts. I didn't notice that until late yesterday. On second key up the power went to 40w at 4.5amps, fairly stable, but very slowly started falling off. Probably somewhat normal as things warm up. After 30seconds or so the power suddenly dropped to 23w with current at 3.0A.
?
In general, once the power profile starts this behavior, and things get warmer, the sudden drops occur more quickly to the point where it's almost so fast as to not be seen on the scope. After a few minutes of rest the power will come up to 40w for several seconds then start its bad behavior again. This is why I believe it's a thermal problem.
?
Yesterday, I went through the entire NA4Y video, "QCX Amplifier Troubleshooting". All checked out pretty much as he describes with the exception of a few voltages in mine were lower; perhaps because mine is a 20m and his was a 40m I think.
?
Additionally, I pulled the attenuator resistors and checked them. All okay as far as ohms values.
I also attempted to check capacitors in circuit using the Extech LCR meter to look at ESR. I used same value caps from my assortment of new caps to get a baseline value and compared that with the ones on the amp. Most all the caps checked just like the baseline measurement. The ones that didn't at least were consistent for all the others of same value on the board. Of these I pulled one and measure it out of circuit and it was good, so I am assuming all the others in the same value group are the same. I suppose something in the circuit caused them not to measure the same as the baseline, I don't know for certain as I have never tried this technique before.
?
?I'm leaning toward ordering some more IRF510s and thermal pads, maybe some new nylon screw insulators, but would like to eliminate any other potential parts, to avoid multiple orders to Digikey.
?
Your thoughts?
--
regards,
Bryan, N0LUF
|
QMX+ receiver clipping distortion on 40M CW during RAC contest...
Hi,
I have been playing with my QMX+ over the last month
or so since I built it, and it has been behaving quite
nicely. Mostly, I have been listening in on W1AW's
high speed CW, hoping to one day have enough confidence
to actually do my first CW QSO since being licensed a
half century ago. W1AW rolls in here very strong.
Anyway, the Canadians were having their RAC contest
over the weekend, so I thought that I would listen in,
and brush up on my experience with contest exchanges.
In the beginning, all was well, but as the contest
built up steam, and there was a forest of CW signals on
the air, I noticed a peculiar thing: The low level
signals all had a raspy random noise signature riding
on top of their CW tones. If a signal was strong
enough to activate the AGC, the noise went away, but for
any signal below the AGC threshold, the noise was there,
no matter where I was tuned relative to the carrier.
At first, I thought the contestants were using poor
quality rigs with bad CW signals... but after the number
of such noisy CW signals reached the dozens, I had to
turn my pointy finger back in my direction.
It has to be my QMX+.
So, this morning, I investigated the QMX+'s low level
signal behavior using a signal generator, to see if I
could duplicate the issue. I couldn't.
The tone was sweet and pure all the way from S9++ down
until it became inaudible.
It would seem that it has something to do with large
numbers of middling strength signals within the
receiver's band.
Note that the noisy signals weren't on the threshold of
disappearing gracefully into the noise. The signals were
all head and shoulders above the background noise. Their
tone sounded like it was being clipped, like a guitar run
through fuzzbox. They would be easy copy if the tone was
clean.
So, to capsulize,
Only a few strong, or weak signals: clean and pure tone.
Lots of middling signals, crackly/clipping noise on top of
the CW tones.
Ideas?
-Chuck Harris - WA3UQV
|
Re: QMX+ No transmit or receive
You are absolutely correct Ludwig and you explained why. ?I also read it, but there is a difference between reading and comprehension. The fault is mine, I scanned the thread saw BS170 and didn’t make the correlation. ?When I sat down and tried to work out how the modulator functioned, I’d forgotten all about it. 73 Nick G0OQK
|
Re: QCX Mini 20m - help please ??
I had that rf problem running my qcx+ on the deck with a 20m efhw. If it is rf and If its a single band antenna,? a 1:1 choke balun 1/4 wave from the antenna will help a remove reflected rf getting in. If that's not convenient, from there each half wave down the feedline will also work.You want the balun to be located at a high current position along the feed line for best efficiency. Or just stick it any convenient place and see if it works before cutting coax. An "ugly balun" could work too.
--
regards,
Bryan, N0LUF
|
I'm using a 3S LiPo battery with mine (which is 12.6V at full charge) and I have a single diode in series to drop the volts to 12.0 and a little less as it discharges.? Works for me and should be a safe combination.
|
Re: Full Band Coverage QMX+
So look in the configuration menu, and see if the offending bands are flagged as disabled.
Roger
Get BlueMail for Android
I did a factory reset, selected 160-6 in the menu, saved that, and then backed out of the menus. ?When I cycle through the bands to operate, the radio stays on 80-20, I don’t see any of the other bands. ?I’ll try again in a couple of days, and post. ?Thank you for the helpful responses, such a great group here!
?
--
Trevor (N0YMA)
|
The Magic of Making PCBs Loaded With SMDs
Here's an example of one factory's process.? It's much more complex than I imagined.
?
https://www.youtube.com/watch?v=24ehoo6RX8w
|
Re: QMX+ No transmit or receive
On Sun, Dec 29, 2024 at 09:50 AM, Nick Garrod wrote:
I hadn’t appreciated that Q507 was switched on during receive.
Like I wrote in post /g/QRPLabs/message/135488 on Dec. 18 in this thread. Why not reading?
?
73 Ludwig
|
Re: Full Band Coverage QMX+
I did a factory reset, selected 160-6 in the menu, saved that, and then backed out of the menus. ?When I cycle through the bands to operate, the radio stays on 80-20, I don’t see any of the other bands. ?I’ll try again in a couple of days, and post. ?Thank you for the helpful responses, such a great group here!
?
--
Trevor (N0YMA)
|
Scott, you can use that power supply if it's well regulated by adding a buck converter or a couple of diodes to reduce the voltage seen by the QMX.
--
73, Dan - W2DLC
|
Re: a soldering station option
I picked up the Weller 70 Watt Digital Soldering Station | WE1010NA.? I've been happy with it.? ?I like the tip selections available for it.? I can get most from Amazon delivered in a day or so.? I bought a cheaper desoldering gun, S-995a, and the tip clogs easily.? Took weeks to get a replacement.? It's all an exercise in patients.
|
Re: A question about L401 windings QMX v1 vs QMX v2
The difference between the V1 and V2 circuits seems to be that the V1 circuit uses both of the multiplexers in IC402 while? the V2 circuit only uses one multiplexer. This doesn't seem to be the same as the 20m fix, this is implemented in both but with different numbers of turns on L401? Could the change to the multiplexer circuit means that different inductances are required?
?
Chris, G5CTH
|
Re: QMX+ No transmit or receive
Thank you Stan! I appreciate you offering me this valuable information. I am going to get another chance later today to do some more testing and I am going to put this info to good use. Nothing like specs in real time to give some sense of direction.
On Sat, Dec 28, 2024 at 02:57 PM, Nick Garrod wrote:
someone else who has looked at the voltages in the modulator under normal operation
Here are some waveforms of Q507 from a working QMX+.? All are of the transmission of a single 30wpm 'dit'.
My supply voltage was about 11.7V, scales as labeled, with zero baseline one line up from bottom.
?
1) note that at idle receive, Q507-gate should be at about 6V (1/2 V+), Q507 is 'on' and Q507-drain is at 12V (V+).? DAC is high (3V).
2) at start of transmit, cpu drives DAC low, to start the transmit shaping by starting the power at 0.? DAC low turns on Q513, which ends up turning off Q514, which makes Q507-gate go high, turning it off, making Q507-drain go to 0V.
3) cpu ramps up DAC, which gradually pulls Q507-gate back down toward 6, increasing the Q507-drain voltage to near full value.
4) to end the transmission, cpu ramps down DAC again, causing the reverse shaping happen at the end of the 'dit'.
5) after transmission is complete, cpu sets DAC high again during receive.
6) note that the Q507-drain voltage modulates the amplitude of the drive signal from 0-100%, but that drive signal comes from the TX and CLK2 signal into the BS170 drive transistors, so when TX is not asserted, no transmit occurs regardless of the Q507-drain voltage.
?
Screenshots:
Yellow trace is Q507-gate in all plots.?
Blue trace is Q507-drain in #1, #2.? Blue trace is DAC in #3.? Blue trace is RF out in #4.
Trigger is near left of screen at start of 'dit' (rising edge of Q507-gate).
|
Re: QMX, one of op-amps self-oscillates at 25 MHz (or picks up the system clock?)
So if you have a 2 channel scope verify if the 25MHz is in perfect phase with the system clock. ? If so I’d be looking for a solder blob or a faulty decoupling capacitor?
|
I assume you mean you sprayed it with freeze spray?
?
To check the thermal theory do some tests?
?
cold
on 30s to 1 min. Off 2 mins on again ?etc?
what power profile ?is seen
?
if the profile looks like the problem takes time to go away it is likely thermal
?
- remember that spraying the IRF plastic cases is a very poor way to cool them?
?
-also I’ve found that supplied plastic flanges were sometimes too long for the heatsink pads supplied. ?Great (unwise with alloy screws) tightening force was needed to actually get significant pressure ?
Between the IRFs tabs and the pad/heatsink
?
Sanding the flanges down a fraction of a mm made everything ok ?
?
?
?
|
Bryan,
?
What is the current draw of the amp when transmitting?? Does it change when it acts up?
?
73
Evan
AC9TU
|
Scott from the QMX manual?
?
?
PLEASE READ THE BASIC ASSEMBLY AND USE INSTRUCTIONS IN
THIS MANUAL VERY CAREFULLY BEFORE APPLYING POWER TO
THE BOARD!
SUPPLY VOLTAGE: 12.0V maximum is recommended, particularly
when you are using Digi Modes with high duty cycle, or an uncertain
antenna SWR situation. 12V supply means 12V (not 13.8V type
generic transceiver supply!). If you wish to drop the supply voltage
the easiest way is some series 1N4001 (etc) rectifier diodes, each
one will drop the voltage about 0.6V.
|
Re: QMX+ No transmit or receive
Thank you Stan, that is most helpful and a very clear explanation. I hadn’t appreciated that Q507 was switched on during receive. Cheers Nick
|
Good evening,
I understand the WMX’s voltage sensitivity, and especially the fact that most Lithium batteries have a nominal “12v” of 13.5vdc +/-.
The power source I’m looking at runs at 12.8vdc.? I’m wondering though: is 12.8v “too much”?
Thanks. ?
Scott?
|
Re: Diagnostics fail with no BS170s installed
#qmx
On 28/12/2024 17:04, Chris via groups.io wrote: I attacked the IC one side at a time, flooded the 5 pins with solder and lifted the pins a little. Chris, I've not known that way of removal to have been described before for SMD! I'd have thought it the best way to ruin a board but you succeeded:-) To me your "tricky bit" is fairly easy, but without spending lots of money on specialised equipment we all find ways that best suit our personal techniques. And, a first for the strange MS5351 fault, it seemed so unlikely. 73 Alan G4ZFQ
|
UPNORTHOF48 :
Daniel Conklin
Bob Brown
Bruce Akhurst