Date   

Re: Drive level for 10W HF Linear PA?

ajparent1/KB1GMX
 

Ken,

10db drop in drive should net 10 db drop in output, that is the
definition of linear.

If you need more drive or power out is low you did something
wrong.  Generally voltage tests are not useful as they can
indicate a lot if not at all functional but in the poor power
state its meaningless.

For example if tT203 has one side not connected the
DC at the IRF510 drains will be fine but the RF power poor.
If either side if T201 or T202 is not connected or worse
has a internal short DC will likely be fine but the power
out will be poor.

FYI if the heatsink is getting hot but T203/T204 not
then the problem is in one or both.  The source of the
heat is the IRF510s but its  case of driving them to
heat up but they can't deliver the power.

It really is important to read and understand the manual
and be able to understand the schematic.

Allison
-------------------------------
Please reply on list so we can share.
No private email, it goes to a bit bucket due to address harvesting


Re: Drive level for 10W HF Linear PA?

ajparent1/KB1GMX
 

Failure to see the correct gain says you munged one of more transformers.
Possibly have one of more parts incorrectly inserted or soldering issues.
Waveform being bad at low power confirms that.

I will make clear from testing the gain should be as stated or maybe a db higher.
IF the transformers are not correct power will be low and other issues will exist.

0dbm in should net 26dbm out is all is calibrated.  Expected power gain is 26-28db.
at 13.8V 1db compression is over 10W (12.1 was typical or my four).
At 12.0V the power will be lower, typical was 9.8W to 10.1W.

To get 40dbm (10W) you must have 14dbm of drive (about 25-35mW).
NOTE the driver must be able to do more than that as the input is not
50 ohms (not intended to be).  The previous stages must accept it.

the amplifier is a two stage system and if the driver stage is bad
(soldering, transformers, parts error) the output stages will not 
make up for it and the ills mentioned will be evident.

Common errors:
+ITs easy to scratch the enamel and have internal sorts on the small cores.
  T201 and T202.  I use a fine stone to take out edged in the holes before
  winding as often they gave molding edges that will nick or scrape the enamel.
+ITs also easy to put the output core T204 in backward (3:2 instead of 2:3).
+If the DC source core t203 is munged or one side wrong or not connected the 
  DC tests will appear right but power/gain/waveform will all be bad.

IT is possible to stuff a lead into a though hole that is not the intended
target.  You will believe its right but it works badly.

There is no if this happens do that trouble shooting.  As in many
closely coupled RF systems an error here may show up there.
The trick if any is make no errors.

One sote is that if you can set the bias the output stages can still be
be wrong. (t203 and t204).

HINT:  for coax to and from I use RG316 teflon.  I fist strip the
outer jacket and then tin the braid to make it solid. Using a
precision knife I score the braid and then snap it and slid it off. 
That give me a nice braid and exposed center insulator/conductor.
For the braid I wrap 2 turns of #22-24 tinned wire around it to give me
a nice wire end for the braid.  Then I strip the center conductor.
If I keep all the dimensions small this works for me to about
2ghz or more.    I find the silver plated RG316 is easier to
solder and does not melt like RG174. 


Allison
-------------------------------
Please reply on list so we can share.
No private email, it goes to a bit bucket due to address harvesting


Re: QCX-mini failure on new build

Timothy Freeze
 

wow, you sure found that quick. Very impressive! 

Thanks and let me know about your costs and how to pay you. I just finished building an 80 meter mini and it works great right after assembly. That was a relief!!

Thanks again. 
Tim

Sent from my Verizon, Samsung Galaxy smartphone
Get Outlook for Android


From: QRPLabs@groups.io <QRPLabs@groups.io> on behalf of Julian N4JO <n4jo@...>
Sent: Friday, July 2, 2021 4:50:04 AM
To: QRPLabs@groups.io <QRPLabs@groups.io>
Subject: Re: [QRPLabs] QCX-mini failure on new build
 
Hi Tim, the unit arrived today, I got around to looking at it late tonight, and I've found the problem with the display: there should have been 5V on pin 2 of the display connector to power the display electronics, but it was missing (your measurement was on the right track!). The connection looked good, but was faulty on the "via" - a feature that connects a pad on the top side of the board to the corresponding pad on the bottom - something you wouldn't have expected to think about, as it looked good visually. The sneaky part was that the voltage looked reasonable when the display was attached, but that was only bleed-through from the signal connections back out the supply line, not real supply voltage: a particularly insidious error that caught me once, many years ago. Taking the display off and looking at the voltages on the pins, it showed up immediately.

I reheated the connection which got it working, but tomorrow I'll think about how to make that connection more robust. I don't know how sound it is, and as it's hidden under the display connector, there's no way of knowing. I'll run a separate wire in parallel with the track from the point that supplies power to that point on the display connector (it's pin 5 on the nearby ISP connector, actually), just to be sure.

Meanwhile, I performed a factory reset and configured it for 20m, and it seems to respond correctly to control inputs. I'll run through the calibration procedure tomorrow to make sure everything else is good, and hook an antenna to it to compare sensitivity to mine. Then I'll send it back to you.

Lastly...
I notice that there is the infamous scratchy noise on tuning with no signal, as reported by other users. I found, while tracing out the missing power problem above, that there is no L6 filter in the Vcc line from 5V like there is in the Classic and the plus; I wonder if that has anything to do with the issue? I'll leave that question to others here to investigate and/or for Hans to ponder.

--

Julian, N4JO.


Re: Drive level for 10W HF Linear PA?

ajparent1/KB1GMX
 

The drive level was accomplished in my testing with external Qbit amp
with its drive level of 0dbm (from sa).

without the block amp 0dbm give you under a watt.

you need 25 to 35mw...

Allison
-------------------------------
Please reply on list so we can share.
No private email, it goes to a bit bucket due to address harvesting


Re: QCX Mini 20 - occasional lockup

K9NUD-Steve
 

I don't recall seeing a "G", but I need to confirm. My antenna is a ground-mounted 6BTV with buried radials, buried LMR400 coax, and , low SWR. I could have added a current balun but it doesn't seem to need it. 


Re: Assembled QCX-mini and QCX+ kits

Michael.2E0IHW
 

And very reliable, well worth the wait.
Thanks, Hans and team!

Michael
. . . . . . . . . . . .

On 02/07/2021 09:10, Hans Summers wrote:
Hi all

As you may know, QCX-mini and QCX+ are also available as ready assembled, tested, adjusted (including power output optimization), calibrated radios, optionally including enclosure and TCXO. See:

We had at one point a very large backlog of orders due to the huge initial rush on QCX-mini kits in December; the estimated waiting time was 10 weeks but many orders took more than 4+ months for us to get assembled etc and shipped. 

The team here have been working very hard to work on this mountain of assembled kit orders and have everything well under control now; the estimated waiting time for assembled QCX+/QCX-mini orders placed today is 4 weeks. 

I am publishing a waiting list of the assembled kit orders here: http://qrp-labs.com/qcxmini/assembled.html

73 Hans G0UPL
http://qrp-labs.com



Re: Drive level for 10W HF Linear PA?

KEN G4APB
 

Hi Geoff, 
No, T203 does not get hot. I ran the PA for two minutes continuously to test. The heatsink gets quite warm, but not too hot to touch. My 100mhz scope says 48v p-p which works out a 5.8 watts at 13.6v supply for a reasonable but not true sinewave waveform, Gonna have to put it through an LPF though, so that will loose a bit too.

I measured voltages all around, bias to the Bs170 drivers is 2.72 V, bias to the finals is 4.32 on one and 4.24 on the other, both were set for 125mA. Supply line is reaching all devices.  Power supply is 30Amp capable.
Dropping the input drive level by10dB drops the output level by 20dB, I expected only a 10dB drop in output, so still something not quite right yet, or am I misunderstanding something still, hence my original question on drive levels?

73 Ken G4APB


Re: QCX-mini failure on new build

 

Hi Tim, the unit arrived today, I got around to looking at it late tonight, and I've found the problem with the display: there should have been 5V on pin 2 of the display connector to power the display electronics, but it was missing (your measurement was on the right track!). The connection looked good, but was faulty on the "via" - a feature that connects a pad on the top side of the board to the corresponding pad on the bottom - something you wouldn't have expected to think about, as it looked good visually. The sneaky part was that the voltage looked reasonable when the display was attached, but that was only bleed-through from the signal connections back out the supply line, not real supply voltage: a particularly insidious error that caught me once, many years ago. Taking the display off and looking at the voltages on the pins, it showed up immediately.

I reheated the connection which got it working, but tomorrow I'll think about how to make that connection more robust. I don't know how sound it is, and as it's hidden under the display connector, there's no way of knowing. I'll run a separate wire in parallel with the track from the point that supplies power to that point on the display connector (it's pin 5 on the nearby ISP connector, actually), just to be sure.

Meanwhile, I performed a factory reset and configured it for 20m, and it seems to respond correctly to control inputs. I'll run through the calibration procedure tomorrow to make sure everything else is good, and hook an antenna to it to compare sensitivity to mine. Then I'll send it back to you.

Lastly...
I notice that there is the infamous scratchy noise on tuning with no signal, as reported by other users. I found, while tracing out the missing power problem above, that there is no L6 filter in the Vcc line from 5V like there is in the Classic and the plus; I wonder if that has anything to do with the issue? I'll leave that question to others here to investigate and/or for Hans to ponder.

--

Julian, N4JO.


Assembled QCX-mini and QCX+ kits

Hans Summers
 

Hi all

As you may know, QCX-mini and QCX+ are also available as ready assembled, tested, adjusted (including power output optimization), calibrated radios, optionally including enclosure and TCXO. See:

We had at one point a very large backlog of orders due to the huge initial rush on QCX-mini kits in December; the estimated waiting time was 10 weeks but many orders took more than 4+ months for us to get assembled etc and shipped. 

The team here have been working very hard to work on this mountain of assembled kit orders and have everything well under control now; the estimated waiting time for assembled QCX+/QCX-mini orders placed today is 4 weeks. 

I am publishing a waiting list of the assembled kit orders here: http://qrp-labs.com/qcxmini/assembled.html

73 Hans G0UPL
http://qrp-labs.com


Re: Using 40 mtr QCX+ on 30 mtrs

ON7DQ Luc
 

We just had a discussion about this, but it was for using a QCX for 17m on 15m
Read it all here 
https://groups.io/g/QRPLabs/topic/qcx_mini_for_17m_usability/83838389?p=,,,50,0,0,0::recentpostdate%2Fsticky,,,50,2,0,83838389
It contains links to my blog where I have more information, with measurements and tips.

Conclusion:
You can very well use a QCX on the "one lower" band, but not on the "one higher" band.
So your 40m rig is OK for 60m use, not for 30m.
In receive , you will only loose abt. 6 dB, so 1 S-point. One trick is to tune your 40m bandpass (T1/C1) a bit on the low side, so when you have found the peak, ADD some capacitance at C1..
TX power will be abt 80% of your power on 40m.

If using a tuned antenna (not resonant on the second harmonic) , you don't even need an extra LPF.
Your antenna must have an acceptable SWR on 60m anyway.

73 de Luc ON7DQ


Re: Drive level for 10W HF Linear PA?

geoffrey pike
 

Hi Ken,
This sounds familiar to me, can you be 100% certain that T203 is wound correctly (does it run warm/hot?)
cheers
Geoff
GI0GDP

On Thursday, 1 July 2021, 17:23:38 BST, KEN G4APB via groups.io <lfoofui.nbz42@...> wrote:


Hi  Guys,
I am a bit disappointed Hans did not include RF connectors for input/output on this, guess they are not needed in the QSX.
Anyway, I managed to short-circuit my input coax. It is a bit difficult soldering both the coax braid AND the earth wire for the trigger signal to a single pad. There are no input/output capacitors so both ports measure as short circuit anyway, so not easy to spot a real short.

However, I still don't see the 26dB gain. For 14dBm (25mW) from my U3S via step attenuator, I only see 35dBm (approx 3 watts) output for just over 1amp input at 13.6volts. Waveform goes horrible if i push the drive up just a little.
Full strip down, IPA clean, and inspection under spyglass shows no shorts, re-soldered all the inductors but theytest as the Manual directs.
Is there a 'troubleshooting' section for this anywhere?

73 Ken G4APB


Re: Using 40 mtr QCX+ on 30 mtrs

Bill Cromwell
 

Hi,

The hams who have reported success install the output low pass filter for the higher frequency band and then used an outboard filter for the lower frequency band. That takes care of output spectral purity but that input transformer ahead of the receiver can only be peaked for one band or the other. I can't see how the receiver will perform well on both bands. A compromise of that transformer will diminish BOTH bands. Some kind of bandswitching would have to be used with two transformers. In that case may as well bandswitch the output filters too. If doing that, buying a multiband rig might be a much better choice.

Just my own opinion and the reasons for it.

73,

Bill KU8H

bark less - wag more

On 7/1/21 2:16 PM, Ed Kwik via groups.io wrote:
Phil,
My advice to you and others that want to operate a qcx on more than one band is:

It is unrealistic to think that a single band rig should be used on any other band than it was designed for.

That is not to say that it cannot be done. But a certain amount of technical knowledge and the ability to test your modifications to insure your operation is within all the required specifications like spurious signals, harmonic signal levels and the like.

The qcx design packs a lot of performance in a very elegant design that is highly optimized. Deviating from the design can cause all sorts of undesirable behavior.

Ed
AB8DF




Re: Using 40 mtr QCX+ on 30 mtrs

wa8yan.radio
 

Hi Ed... thanks for your comments.  You are confirming what I have been concerned about.  Thanks again.
Phil WA8YAN




Sent from my Verizon, Samsung Galaxy smartphone


-------- Original message --------
From: "Ed Kwik via groups.io" <ekwik@...>
Date: 7/1/21 2:16 PM (GMT-05:00)
To: QRPLabs@groups.io
Subject: Re: [QRPLabs] Using 40 mtr QCX+ on 30 mtrs

Phil,
My advice to you and others that want to operate a qcx on more than one band is:

It is unrealistic to think that a single band rig should be used on any other band than it was designed for.

That is not to say that it cannot be done. But a certain amount of technical knowledge and the ability to test your modifications to insure your operation is within all the required specifications like spurious signals, harmonic signal levels  and the like.

The qcx design packs a lot of performance in a very elegant design that is highly optimized. Deviating from the design can cause all sorts of  undesirable behavior.

Ed
AB8DF






Re: connection qlg2 to qcx #qcx #qlg

Gerald Ball
 

Hi Rob.                                                                       Yes I use a 6 pin din to connect qlg. Very convenient.                                                               Gerry G4Ojf                                  


--
gerry


Re: Using 40 mtr QCX+ on 30 mtrs

Ed Kwik
 

Phil,
My advice to you and others that want to operate a qcx on more than one band is:

It is unrealistic to think that a single band rig should be used on any other band than it was designed for.

That is not to say that it cannot be done. But a certain amount of technical knowledge and the ability to test your modifications to insure your operation is within all the required specifications like spurious signals, harmonic signal levels and the like.

The qcx design packs a lot of performance in a very elegant design that is highly optimized. Deviating from the design can cause all sorts of undesirable behavior.

Ed
AB8DF


Using 40 mtr QCX+ on 30 mtrs

wa8yan.radio
 

I know I have read other comments here about using a 40 mtr rig on 30 mtrs but i wasn't paying attention to it at that time.  Now I'm getting close to getting a QCX+ 40 on the air and the idea of 30 mtrs has been revived in my head.  Would love to hear of others experience doing this.  How well does it work, what are the drawbacks and/or hazards, should i try it or leave it alone, what are your experiences doing this, etc.
Thanks in advance
Phil WA8YAN



Sent from my Verizon, Samsung Galaxy smartphone


Re: Drive level for 10W HF Linear PA?

KEN G4APB
 

Hi  Guys,
I am a bit disappointed Hans did not include RF connectors for input/output on this, guess they are not needed in the QSX.
Anyway, I managed to short-circuit my input coax. It is a bit difficult soldering both the coax braid AND the earth wire for the trigger signal to a single pad. There are no input/output capacitors so both ports measure as short circuit anyway, so not easy to spot a real short.

However, I still don't see the 26dB gain. For 14dBm (25mW) from my U3S via step attenuator, I only see 35dBm (approx 3 watts) output for just over 1amp input at 13.6volts. Waveform goes horrible if i push the drive up just a little.
Full strip down, IPA clean, and inspection under spyglass shows no shorts, re-soldered all the inductors but theytest as the Manual directs.
Is there a 'troubleshooting' section for this anywhere?

73 Ken G4APB


Re: BSS123 Where to buy?

Christopher Maness
 

That would be great!  TU Chris KQ6UP 

On Thu, Jul 1, 2021 at 11:14 AM Richard - N1RBD <groups.io@...> wrote:
I ordered 5 when I needed to replace mine.  Let me see if I can scrounge the spares up.  If I can, I'll drop a couple in the mail to you.


Re: BSS123 Where to buy?

Richard - N1RBD
 

I ordered 5 when I needed to replace mine.  Let me see if I can scrounge the spares up.  If I can, I'll drop a couple in the mail to you.


Re: QCX Mini 20 - occasional lockup

Richard - N1RBD
 

Does the display show a "G"?

17241 - 17260 of 84397