Topics

QCX 20M with very low power


chmims <chmims@...>
 

I have recently completed assembly of a QCX 20M.  All the voltages appear to be reasonably appropriate,  there is good continuity L1-L3, T1 voltages are appropriate.  However RF output  measures only about 0.12 W.  After removing  2 turns from L3, RF output now measure about 0.2W and is stable.  Drains on Q1, Q2, Q3 all measure 0.25 W.  With so little help with removing turns on L3 I am hesitant to remove more turns.  Any suggestions as to what might help would be appreciated.

Charles
KG5ZLH


mike.carden
 


On Thu, May 7, 2020 at 4:38 AM chmims via groups.io <chmims=yahoo.com@groups.io> wrote:
I have recently completed assembly of a QCX 20M.  All the voltages appear to be reasonably appropriate,  there is good continuity L1-L3, T1 voltages are appropriate.  However RF output  measures only about 0.12 W.  After removing  2 turns from L3, RF output now measure about 0.2W and is stable.  Drains on Q1, Q2, Q3 all measure 0.25 W.  With so little help with removing turns on L3 I am hesitant to remove more turns.  Any suggestions as to what might help would be appreciated.

Charles
KG5ZLH


chmims <chmims@...>
 

I went over the low power FAQ.  L1, L2, L3, & L4 all have good continuity.  Rechecked and reheated the solder on C25-28,30.  Then pulled and decreased by two turns the windings on L1, L2, and L3.  The results was no increase of RF output it remaining 0.2 W.  However the drains on Q1,Q2,Q3 now show 1.0W.

I wondering if some component is at fault?

Charles
KG5ZLH


K5DH
 

Replace the two 390 pF caps at C25 and C26 with silver mica caps.  In addition to taking a turn or two off the LPF coils, I've had to replace those caps on two 20m QCXs to get the power up where it should be.  There have been many discussions here on the group on this subject.  Search and learn!  :) 
--
73/72,
Dean K5DH
20m and 40m QCXs


ajparent1/KB1GMX
 

Something else is very wrong and usually if the caps are an issue the
power is in the 1W out range or more.  

IF anything check the DC voltage at the emmitter and collector of Q6
it should very near 12V when transmitting.  

Shorts to ground in the TX section could be a problem.

Other possible issue is the fets Q1-3 are in back ward or rotated 120 degrees.

Allison
-------------------------------
Please reply on list so we can share.
No direct email, it goes to bit bucket due address harvesting in groups.IO


Charles Mims <chmims@...>
 

Many thanks. I wondered if I might have the bad capacitors.  But my power did not slowly increase. 

Charles
KG5ZLH


On May 7, 2020, at 6:06 PM, K5DH <mustang_maniac@...> wrote:

Replace the two 390 pF caps at C25 and C26 with silver mica caps.  In addition to taking a turn or two off the LPF coils, I've had to replace those caps on two 20m QCXs to get the power up where it should be.  There have been many discussions here on the group on this subject.  Search and learn!  :) 
--
73/72,
Dean K5DH
20m and 40m QCXs


George Korper
 

Look at all the PCB traces with a strong magnifier. Or bring in a ground with a wire near the LPF Simply put it may well be a poor ground path which shows up on 20. 


On Thu, May 7, 2020, 7:43 PM Charles Mims <chmims@...> wrote:
Many thanks. I wondered if I might have the bad capacitors.  But my power did not slowly increase. 

Charles
KG5ZLH


On May 7, 2020, at 6:06 PM, K5DH <mustang_maniac@...> wrote:

Replace the two 390 pF caps at C25 and C26 with silver mica caps.  In addition to taking a turn or two off the LPF coils, I've had to replace those caps on two 20m QCXs to get the power up where it should be.  There have been many discussions here on the group on this subject.  Search and learn!  :) 
--
73/72,
Dean K5DH
20m and 40m QCXs


Timothy Fidler
 

"Many thanks. I wondered if I might have the bad capacitors.  But my power did not slowly increase. "

Power increase or decrease ?  

If they were lossy I'd expect power out to decrease as they heated up. (??)

Alison's idea re Mosfets in backwards is worth looking at but it is very hard to make that error with a soldermask on the PCB.
The interesting thing is that some FETs (junction FETS) are specified to work with Drain and source  connections reversed.  I have never heard of a mosfet where this is the case . If you did such a connection for a BS170  the bulk substrate connection would be grossy above the gate quiescent voltage and the static protection diode would blow the first time you applied power (in the QSX circuit).  That diode is supposed to zenner break under high impedance sources a lover 60V  it is not expected to withstand hundreds of mA of forward current.  




Timothy E. Fidler : 
Telephone Whangarei   022  691 8405
e: Engstr@...



----- Original Message -----
From:
QRPLabs@groups.io

To:
<QRPLabs@groups.io>
Cc:

Sent:
Thu, 7 May 2020 19:43:18 -0500
Subject:
Re: [QRPLabs] QCX 20M with very low power


Many thanks. I wondered if I might have the bad capacitors.  But my power did not slowly increase. 

Charles
KG5ZLH


On May 7, 2020, at 6:06 PM, K5DH <mustang_maniac@...> wrote:

Replace the two 390 pF caps at C25 and C26 with silver mica caps.  In addition to taking a turn or two off the LPF coils, I've had to replace those caps on two 20m QCXs to get the power up where it should be.  There have been many discussions here on the group on this subject.  Search and learn!  :) 
--
73/72,
Dean K5DH
20m and 40m QCXs


Charles Mims <chmims@...>
 

My mistake I meant decrease. My seemed stable as low as it was. 


On May 7, 2020, at 8:07 PM, Timothy Fidler <engstr@...> wrote:


"Many thanks. I wondered if I might have the bad capacitors.  But my power did not slowly increase. "

Power increase or decrease ?  

If they were lossy I'd expect power out to decrease as they heated up. (??)

Alison's idea re Mosfets in backwards is worth looking at but it is very hard to make that error with a soldermask on the PCB.
The interesting thing is that some FETs (junction FETS) are specified to work with Drain and source  connections reversed.  I have never heard of a mosfet where this is the case . If you did such a connection for a BS170  the bulk substrate connection would be grossy above the gate quiescent voltage and the static protection diode would blow the first time you applied power (in the QSX circuit).  That diode is supposed to zenner break under high impedance sources a lover 60V  it is not expected to withstand hundreds of mA of forward current.  




Timothy E. Fidler : 
Telephone Whangarei   022  691 8405
e: Engstr@...



----- Original Message -----
From:
QRPLabs@groups.io

To:
<QRPLabs@groups.io>
Cc:

Sent:
Thu, 7 May 2020 19:43:18 -0500
Subject:
Re: [QRPLabs] QCX 20M with very low power


Many thanks. I wondered if I might have the bad capacitors.  But my power did not slowly increase. 

Charles
KG5ZLH


On May 7, 2020, at 6:06 PM, K5DH <mustang_maniac@...> wrote:

Replace the two 390 pF caps at C25 and C26 with silver mica caps.  In addition to taking a turn or two off the LPF coils, I've had to replace those caps on two 20m QCXs to get the power up where it should be.  There have been many discussions here on the group on this subject.  Search and learn!  :) 
--
73/72,
Dean K5DH
20m and 40m QCXs


Alan G4ZFQ
 

Charles,

My mistake I meant decrease. My seemed stable as low as it was.
You were correct the first time.
Low power (~1 watt) with power increasing as the capacitors get warm is the symptom reported by those who have had faulty capacitors.
Note that replacements can be the same type. Get from a reliable supplier. Hans has changed his supplier so the problem should not occur on new kits.

73 Alan G4ZFQ


Mike Easterbrook
 

Just to adding my experiences with QCX20/low power -  & some help still needed.

I had a spare QCX 30 kit with V4 board which I  have now built up for 20m using latest updates i.e. R41/42 150R and 1.2K, C31 at 2.2uF;  R21 at 10K, C25/26 390pF ex Vishay/Digikey.

RF out per SWR/Power Meter (has been side-by-side compared with known(?) to be accurate meter in 1.0 - 5.0 W range) and found "similar". Following measurements made at 13.8 V at the connector i.e. before D3:


L1/2/3 wound per manual   ~1.0W
L1/2/3 wound to manual inductance tables using recently purchased Peak LCR40  ~1.5W
L1/2/3 adjusted for max output by squeezing/distancing turns on toroid and subtracting turns if necessary to peak output  ~ 4W  (final turns no. L1=13, L2=15, L3=13)  Observation was that only L3 adjustment had significant effect on power out. Reducing L3 to 12 turns gave output~5W.  However at this level there some evidence of intermittent instability/oscillation accompanied by immediate increase in total current from ~540 to ~990ma rapid heating of the finals and maybe 50% less power out.  Sending a series of dits about 1 in 20 were of much reduced power. The problem disappeared on reducing supply volts to 12 or increasing L3 to 14 turns. (i.e. keeping power out to below 4W)

On the assumption that the instability was in the key shaping section, I looked at the output with the scope and found complete absence of shaping.  I compared this with my QCX40 (on a V5 board built according to manual but with L1/2/3 turns reduced to match quoted impedance (17/22/17). This is measuring 5W out at13.8V with apparent good stability and OK shaping (see attached pics)  Grateful for any ideas on where to investigate the key shaping - component values and soldering look good.

Maybe unrelated incident - the LCD developed a fault - screen contrast suddenly very low but can be adjusted via R47 but random gibberish characters displayed and no communication with IC2. If I transfer this LCD to another QCX the fault remains.  Fault disappears with same board and a working LCD.  Not aware that the LCD was subjected to any trauma other than multiple plug-in/plug-outs while doing the L1/2/3 adjustments - anyway need to replace.  Anyone got a part no. for this item so I can order locally?  Don't want to add to my stock of wrongly ordered displays from old (non-QCX) project (pin outs not lining up, wrong colour, wrong backlight etc etc) 

Thanks
Mike 9M2LXM


On Fri, May 8, 2020 at 8:11 AM ajparent1/KB1GMX <kb1gmx@...> wrote:
Something else is very wrong and usually if the caps are an issue the
power is in the 1W out range or more.  

IF anything check the DC voltage at the emmitter and collector of Q6
it should very near 12V when transmitting.  

Shorts to ground in the TX section could be a problem.

Other possible issue is the fets Q1-3 are in back ward or rotated 120 degrees.

Allison
-------------------------------
Please reply on list so we can share.
No direct email, it goes to bit bucket due address harvesting in groups.IO


ajparent1/KB1GMX
 

A note:  L4 is not a choke for supplying DC its part of the C30/L4 pair that needs
to be tuned to in this case about 14.03mhz.

Seems to be an aversion to making sure that's correct too.

Allison
-------------------------------
Please reply on list so we can share.
No direct email, it goes to bit bucket due address harvesting in groups.IO


Mike Easterbrook
 

Hi Allison
Does the tuning of L4 have possible relevance to the key shaping &/or instability I am seeing!
I guess power out is easily appreciated (if not easily measured precisely) thus the obsessions with the LPF - I would not have gone looking for a shaping problem but for the instability & having a scope available.
I'm actually happy to keep power out @ 3.5 W to avoid the Instability but now I'm aware of the keying issue am keen to understand it.
Thanks for input
Mike 9M2LXM


chmims <chmims@...>
 

I have now replaced c25 and c26 with silver mica caps.  Unfortunately it has been of no benefit.  I only get 0.20 watts of RF power.  Not sure where to go at this point.  The measurements per the manual  continue to be satisfactory except for the RF power and L1-3, L4 have good continuity.  I have noted when I key down the power goes from 0.01 to 0.20 watts.  I have noticed even when I release the key the measurement stays at 0.20 watts.  The drains on Q1-3 do show 1.0 watts and they seem to hold that reading also when the key is released.  On Q6 the emitter and collector both show 12.26 v.  I have Han's 50 ohm dummy load and it measures about 3.5 v with key down.  I'm open to any idea.

Charles


Jim Allyn - N7JA
 

On Mon, May 18, 2020 at 12:59 PM, chmims wrote:
I have noticed even when I release the key the measurement stays at 0.20 watts.  The drains on Q1-3 do show 1.0 watts and they seem to hold that reading also when the key is released.  On Q6 the emitter and collector both show 12.26 v.

Your transmitter is keyed up all the time, if there is power when the radio is unkeyed.  Possibly shorted Q6 or Q4.


Alan G4ZFQ
 

Your transmitter is keyed up all the time, if there is power when the radio is unkeyed.  Possibly shorted Q6 or Q4.
Also, no power out may well indicate blown BS170s.

73 Alan G4ZFQ


chmims <chmims@...>
 

I now have fair output, 2.5W.  I did replace Q1-3 BS170 transistors.  This did not help. I still had 0.20 W RF power and which persisted  even after I released the key.  I then elected to replace Q6 and Q4.When I plugged the QCX again I noticed it kept streaming a T across the lower screen.  So before checking the RF power I decided to just resolder the key.

As noted my QCX is now putting out 2.5W.  Unfortunately I'm not sure if it is from replacing Q6 and Q4, or resoldering the key.  Anyway thanks for the help.

Charles
KG5ZLH


N3MNT
 

The "T" scrolling across the bottom of the screen is the CW decoder decoding noise so no worry.  Also 2.5W is what most people are seeing at 20M.  So sounds like you are there.