Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
#QMX #HighBand #LowPower QMX High-Band Working but with Low Power out
#qmx
#HighBand
#LowPower
|
I just completed my 3rd QMX build, this being my first High-Band QMX. It is a 12 Volt 2023 Rev 2 board with 3 modifications done by me, the addition of the diode (surface mount) for potential parasitic oscillations, and the jumper for the PTT Out grounded functionality, and 3) replaced the power socket with a power cable with power-poles. It is running the _17 firmware with AGC enabled.
Everything appears to work correctly and it appears to be receiving OK ... except that I'm getting about half the power out that I should be expecting. What should I be looking at that I may have done wrong, or that should be adjusted? What other tests can I perform to determine the issue? Here's my data while the QMX is connected to a QRP-Labs dummy load: It's drawing 212 mA powered up and idle, which is high! And it feels warm (I'll look at it w/ infra-red camera next). My low-band QMX draws around 93 mA at idle. When transmitting (CW key down) on: 20M drawing 716 mA 17M drawing 775 mA 15M drawing 833 mA 12M drawing 740 mA 11M drawing 760 mA 10M drawing 755 mA Diagnostics Power Out into Dummy Load 12V power input BIAS SMPS (during transmit) 30.4 mA Duty Cycle 14% ? 20M 2.4W 10.1V SWR 1.09 17M 2.5W 10.1V SWR 1.22 15M 2.2W 10.1V SWR 1.40 12M 1.7W 10.1V SWR 1.60 11M 1.6W 10.1V SWR 1.70 10M 1.6W 10.1V SWR 1.75 ? ? Audio Filter Sweep ? ? ? ? ? ? ? ? ? RF Filter Sweep ? ? ? ? ? ? ? ? ? ? Low Pass Filter Sweep ? ? ? ? ? ? ? ? Image Sweep ? ? ? ? ? ? ? ? ? SWR Sweep (into QRP-Labs Dummy Load) ? ? ? ? ? ? ? ? ? 73 de Roy - KI0ER Littleton, Colorado USA |
|
Hello Roy,
It's drawing 212 mA powered up and idle, which is high! Indeed. Around 100 mA here. |
|
Another clue .... I found where all the excess current is going ... the PA transistors are very hot while the QMX has only been in receive (idle) mode!? ?
I've re-checked all my solder joints under strong magnification while also looking for some stray short, but nothing is popping out at me as problematic. Does this clue help narrow things down? I'm open to suggestions! -- 73 de Roy - KI0ER Littleton, Colorado USA |
|
Roy, Check the gate signals at the BS170 PA transistors. You should see zero volts in RX, and a 5V 50% duty cycle RF square wave in TX.? If you are using a DMM to measure, that TX signal will register as 2.5 volts, an averaged value. If you see voltages substantially different, something is wrong for sure. Either the driver circuit has a problem or an internal short has occurred at a PA transistor(s). GL JZ KJ4A? On Wed, Apr 3, 2024, 10:08?PM Roy - KI0ER <ki0er@...> wrote: Another clue .... I found where all the excess current is going ... the PA transistors are very hot while the QMX has only been in receive (idle) mode!? ? |
|
Roy,
please check also the voltage at each connection of T501 in RX mode and let me know. Reference to the transformer document at page 10 all the connections marked "P" should be on the same value below 1 V (0.6 V or so). Both "S" should show 0 V. If not, T501 is wrong or the voltage source of the PA. If the PA transistors are hot only from 100 mA additional current a PA voltage of 0.6 V is not enough I think (only 60 mW of heat power). If I would search for the transistors producing the heat I would do the following: Let the PA cool down (QMX off) and remove the Display and the steel screw, nut and washer. Switch the QMX on for a few seconds (don't need the Display) and after switch off check temperature of each transistor carefully with the finger tip (it could be hot!!!). Maybe some more seconds ON are necessary for enough temperature. 73, Ludwig |
|
Roy, Ludwig, If the fault is on the gate driver side, the excess current will be via the 5V SMPS. At the 12V feed it will appear less because of the SMPS transformation. Much more than 60mW could be in play. If that is the case, the driver chip IC503 should also be getting warm during RX. One or more BS170 would likely be found to be damaged as well. 73 JZ KJ4A? On Thu, Apr 4, 2024, 7:16?AM <DH8WN@...> wrote: One thing more. Because the PA is on in some way also in RX mode, please connect the dummy load the whole time. It's only as a precaution. |
|
John,
ok, If the current is flowing from 5 V trough the PA transistor, the current on 5 V could be around 200 mA and the heat power 1 W or so. This is much more than 60 mW and will clearly heat up the PA. It would be good to know the Duty cycle for 5 V (Terminal Diagnostics). If the fault current is flowing via the 5 V some of the outputs of IC503 have to be ON. So IC503 could be faulty in this case. But I guess, not the whole IC503 is faulty because the PA is working partially. An other possibility would be if the signal "TX" to IC503 is not LOW and so one pair of BS170 is ON the whole time. It depends also on the signal "CLK2" to IC501 and IC502. Are there pulses in RX mode? Not so easy to predict the faulty part. 73, Ludwig |
|
With display removed, I powered up my QMX and watched the PA transistors with an infrared camera. One of them lit up in the camera within seconds of power up. It was the top-left PA transistor that began warming up. I'll remove it to see if it's bad, shorted internally or something.
I'll do other tests later this morning. -- 73 de Roy - KI0ER Littleton, Colorado USA |
|
Ludwig, A good step towards figuring this out is to observe or measure the gate signals at each bank of BS170s. If it is not what is expected, we can begin to peel down the possibilities at that point. JZ On Thu, Apr 4, 2024, 10:02?AM <DH8WN@...> wrote: John, |
|
Roy, Is IC503 warming? during RX as well? JZ On Thu, Apr 4, 2024, 10:17?AM John Z via <jdzbrozek=[email protected]> wrote:
|
|
On my "hot" BS-170, the Drain and Gate show a direct internal short.
So the question is, if I just replace that one, am I good, or more likely, I need to find the root cause of why it went bad ... unless it's just that I over tightened the heat sink assembly? Thank you so much John and Ludwig for jumping in to help this novice EE (I'm a retired software engineer actually)! -- 73 de Roy - KI0ER Littleton, Colorado USA |
|
Roy, I strongly advise you try to figure out root cause before you replace the bad transistor. ?At a minimum, disable the power source to the PA by lifting a leg of L502, or removing it completely.? Now you can measure gate drive voltages while not risking another transistor failure. Make certain IC503 has not failed. JZ On Thu, Apr 4, 2024, 11:01?AM Roy - KI0ER <ki0er@...> wrote: On my "hot" BS-170, the Drain and Gate show a direct internal short. |
|
That's encouraging!? While L502 is still removed also measure the gates while TX is active. It will probably read OK but one-step-at-a-time caution is still prudent. JZ On Thu, Apr 4, 2024, 11:45?AM Roy - KI0ER <ki0er@...> wrote: Hi John. |
|
Looks good! On Thu, Apr 4, 2024, 12:09?PM Roy - KI0ER <ki0er@...> wrote: I'm seeing 2.436 V gate to ground on the 2 BS-170's near board edge, and 2.46 V on the "lower" 2. I am running my 12V QMX at 9 V for these tests. |
|
I re-checked continuity of T501, and looks good. Then checked voltages, and all P & Secondary are zero volts with the QMX powered on in RX. But that's with L502 removed, so I'm guessing that's why (without looking at the schematic).
John, Should I reinstall L502 and see what happens? -- 73 de Roy - KI0ER Littleton, Colorado USA |
to navigate to use esc to dismiss