开云体育

Greetings all,

I have hit a point in my QDX build where I am able to receive fine on 40m and 20m (well, before I shorted out IC5 as described below - I haven’t tried since), but still cannot transmit at all, save you count a few milliwatts. To summarize what I have done...

?- The QDX is built as per the manual, only with the LPF normally used for 80m outfitted for 15m and a tap after the 16th turn on L12 that goes over and connects to the pad normally used for 80m.

?- I traced the source of the high current draw on receive (340 mA @ 12V) to IC5. Replacing IC5 dropped the receive current draw to ~150 mA.

?- When I replaced IC5, as I suspected that at least one BS-170 might have been damaged, I replaced all 4 BS-170s. After I applied power to the QDX, I found Vcc on the correct pins and the BS-170s’ gates at 0V in receive. Using WSJT-X to generate a test tone*, I probed the gates and found them to all be at ~2.46V during transmit, so all was well. At least that far in the signal chain, for I was still getting no RF output, even though the front LED was blinking 3 times, indicating that it was receiving sufficient audio.

?- I then managed to kill IC5 when I was probing one of the BS-170s from the top side of the board (read: easy to short together pins) and my probe bridged the drain and gate on one of the BS-170s. A small pop was heard with the telltale ‘dead IC smoke’ being then smelled. Probing the gates found them to all be over 6V (6.6V?).

?- I forgot to check the gates AFTER I removed IC5 but BEFORE I replaced it, so I can’t say what the gates’ voltage would have been with the BS-170s still in, but without IC5. What I can say is that after I replaced IC5, I found that the gate voltage was still greater than 6V.

?- Removing the BS-170s, I metered the gate voltages at 4.99V - and this is all in receive. I also saw the buck converter’s display showing that with no BS-170s and the new IC5, the current draw was 250 mA.

At this point, I am at a loss as to what to check and/or do. I can understand the raised current draw being because the PA gates are at 5V, but as I’ve never seen them go above 2.5V with a good IC5, I don’t understand why it is that high. I’ve checked the voltage regulators for +5V and +3.3V and they are metering fine. All of the switching diodes are switching correctly, though I do not think that I have verified that since The Short that killed IC5. The only check I can think of that I have never done is to measure the switching diodes during transmit, so as to verify that the signal is getting through T1 and at least entering the appropriate LPF. Also, it should be noted that when I power up the QDX, the front LED blinks rapidly several times and then goes to always on, so it would seem that the microprocessor part of the unit is still working.

Any thoughts and suggestions as to what to do next would be greatly appreciated. I’m really trying to not have to write off this board and start afresh :-)

Regards,

Charles Johnson
KF4AYT

* I realize that there is a test utility in the QDX that will generate a test tone, but since it requires that the appropriate key be pushed the entire time and I rarely have a second set of hands that could push and hold it, WSJT-X is my only option.

Hello Charles,

The only way to get 5 volts on the gates is if there is a part failure, most likely IC5, though it could also be from a failed BS170.? The current when in receive mode should be under 160mA.? This does depend on the board revision.? Rev1 at 100mA is much lower than rev3.? I have not measured rev2 boards, so I do not know what the receiver current should be.

As I read your post, it looks like you were almost there when the shorted gate occurred.? That happened to me.? When it did, it took out the two BS170s on the side that was shorted AND IC5.? At this point, I would remove the 4 BS170 transistors and verify that the receiver current drops below 160 MA and IC5 is working correctly (0 volts when receiving and 2.5 volts when transmitting).? If that test fails, you need to replace IC5.? The next step is to verify that the receiver is working before installing the BS170 transistors.

Assuming all is good, the last step is to install the BS170 transistors.

Can you verify which board revision you have?? Also, the supply voltage when testing.? It is a good practice to start with 7.5 volts supplied to the QDX until you verify that the transmitter section is working.? This reduces the chance of overloading a device.? 7.5 volts is the lowest, so you can ensure enough headroom for the 5-volt regulator.

73
Evan
AC9TU

Charles Johnson <cjohnson@...> wrote:

- I *forgot* to check the gates AFTER I removed IC5 but BEFORE I
replaced it, so I can't say what the gates' voltage would have been with
the BS-170s still in, but without IC5. What I can say is that after I
replaced IC5, I found that the gate voltage was still greater than 6V.

- Removing the BS-170s, I metered the gate voltages at 4.99V - and this
is all in receive. I also saw the buck converter's display showing that
with no BS-170s and the new IC5, the current draw was 250 mA.

Do I understand correctly that you removed BS170s, installed new IC5 and
still have 5 volts on the gates of BS170s, even when receiving?

At this point, I am at a loss as to what to check and/or do.

If IC5 produces incorrect voltage(s) and has been replaced, I'd check what
appears on its inputs. If TX is high during receive (and it has to be for
IC5 to produce 5 volts), you might have fried the STM32... or at least a
PCB trace.

I'd also check if IC5 has proper ground connection.

The only check I can think of that I have never done is to measure the
switching diodes during transmit , so as to verify that the signal is
getting through T1 and at least entering the appropriate LPF.

I'd first try to fix the problem with incorrect voltage on gates. Then, if
finals produce the correct signal, I'd check futher.

Also, it should be noted that when I power up the QDX, the front LED
blinks rapidly several times and then goes to always on, so it would
seem that the microprocessor part of the unit is still working.

This blinking is normal. The MCU might be working, but output ports (I'm
thinking of the TX port) might have been damaged...

On Mon, Dec 5, 2022 at 07:35 AM, Adam wrote:

Charles Johnson <cjohnson@...> wrote:

- I *forgot* to check the gates AFTER I removed IC5 but BEFORE I
replaced it, so I can't say what the gates' voltage would have been with
the BS-170s still in, but without IC5. What I can say is that after I
replaced IC5, I found that the gate voltage was still greater than 6V.

- Removing the BS-170s, I metered the gate voltages at 4.99V - and this
is all in receive. I also saw the buck converter's display showing that
with no BS-170s and the new IC5, the current draw was 250 mA.

Do I understand correctly that you removed BS170s, installed new IC5 and
still have 5 volts on the gates of BS170s, even when receiving?

To the best of my memory, the IC5/BS-170 'dance' was as follows (note: The A, B, C denotes the unit # - ie IC5 (A) was the IC5 that came preinstalled):

?- IC5 (A) + 4x BS-170s (A) - No RF output, receive current draw was 340 mA (for a 12V build).

?- Removed IC5 (A) and the receive current draw dropped to 150 mA.

?- Replaced IC5 (A) and, just for good measure, BS-170s (A).

?- IC5 (B) and BS-170s (B) worked properly, insofar as the gate voltages rising to 2.45V on transmit. Still no RF output.

?- Accidentally shorted drain to gate for one of the BS-170s (B) and smelled That Smell. Checked IC5 (B) and I saw the small dimple on it. Checked the gate voltages and found them to be >6V.

?- Removed IC5 (B) and BS-170s (B). Installed IC5 (C).

?- Measured gate voltages at 4.99V (no BS-170s installed)

?- Removed IC5 (C). Measured no voltage on the gates (as expected).

Currently, no BS-170s or IC5 are installed. It's also still drawing 250 mA on receive.

At this point, I am at a loss as to what to check and/or do.

If IC5 produces incorrect voltage(s) and has been replaced, I'd check what
appears on its inputs. If TX is high during receive (and it has to be for
IC5 to produce 5 volts), you might have fried the STM32... or at least a
PCB trace.

I'd also check if IC5 has proper ground connection.

I always checked the Vcc and GND voltages when I installed the new IC5 chips, but as I understand it, the gates should be seeing 2.5V, not 5V. I don't see anywhere in the schematic that shows where the voltage is halved, as it's only logical (pun intended) that the 74ACT08 should output 5V, but everything I've read says that the gate voltage should be 2.5V on transmit.

As for checking the input/output values on IC5 during transmit, at the moment, I can't, as no IC5 is currently installed.

Also, it should be noted that when I power up the QDX, the front LED
blinks rapidly several times and then goes to always on, so it would
seem that the microprocessor part of the unit is still working.

This blinking is normal. The MCU might be working, but output ports (I'm
thinking of the TX port) might have been damaged...

In going over the board, when I checked Q6, which is supposed to have a drain value of Vcc (not Vdd), whether I had the supply voltage at 10V or 12V, the voltage on the drain was 3.29V. Due to having the wrong computer on hand to connect to the QDX, I couldn't connect over the serial port or use WSJT-X to trigger transmit and see what happened. I don't know if 3.29V would be enough for the T/R switch (if not, then that would explain the utter lack of RF output), but it not being Vcc points to a problem elsewhere.

As for the MCU, thus far, I've been able to connect to it via the serial port with no problems, so while that doesn't rule out an output pin (or more) being damaged and/or damaging something downstream, I think I can at least say that I haven't fried the MCU :-)

-------

On a related note, with IC5 and the BS-170s out, even though the audio sweeps should not be affected at all, the last time I ran the sweeps (I was moving through the bands to verify that the switching diodes were switching correctly), the results I saw were horrible. The USB and LSB traces were a mess and all intermingled. Can anyone explain that?

(Puzzled) Regards,

Charles Johnson
KF4AYT

Charles,
You asked about why 2.4V on IC5's outputs. It's because in transmit it's outputting a 5V square wave at the the operating frequency. On a DVM that high frequency square wave will read ~2.5V.

Are you doing 'sweeps' testing with a dummy load attached? If so that could explain your zany results.

At this point, are you able to receive signals over the air and on all bands?

As far as current draw during receive (at idle) if you are using one of the buck converters to read this it might be better to do this the traditional way using a DVM inline with the positive lead of the power supply.?

Thanks!

--Al
WD4AH