Whoa! Sounds far fetched but reported DC behaviour on D3 and D6 is inverted logicwise to what I would expect and is definitely the odd one out. Now I notice the same on your original post so not related to you now disconnecting cathodes.?
Can you document/verify please the signals on X6 as you change band selection? Am traveling far from home in Ireland so can’t verify what should be but I think active low for each band?
Could you also verify please that Q3, Q4 and Q5 have identical part markings?
On 29 Aug 2023, at 4:18 pm, Evan Hand <elhandjr@...> wrote:
?First, it is considered good form always to post your name and call at the end of a post.? Then, others can direct their posts to the correct person.? Also, when posting in response, direct the post to the person you are responding to.
To the original poster:
The QDX has gone through 5 revisions of the circuit board.? The differences can impact the usefulness of the help from the forum.? Please post the board revision and the model to help us give you the best information.
To answer your question on how the LPF selection works, there is a detailed description in the construction manual on page 73 for the Rev 5 board.? Others will be around the same page with the same heading.? You can get the correct manual from the QRP-Labs QDX web page:
Basing the selection diodes forward in a simplified description connects the two leads like a switch.? Reverse bias opens the switch.? The caveat is that the RF peak voltage must not exceed the reverse bias voltage, or the diode will conduct.? ? For receive, that is in the millivolts, and the 5-volt supply is enough.? When transmitting, the voltage can get over 45 volts peak to peak or 23 volts peak.? That means the reverse voltage on the diode must be above 24 volts for safety.? There is no 24-volt DC supply in the QDX, so Hans developed a voltage doubler circuit for transmitting.? That doubler is D8 and D9 with capacitor C27.? This happens only when transmitting and does require a good RF output.? If there is no RF, The doubler does not work, and D7 provides the input voltage supplied to the QDX until the doubler kicks in.
The supply to forward bias the LPF diodes is different for different boards.? The original board (Rev 1) used the input voltage through a MOSFET switch and then a resistor to bias the selected diodes forward.? The same circuit was used up to Rev 4 when Hans came up with using the processor to act as the controller for a switching buck regulator to reduce the power dissipation required by the series resistors.? There are other aspects of the new design that benefited the QDX operation.? This is an example of why it is important to list the board revision when asking for help.
To address how to verify that the LPF selection is working, first, confirm in receive mode that only two of the diodes (the two that select the band) have the band (cathode) close to zero volts.? The other four should be at 5 volts.? That would verify that the selection process is working.? At that point, you should be able to receive and decode digital signals with the QDX.
To test in transmit, first verify that there is DC over 15 volts (this would depend on the voltage supplied) on C27.? As in the receive selection test, verify that only two diodes are at zero volts DC and the others are close to the voltage measured on C27.
There is a diagnosing help page on the QRP-Labs web pages:
I? hope that the above helps.? Good luck in finding the issue.