Re: How to connect ubitx microphone jack to soundcard output; now unsuppressed carrier in SSB or digital modes.

Jerry Gaffke
 

For those trying to follow the conversation, it started in this other thread:
    https://groups.io/g/BITX20/topic/30204401
Karl's adio had been working well as an SSB transmitter, but after trying to adapt it to a PC soundcard
for use on digital modes, it now has a strong residual carrier.
My understanding is that this undesired carrier is roughly as strong as the desired sideband.
Karl tried replacing the BAT54S diode pair in the modulator, that apparently did not help.

This fault may have nothing to do with the attempts at a soundcard interface, 
could just be from jostling and banging around the board.

Curious that the carrier is not getting at least somewhat suppressed by the 12mhz crystal filter.
I'd try shorting across R114 to disable the output from the modulator, then see if the carrier is still
there when the mike is keyed.  Could be that CLK0 is somehow getting coupled in after the filter.

Karl reports "At the C50/C63 node, I am getting 1.4 mV without the microphone being keyed and 1.2 to 1.5 mV when the microphone is keyed".
That node is going into the modulator from the mike amp.
Not clear if that is a DC level, or some sort of audio or RF AC.
If it's AC, that might just be local noise (perhaps from CLK0) getting picked up by a scope ground lead.
If it's DC, I would not expect that, but then I would not expect a mV or so to cause a very strong carrier.
Might be worth removing C50 and C63 from the board using two soldering irons, clean the board
in that area with alcohol in case some flux is providing a conductive path, then power up and see if
there is still a significant carrier when the mike is keyed. 
Check those caps with an ohmmeter to verify they are not leaking.

Perhaps windings on T7 have shorted out?
Might be worth rewinding T7 with new magnet wire, see if that helps.

Could replace that modulator with a commercially available diode ring mixer such as an SBL1 or ADE1.

One of the cheap AD8307 RF sniffers on ebay would be useful for looking at these low level RF signals. 
Solder coax into the uBitx keeping signal and ground as short as possible, run that coax into the sniffer.
    https://www.ebay.com/itm/AD8307-1MHz-to-500MHz-Wideband-RF-Detector-RF-Power-Meter-Strength-Meter-/282698011091

Hacking at boards to investigate problems like this must be done carefully.
Would be very easy to create a second reason for this strong residual carrier.

Jerry, KE7ER


On Sun, Apr 7, 2019 at 02:47 PM, Karl Heimbach wrote:
Jerry,

I have the v3 board as do you.  I bought 10 diodes from Mouser and installed a replacement (twice) this afternoon.  No improvement and still have the unsuprressed carrier.  I also heated up the pins on T7 to check for a cold solder joint, looked over the entire board with a magnifying glass to see if I missed something.  Loaded version CEC version 1.1 software, recalibrated, etc., without any success.

At the C50/C63 node, I am getting 1.4 mV without the microphone being keyed and 1.2 to 1.5 mV when the microphone is keyed.  I looked very closely at the microphone jack for bent pins and can't see any.

As before the radio works fine on CW and receives well.

Any other thoughts?

Karl - W5QJ

Join BITX20@groups.io to automatically receive all group messages.