Tinypico Radio Part 10a - FST3253 Transmit Mixer

This is an additional video to Part 10 - the double balanced mixers. In this video I use the FST3253 Mux/Demux as a transmit mixer.
Schematics here:
github.com/tha...
The mixer is first testing via input from the Signal Generator and then using the Microphone Amplifier built in Part 11 of the series

[@HB9GZE]

Hello Ian, why do you think do we need for transmit an audio signal in all 4 quadrants (0, 90, 180 and 270)? If a mixer works as down-mixer, it should also work in up-mixing the same way I thought so I was expecting to use 1:1 the RX mixer configuration. Did you try the RX mixer with the 4 quadrants input?

[@na5y]

Hi Stefan. My understanding is the four quadrant approach is used to reduce loss in the mixer. The 0 and 180 effectively add together (as do the 90 and 270) . That's at least my laymans understanding anyway. The receive mixers do the same thing - check this out (page 3) www.norcalqrp.org/files/Tayloe_mixer_x3a.pdf Thank you for the comment and question. I hope the above helps

[@HB9GZE]

@@na5y I will try it out and let you know...

[@788home]

I am a bit late watching your video - and I have to assume you have diagnosed the source of the additional spurious responses. If not: You are using a switch as a mixer - and the pulse shape of the driving waveform has a spectrum of lines rather than a single line (as in a sine wave). As someone else observed, you may also be overdriving the op amps, causing them to go into limiting on part of the cycle - which also generates set of spectral lines. I believe you may be able to do a little DSP . Perhaps generate the SSB signal at a lower IF, where you can use a Digital filter after the modulator and then translate it up. More complexity.

[@na5y]

HI John - thank you for the comment. The biggest source turned out to be the opamps I was using (LM358s) I switched to LM4562s and a good portion of the spurious output disappeared.

[@4youian]

Hi Ian, I just discovered your channel, very interesting, thanks! I wonder if the harmonics are being generated by the op-amps? I'm not sure if you said - sorry if I missed it - did you probe the outputs of the op-amps to see if the harmonics were being generated there and not in the mixer? Just wondering if you discovered the cause, thanks, VK7IAN.

[@na5y]

You are exactly right Ian! I did find that out that in a subsequent video. Those cheap LM358 opamps were introducing a pile of harmonics. I replaced them with LM4562s and all those harmonics pretty much went away. Thank you for the comment!

[@4youian]

@@na5y Awesome. Good to know! I'll keep watching.

[@cliffwiener4681]

I would like my team to implement your designs in an SOC in 40nm and below. We have the EDA tools to do it (Analog/Digital Rails. How do I contact you?

[@cliffwiener4681]

We will embed an open source 32 bit risc-V processor, along with our PLL and interlacing pipeline ADCs

[@na5y]

Cliff, to be clear they aren't my designs and I have no desire to commercialize them. I have just picked up bits of knowledge here and there through kits I have built and articles. Having said that happy to chat on my email address which I have updated in the channel "About"