3
$\begingroup$

I have read about double-balanced mixers and as far as I got the theory right they should suppress carrier frequency.

I am trying to build my own transceiver project and I cannot suppress the carrier. See the picture: On air

I guess this is coming from a first mixer and partially gets suppressed by the crystal filter.

My first mixer looks like this: Mixer

Link to a full schematic.

Audio comes from audio-compressor but I tried to inject it directly which is not helpful at all.

Unfortunately I cannot inject a balanced audio so I tried to be creative and injected only on one side of it. You might notice C108 has small capacity of 2.2nF while C109 is 1uF. The carries gets supplied from VFO1 (I flip VFO1 & VFO2 frequencies between RX & TX). In TX mode I supply a constant offset for intermediate frequency amplifier to have a static level of gain, it helps a little bit with carrier frequency suppression but not a lot.

Theoretically dis-balance might occur in T2 but I am not sure how to compensate this.

Also, I noticed that level of the carrier depends on the level of input signal and it gets lower but I think this is just a work of automatic gain control inside IF amplifier.

Have a look at it:

carrier

I am lost right and I will appreciate any advice or links/references to read.

$\endgroup$
1
  • 2
    $\begingroup$ I can't help with your question but would like say I really like the effort you put in your schematic/notes... $\endgroup$ Dec 25, 2021 at 21:04

1 Answer 1

1
$\begingroup$

So far I came up with trimmer potentiometer of 470 ohm. It got the job done and I had a good carrier suppression that is not visible on the SDR receiver.

Solution

I guess there is nothing much I can do about. I guess I have to blame MABAES0060 for imperfections. I had to move the tap quite far.

enter image description here

$\endgroup$
5
  • 2
    $\begingroup$ LO leakage will be reduced if the LO is driven differentially, says data sheet... Curious that your trimpot is so far from midpoint. I wonder how 40mA divides between Gilbert cell's two main branches (DC current out pins 4,5)? Does the balance point change if you lift a pad of L29? Does balance point change with LO drive level? You're using this chip near the bottom-end of its intended frequency range. $\endgroup$
    – glen_geek
    Dec 26, 2021 at 17:55
  • 1
    $\begingroup$ +1. If no one else posts an answer after a few days, then please mark your own answer as accepted. This will prevent the bot from constantly bringing this to the top. We look forward to more from you! :-) $\endgroup$ Dec 26, 2021 at 23:08
  • $\begingroup$ @glen_geek I think if trimpot is about 1/4 so it will translate into something like 150 Ohm. I haven't taken any measurements but I guess that will be just a few dB difference. I am not planning to measure current on the mixer output because the transformer on the other side and it is difficult to desolder it while preserving parts on the bottom from falling. I might consider doing it later but not right now. I agree on the using the part on the lower end of the range. I guess matching there is non trivial. Sadly there are not that many mixers with such dynamics for these frequencies. $\endgroup$
    – k1zmt
    Dec 27, 2021 at 21:23
  • 1
    $\begingroup$ Minicircuits JMS-1H+ has pretty good balance at low frequencies (60 dB L0-IF isolation). Your Si5351 should be able to drive its L,O. requirement if you use 2 Si5351 outputs: one inverted from the other, but same frequency. Add some attenuation to get +17 dBm. It can be used as a modulator, since I.F. goes right down to DC. $\endgroup$
    – glen_geek
    Dec 27, 2021 at 23:36
  • $\begingroup$ @glen_geek oh, nice. I remember, I saw earlier ADE-1 mixer but I thought using LTC5562 would be nice because it is smaller and doesn't require high amplitude to drive. $\endgroup$
    – k1zmt
    Dec 28, 2021 at 3:18

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .