开云体育

ctrl + shift + ? for shortcuts
© 2025 开云体育
  1. BITX20
  2. Topics

Spurious output.

 

I am using a BITX40 with Raduino. The only changes have been around the mic amp and getting rid of the tuning noises. This is related to the rx birdie just below 7.2MHz, which annoys me but I can live with that. However if I transmit on 7.15MHz I can measure an upper sideband signal at 7.4MHz which is only about 33dB down. This is the 4th harmonic of the vfo mixing with the 12MHz IF.

Can anyone with modern test gear confirm this? I would love to have a Rigol but have to make do with a receiver and switched attentuators. I am measuring the level across the R136 pot with no power applied to the IRF510. Twin-T af oscillator into the mic input.

Does the square wave from the Raduino make this worse than using a clean sine wave from the analogue vfo? Is there a conclusion as to whether a diode mixer should be driven with a sine wave or square wave?

73 Brian.
 


The image from the 4'th VFO harmonic, transmit or receive, can be cured by running the VFO at 19mhz, not 5 mhz. ?(We've discussed receive images from this fourth harmonic previously, hadn't occurred to me that it would also affect transmitting.) ?The analog vfo is not stable enough for this, but the Raduino is plenty steady. ?Only issue is you would then be on the upper sideband. ?Can get back to lower sideband by moving the BFO down in frequency about 3khz, perhaps by just adding capacitance at C103, though it may take more extreme measures to 3khz with the crystal BFO oscillator. ?A much better solution is to use the first channel of the Si5351 to drive the BFO, and hack the Raduino sketch to put it at about 11.970mhz. ?(Use the BFO scheme from the new uBitx schematics.) ? With the Si5351 driving the BFO, we can now easily switch between USB and LSB, and put the carrier right in the middle of the filter when transmitting CW (by unbalancing the modulator, much like the uBitx unbalances the first mixer), this also allows much easier calibration procedures that scale properly to other bands.

> Is there a conclusion

There never has been a question. ?Diode ring mixers are comutating mixers, should be operating as switches. ?Square wave is best. ?If you were to feed them with a pure sine wave, the diodes would be passing through their transition region very briefly, and they would either be conducting or not conducting anyway. ?A square wave approaches the ideal of switching the diodes instantaneously and simultaneously, a sine wave only adds to the crud coming out off the mixer. ?If still curious, see my rant here: ?/g/BITX20/message/22521

Jerry, KE7ER


On Sat, Mar 11, 2017 at 09:51 am, <vk4bap@...> wrote:

I am using a BITX40 with Raduino. The only changes have been around the mic amp and getting rid of the tuning noises. This is related to the rx birdie just below 7.2MHz, which annoys me but I can live with that. However if I transmit on 7.15MHz I can measure an upper sideband signal at 7.4MHz which is only about 33dB down. This is the 4th harmonic of the vfo mixing with the 12MHz IF.

Can anyone with modern test gear confirm this? I would love to have a Rigol but have to make do with a receiver and switched attentuators. I am measuring the level across the R136 pot with no power applied to the IRF510. Twin-T af oscillator into the mic input.

Does the square wave from the Raduino make this worse than using a clean sine wave from the analogue vfo? Is there a conclusion as to whether a diode mixer should be driven with a sine wave or square wave?

?
 

Well, never been a question for me anyways. ?Other's have asked if a sine wave VFO might be better. ?I'm convinced a square wave into the diodes is ideal. ?You could make a case for a sine wave being better if it's to be transported around a radio by a big random jumble of 2 foot long wires. ? Better still is to just use coax.


On Sat, Mar 11, 2017 at 10:20 am, Jerry Gaffke wrote:

There never has been a question.

?
 

Your RF mixer might be defective, check that the two transformer windings are not shorting.

-- f

On Sun, Mar 12, 2017 at 12:45 AM, Jerry Gaffke via Groups.Io <jgaffke@...> wrote:

Well, never been a question for me anyways.? Other's have asked if a sine wave VFO might be better.? I'm convinced a square wave into the diodes is ideal.? You could make a case for a sine wave being better if it's to be transported around a radio by a big random jumble of 2 foot long wires. ? Better still is to just use coax.


On Sat, Mar 11, 2017 at 10:20 am, Jerry Gaffke wrote:

There never has been a question.

?


 

Visually the transformers look OK. That's why I ask whether anybody else has measured this before going any further.
 

I did not notice anything on 7.4 but I'll check again on Tuesday when I am back in town.

Try getting a neighborhood ham to listen in an see what he reports on 7.4.

At 11/03/2017, you wrote:

I am using a BITX40 with Raduino. The only changes have been around the mic amp and getting rid of the tuning noises. This is related to the rx birdie just below 7.2MHz, which annoys me but I can live with that. However if I transmit on 7.15MHz I can measure an upper sideband signal at 7.4MHz which is only about 33dB down. This is the 4th harmonic of the vfo mixing with the 12MHz IF.

Can anyone with modern test gear confirm this? I would love to have a Rigol but have to make do with a receiver and switched attentuators. I am measuring the level across the R136 pot with no power applied to the IRF510. Twin-T af oscillator into the mic input.
 

Brian,

See my latest mail on the 4th harmonic of the VFO mixing with BFO and producing products in the 7-8Mhz band.

Change the C72 to 10pf - your power out will drop but you will have a cleaner signal. Select a value that you

can compromise the PO and spurious above 7.2!

Cheers

Raj, vu2zap
 

Raj,

Thanks for that. I have already started on the analogue vfo. I have lots of 1960's components! Should be less of a compromise.

The Raduino might be used later at 19MHz to produce usb for data/wspr. Could one make a dual 40/30 meter usb transceiver by adding 22MHz to the Raduino? Would a 10MHz bandpass filter reject the 12 MHz IF sufficiently?

73 Brian.
 

Brian,

I don't think you will need to reject the IF.

I analysed the TX this morning I found oodles of spurs. Nearest ones to 7.1Mhz (TX) being ~7.6 and ~6.6 about 30db down.
The mod I suggested to improve the image problem makes the spurs slightly worse.

As Jerry mentioned in another mail a square wave has only odd harmonics, a perfect square VFO will clean up the 19MHz
harmonic of VFO and probably make others worse! Must try this with a sn74lvc1g17 Schmitt device.

I will feed a square wave direct from function generator and see what results I get.

73 Raj

At 16-03-2017, you wrote:

Raj,

Thanks for that. I have already started on the analogue vfo. I have lots of 1960's components! Should be less of a compromise.

The Raduino might be used later at 19MHz to produce usb for data/wspr. Could one make a dual 40/30 meter usb transceiver by adding 22MHz to the Raduino? Would a 10MHz bandpass filter reject the 12 MHz IF sufficiently?

73 Brian.

1 - 9 of 9