¿ªÔÆÌåÓý

Re: Spectrum purity of uBITX and BIT40


 

A two tone test will check for IMD (intermodulation distortion, google it!).
It does not check for spurs or harmonics, which are the primary issues with the stock uBitx.

IMD can make your signal sound bad, and cause interference to those operating nearby on frequency.
If you have bad IMD, you will probably be told about it.
The stock uBitx with the mike gain set to a reasonable level will have enough IMD to be
slightly alarming when looking for it with a spectrum analyzer.
But the IMD products are way down (50dB or so) once you get a couple khz away.?
And thus it's a legal transmission anywhere AM phone is legal (at least with respect to IMD).
When transmitting at 5 or 10 watts, the IMD is probably not perceptible to someone using an adjacent channel.

If you have bad spurs and/or harmonics, you will be interfering with operations
on totally different frequencies.
For example, when operating at 28mhz, there's a? spur at 45mhz-28mhz=17mhz.

This particular spur is due to the 2*45mhz second harmonic of the first IF?
mixing with the high side clk2 VFO.? The VFO is always at 45mhz+DialFreq.
In this case, the spur is at? 2*45mhz - (45mhz+28mhz) = 45mhz - 28mhz = 17mhz.?

We also have a spur at 12mhz, due to crosstalk within the si5351? from the BFO at clk0
into the VFO at clk2.? This 12mhz energy on clk2 goes through the mixer at T2 unmolested,
and continues on through the transmit power amp and the transmit low pass filters.

Harmonics are simple integer multiples of the operating frequency.
For example, when operating at 7mhz, we get harmonics at 2*7=14mhz and 3*7=21mhz.
The transmit low pass filters at L11 through L22 should take these harmonics out,
but the stock uBitx relays (and to some extent, the board placement and routing)?
allow the harmonics to sneak past the low pass filter.

Also, for various reasons, we have some carrier leakage, though not too bad.
This results in a not quite suppressed carrier riding along with our SSB signal,
sort of like an AM phone signal.? Not really much of an issue at 5 or 10 watts,
but it does make our signal a bit wider than it should be.

Spurs, harmonics, IMD, and residual carrier are why you should think twice
before?adding a linear amp to the uBitx.
Like any rig without ALC (automatic level control), increasing the mike gain
too much (or shouting into the mike) will result in IMD.

The Bitx40 has a much easier job with respect to spurs and harmonics.??
There is a 7mhz bandpass filter before the transmit power amp, which knocks out all those spurs.
And with only one transmit low pass filter there is?no easy path for harmonics to leak out.
With too much mike gain it will have IMD, and there is some carrier leakage,
but these are relatively minor issues.

Jerry, KE7ER


On Sun, Dec 2, 2018 at 09:27 PM, Wayne Leake wrote:
?I notice that nobody mentions the best way to check an SSB outgoing signal.
?You need a good two tone audio signal generator. Along with a wide band scope, check that you see a good sine wave pattern.
??

Join [email protected] to automatically receive all group messages.