¿ªÔÆÌåÓý

  1. BITX20
  2. Topics

Zbitx musings #2

 

Hi, Just some ramblings...
The test sheet with the Zbitx states -105 dBm is the minimum signal. I am not sure how this is measured. I found it difficult? with the chuff/chuff of screen update, so by ear I listened to a? very weak carrier (from a signal generator) of -105 dBm which appeared on the noise level (with display set in menu mode to minimise chuffing)
?
I extended the display board away from the RF board with Arduino? Male-female test leads so I could do some testing
?
I looked at the datasheet of the WM8731 and it said the S/N ratio was -90dB for a FSD level of 1V RMS. The is about a 32uV noise floor. After doing some tests with the display extended away from the RF board, I was able to determine that the gain of the low IF amplifier was in the vicinity of 72x and that the RF Gain/Loss from RF input to that section of the board was? about -8 dB as best as I could determine.
The low frequency section appears to have a 50ohmish input Z,? so 32uV/72 is? about 0.45uV at 50 ohms or -114 dBm. With 8 dB loss added to that, the noise floor would be about -106 dBm. I figured that I could try some extra gain to bring the input level up. This was done by adding a 27 ohm and a 4.7uF tantalum from Q8 emitter to ground. This increased the gain, but had a marginal increase in sensitivity. (possibly due to earlier gains/losses)
?
I looked at Q4, the 40MHz? RX amplifier, and it was running at about 2.3 mA emitter current, vs the TX side Q3 amplifier which calculates to around 19mA. I decided to change the R27 base resistor from 470 ohm to 1k ohm which increased current to around 8.5 mA.?
?
I found I could now hear carrier around -114 dBm now, which is probably? due to both the 40 MHz and Low IF gain being increased. It affects the S meter readings. I haven't done any measurements of impedances of the 40 MHz amplifier, and would have to try a breadboard setup to determine that.
?
I think the 2.3mA emitter current on Q4 is probably marginal for a RX IF preamp. The increased gain will also affect the maximum signal that the receiver can handle.
?
Tests were done on the RX front end with a signal generator at high-ish level, and a -20 dB RF probe, into a TinySA Ultra. The tests at the low IF were viewed with a oscilloscope in x1 mode. Clipping appeared on Q8 output at about 3v p-p, probably due to the 3.3V zener clamp.
?
Thoughts anyone...
?
Kevin
?
?
 

Yep, I think they are being straightforward.? ?From what I saw, -105 dBm is about 1.26 microvolt.? ?Useful but not stellar sensitivity.? ?I ran into difficulties using a transverter trying to listen to satellites and thought this was related.? ?Usually?the FIRST stage sets the noise figure for the entire receiver so I added a low noise GaAs FET in front for the VHF/UHF signals and found around 15dB gain I was now having results where I could finally hear a difference when I connected/disconnected the antenna.? ?That could be both due to the transverter as well as the sBitx -- but think Ashhar has been straightforward?always pointing that the dynamic range is pretty good and the sensitivity?is "adequate"

The ITU accepted values for background noise:??
Note the units.? ?It takes some math to get to the equivalent dBm sensitivity

dBm = E dBuV/m - 107 - Antenna Factor(in dB)
(I had to work this out for a paper that I wrote on the analysis of an RFI problem.? ?However, QEX wasn't interested in the topic....oh well.)



For an 80 meter full size dipole, AF is approx -19 dB.? ?For a 10 meter full size dipole AF is approx 0.? ?

If I did the math right, you only need -88dBm minimum discernible signal on 80 meters to equal the typical noise floor.
On 10 meters, the math isn't so friendly to the sBitx and you might prefer to have a lower noise preamp in front.? ?I think.? ?This math is somewhat complicated.
Others might do better than I.? ?But on MOST bands I think you're going to find it quite acceptable.??


On Mon, Apr 28, 2025 at 6:11?AM zl1ujg via <zl1ujg=[email protected]> wrote:
Hi, Just some ramblings...
The test sheet with the Zbitx states -105 dBm is the minimum signal. I am not sure how this is measured. I found it difficult? with the chuff/chuff of screen update, so by ear I listened to a? very weak carrier (from a signal generator) of -105 dBm which appeared on the noise level (with display set in menu mode to minimise chuffing)
?
I extended the display board away from the RF board with Arduino? Male-female test leads so I could do some testing
?
I looked at the datasheet of the WM8731 and it said the S/N ratio was -90dB for a FSD level of 1V RMS. The is about a 32uV noise floor. After doing some tests with the display extended away from the RF board, I was able to determine that the gain of the low IF amplifier was in the vicinity of 72x and that the RF Gain/Loss from RF input to that section of the board was? about -8 dB as best as I could determine.
The low frequency section appears to have a 50ohmish input Z,? so 32uV/72 is? about 0.45uV at 50 ohms or -114 dBm. With 8 dB loss added to that, the noise floor would be about -106 dBm. I figured that I could try some extra gain to bring the input level up. This was done by adding a 27 ohm and a 4.7uF tantalum from Q8 emitter to ground. This increased the gain, but had a marginal increase in sensitivity. (possibly due to earlier gains/losses)
?
I looked at Q4, the 40MHz? RX amplifier, and it was running at about 2.3 mA emitter current, vs the TX side Q3 amplifier which calculates to around 19mA. I decided to change the R27 base resistor from 470 ohm to 1k ohm which increased current to around 8.5 mA.?
?
I found I could now hear carrier around -114 dBm now, which is probably? due to both the 40 MHz and Low IF gain being increased. It affects the S meter readings. I haven't done any measurements of impedances of the 40 MHz amplifier, and would have to try a breadboard setup to determine that.
?
I think the 2.3mA emitter current on Q4 is probably marginal for a RX IF preamp. The increased gain will also affect the maximum signal that the receiver can handle.
?
Tests were done on the RX front end with a signal generator at high-ish level, and a -20 dB RF probe, into a TinySA Ultra. The tests at the low IF were viewed with a oscilloscope in x1 mode. Clipping appeared on Q8 output at about 3v p-p, probably due to the 3.3V zener clamp.
?
Thoughts anyone...
?
Kevin
?
?

1 - 2 of 2