开云体育

ctrl + shift + ? for shortcuts
© 2025 开云体育
  1. Softrock40
  2. Topics

Two switch QSD test

kb9yig
 

Good Evening All,

I am testing a two switch QSD circuit for use in the v5.0 SoftRock.
The switches in the QSD are each on for 50% of the clock cycle and
staggered by 90 degrees. All other circuits in this test version of
the SoftRock are the same as the v4.0 SoftRock.

In the present v4.0 SoftRock there are four switches with each
switch on for 25% of the clock cycle. Two switches, staggered by 180
degrees, connect alternate ends of transformer T1,(+/-RF), to the I-
signal integrating cap. In the same maner but 90 degrees offset from
the I-clocking, two switches provide the input to the Q-signal
integrating cap. Typical performance with the v4.0 is a noise
baseline of about -139dBm when PowerSDR is calibrated for -73dBm at
50uV RMS.

With this two switch QSD SoftRock I am seeing a 3dB decrease in
signal strength compared to a v4.0 when checking with a -73dBm
source. The noise baseline comes up to -135dBm when PowerSDR is
calibrated with the two switch QSD to -73dBm. Some loss of
sensitivity but still a MDS on the order of -131dBm.

Still working on the schematic for the v5.0 SoftRock, but hope to
have it ready to post soon.

73,
Tony KB9YIG
 

kb9yig wrote:

Good Evening All,
I am testing a two switch QSD circuit for use in the v5.0 SoftRock. The switches in the QSD are each on for 50% of the clock cycle and
staggered by 90 degrees. All other circuits in this test version of
the SoftRock are the same as the v4.0 SoftRock. [snip]

Tony,
which analog switches are you using ? Just curious...
Have you given a look at the Analog Devices ADG704 ?
They have a very low RdsOn, I have used them with success
(were it not for the pesky 0.5mm spacing between pins...)

73 Alberto I2PHD
Tony Parks
 

Hi Alberto,

The switches used are two of the switches in a FST3126. I am needing to stay with this part for the v5.0 SoftRock to use a large supply of the parts I had to buy as part of a minimum order.

Would a lower on resistance improve the performance of a QSD circuit?

73,
Tony KB9YIG

----- Original Message -----
From: "Alberto I2PHD" <i2phd@...>
To: <softrock40@...>
Sent: Monday, October 17, 2005 9:29 AM
Subject: Re: [softrock40] Two switch QSD test


kb9yig wrote:

Good Evening All,

I am testing a two switch QSD circuit for use in the v5.0 SoftRock.
The switches in the QSD are each on for 50% of the clock cycle and
staggered by 90 degrees. All other circuits in this test version of
the SoftRock are the same as the v4.0 SoftRock.
[snip]

Tony,
which analog switches are you using ? Just curious...
Have you given a look at the Analog Devices ADG704 ?
They have a very low RdsOn, I have used them with success
(were it not for the pesky 0.5mm spacing between pins...)

73 Alberto I2PHD







Yahoo! Groups Links
 

Tony Parks wrote:
Hi Alberto,
The switches used are two of the switches in a FST3126. I am needing to
stay with this part for the v5.0 SoftRock to use a large supply of the parts
I had to buy as part of a minimum order.
Would a lower on resistance improve the performance of a QSD circuit?
Hi Tony,

the RdsOn is practically in series with the 50 ohm of the antenna or
the output impedance of a preceding preamplifier. So for the maximum of
efficiency you want to keep the switch resistance as low as possible.
The FST3125 is rated for an RdsOn of 4 ohm, while the ADG704 (or the ADG711 if you prefer 4 separate switches instead of a mux) is rated
for 2.5 ohm. Of course if you design your preamplifier with a somewhat higher output impedance, then the issue is much less relevant...

73 Alberto I2PHD
 

Hello Alberto and tony

The other parts of the switch will come handy for a TX.

the SR40 has a 10 ohm resistor in series with the source impedance so
the influence of 2 ohms in minute. (you could always connect the other
half in parallel...)

p.s Alberto I want to thank you for the very nice and
excellent program for SDR and for the other weak signal programs in
your web site.


73 de Joe
Cecil Lindsey
 

Tony & Alberto

What about paralleling the unused switches in the '3126, since you are only using to in the new configuration?

Cecil - WA6RDY

----- Original Message -----
From: "Alberto I2PHD" <i2phd@...>
To: <softrock40@...>
Sent: Monday, October 17, 2005 10:11 AM
Subject: Re: [softrock40] Two switch QSD test


Tony Parks wrote:
Hi Alberto,

The switches used are two of the switches in a FST3126. I am needing to
stay with this part for the v5.0 SoftRock to use a large supply of the
parts
I had to buy as part of a minimum order.

Would a lower on resistance improve the performance of a QSD circuit?
Hi Tony,

the RdsOn is practically in series with the 50 ohm of the antenna or
the output impedance of a preceding preamplifier. So for the maximum of
efficiency you want to keep the switch resistance as low as possible.
The FST3125 is rated for an RdsOn of 4 ohm, while the ADG704 (or the
ADG711 if you prefer 4 separate switches instead of a mux) is rated
for 2.5 ohm. Of course if you design your preamplifier with a somewhat
higher output impedance, then the issue is much less relevant...

73 Alberto I2PHD







Yahoo! Groups Links

1 - 6 of 6