|
Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members. |
|
Thread Tools |
6th Jun 2021, 6:47 pm | #61 |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: RF preamp circuit selection
If there really is just a series 5500R bias resistor at each base and this resistor does feed from a noisy 1.4V reference inside the chip then this 5500R resistance can probably be used to predict the RC noise filtering effect of adding (say) 10nF or 1uF to pin 2 in single ended mode.
Leaving it completely unfiltered in balanced mode clearly has noise problems but maybe it doesn't need a huge tant cap at the centre tap of the transformer to give good results. I'm out of time looking at this stuff for now as I'm getting my bench back ready for home working on Monday but I could repeat the noise plots using various cap values for the centre tap in balanced mode if there is any interest? I think I used a whopping 47uF for the earlier analyser plots of the noise. Balanced mode does offer the advantage of increased conversion gain and this is probably desirable in a direct conversion receiver as it probably helps to reduce the system noise figure. Any benefit depends on the noise performance of the first AF stage.
__________________
Regards, Jeremy G0HZU |
6th Jun 2021, 6:57 pm | #62 |
Pentode
Join Date: Apr 2018
Location: Canterbury, Kent, UK.
Posts: 189
|
Re: RF preamp circuit selection
I for one would be interested in an estimate of how little centre tap capacitance is "just enough", both for my 10KHz design and also for more general direct conversion receiver CW/SSB applications. Certainly 1uF in my receiver made a significant improvement.
Alan |
6th Jun 2021, 10:56 pm | #63 |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: RF preamp circuit selection
I managed to do this tonight so see below for a comparison plot of 10nF (green trace) vs 47uF (yellow trace) at the centre tap of the transformer.
The other plot shows a comparison of 100nF (green trace) vs 47uF (yellow trace) at the centre tap of the transformer. 10nF clearly isn't enough but 100nF seems quite good unless trying for a cw receiver I guess. Note that I had to use the PP3 battery with an RC cleanup for the main PSU connection to pin 8 because my linear PSU was adding some noise down in the comms audio range. I also had to use the RTSA in baseband mode to keep a low noise floor down close to DC. The RTSA feeds the RF input straight the the digital IF like this so there is no RF converter involved. It's all sampled direct with a big ADC and then it runs an FFT. Note that these plots are scaled at 2dB/div rather than the 1dB/div of the previous plots. I suspect most people would choose to use 1uF here if trying for a cw/SSB receiver but in your case the 100nF response looks fine by 5kHz. I used a 100nF X7R class 2 ceramic cap for this test. The 10nF cap was the same type. The 47uF was an SMD tant cap.
__________________
Regards, Jeremy G0HZU Last edited by G0HZU_JMR; 6th Jun 2021 at 11:08 pm. |
6th Jun 2021, 11:19 pm | #64 |
Pentode
Join Date: Apr 2018
Location: Canterbury, Kent, UK.
Posts: 189
|
Re: RF preamp circuit selection
Oh good. So I don't have to worry that I somewhat arbitrarily chose 1uF and 10nF in parallel.
Ta, Alan |