NZ0I
Looking at relative RSSI values: the RSSI value being printed to the LCD display under the above conditions bounces between 2 and 8, corresponding to ~400mV RSSI suggesting slight instability. When I cut the signal path completely (i.e., remove the gain-killing resistors entirely) the LCD displays an RSSI that bounces between 0 and 4, corresponding to ~210 mV RSSI indicating stable conditions. If I supply an unattenuated receive-frequency signal (e.g., 144.25 MHz) from an external Si5351 signal source to the SA605 input with the resistors in place, the RSSI jumps up to 24 to 31, and I can hear a loud steady carrier coming from the audio output.
This is the first time that we've tried using both gain stages of the SA605 together, and it isn't surprising that we are observing instability when all 100dB+ of gain is enabled, especially with a 10.7 MHz IF frequency. But I am a little surprised that such high resistor values are required in order to achieve stability. When I have a better estimate of the minimum values for R111 and R112 I will look more carefully at the RF input matching network, and then attempt to get a better estimate of what sensitivity level is being achieved, probably by comparing the Receiver board performance with a standard receiver when listening to a signal over the air.
I've done a little poking around (literally) in an attempt to see if I can find any clues to the feedback path responsible for the observed instability. So far, no success. Any suggestions or comments are welcome... otherwise I will proceed to follow (loosely) the procedures described in AN1995.
Gerald Boyd
Just fired it up and re-checked.
From last testing and tonight get near same values. Example tonight my zero is .025 and last test was .026 in 1-22 and 1-23 test notes.
Below is the current setting of the gain resistor R212 ( old board schematic).
I did do a test in the past( in test report) with gain killing resistor at zero ohms and the IF was stable and more sensitive. ( see below).
Original test report is on the share drive.
I do notice tonight that if I disconnect the coax from the signal generator and let the RF input float the RSSI does go up from .025 to .6 volts.
Putting a 50 on load on the input its goes back down to .032 volts.
Also did a test where I shut down the Linux computer ( and the clock gen eval board) the RSSI is .033 volts regardless of what the configuration of the RF input is ( signal gen connected at zero output, RF input floating or 50 ohm load). In this configuration LO and BFO are off.
I am using an older schematic / board ( below schematic) with values modified as noted above.
I wonder if there is something going on between the different input circuits and how the IF is connected?
New schematic inputs for LO and BFO.
My LO and BFO inputs are:
My RF input is the older email version that was wired Manhattan stile on the test adaptor board.
Could we have some type of feedback path in the new layout?
Where are L202,203,T201 located and oriented in relationship to other components like the RF input circuits and the other side of the chip where the crystal filter and gain resistor Is located?
Jerry
From: receiver-devel...@googlegroups.com [mailto:receiver-devel...@googlegroups.com] On Behalf Of NZ0I
Sent: Monday, April 24, 2017 7:46 AM
To: Receiver Development Platform <receiver-devel...@googlegroups.com>
Subject: [receiver-development-platform] Receiver Hardware Debugging
I've been going (somewhat haphazardly) through the steps detailed in the application note AN1995: https://drive.google.com/open?id=0B-_v_AmmxnY5akxBSGdwSlpRUVU
One important tool that is working: the SA605's signal strength indicator (RSSI) output. It seems to be putting out a sensible voltage that does change in a reasonable manner when I apply a signal to the SA605 input. The application note AN1995 says that the RSSI level, when no external signal is applied, should be 250 mV rms or less, and that any value greater than 250mV indicates regeneration or instability in the SA605.
With all the gain-killing resistors set to zero in the signal path, both between the first mixer and the IF amp and between the IF amp and the limiter, the RSSI puts out a voltage of about 1.3V (1300 mV). I think that is a pretty clear indication that the SA605 is experiencing instability. Also, the audio output under those conditions is very low, rather like a receiver experiencing "desense" from a strong nearby station. I think that is more evidence that regeneration is occurring.
I can get the RSSI below 250 mV by cutting the signal path, but I'm still trying to establish what minimum level of attenuation is required to achieve stability. The schematic below shows values that result in ~400mV RSSI output with zero signal input, and with U207 shorted across (providing no attenuation), R211 and R212 are the two "gain killing" resistors:
Looking at relative RSSI values: the RSSI value being printed to the LCD display under the above conditions bounces between 2 and 8, corresponding to ~400mV RSSI suggesting slight instability. When I cut the signal path completely (i.e., remove the gain-killing resistors entirely) the LCD displays an RSSI that bounces between 0 and 4, corresponding to ~210 mV RSSI indicating stable conditions. If I supply an unattenuated receive-frequency signal (e.g., 144.25 MHz) from an external Si5351 signal source to the SA605 input with the resistors in place, the RSSI jumps up to 24 to 31, and I can hear a loud steady carrier coming from the audio output.
This is the first time that we've tried using both gain stages of the SA605 together, and it isn't surprising that we are observing instability when all 100dB+ of gain is enabled, especially with a 10.7 MHz IF frequency. But I am a little surprised that such high resistor values are required in order to achieve stability. When I have a better estimate of the minimum values for R111 and R112 I will look more carefully at the RF input matching network, and then attempt to get a better estimate of what sensitivity level is being achieved, probably by comparing the Receiver board performance with a standard receiver when listening to a signal over the air.
I've done a little poking around (literally) in an attempt to see if I can find any clues to the feedback path responsible for the observed instability. So far, no success. Any suggestions or comments are welcome... otherwise I will proceed to follow (loosely) the procedures described in AN1995.
--
You received this message because you are subscribed to the Google Groups "Receiver Development Platform" group.
To unsubscribe from this group and stop receiving emails from it, send an email to receiver-development...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/receiver-development-platform/ae93e6a8-1e34-4692-bd54-f47e9b2cac65%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.