jmpac
Junior Member level 2
- Joined
- Nov 4, 2010
- Messages
- 24
- Helped
- 6
- Reputation
- 12
- Reaction score
- 6
- Trophy points
- 1,283
- Location
- Spain
- Activity points
- 1,463
Hi all. I have implemented on PCB the circuit attached (LNA with bypass) but it's not working properly.
If I open the circuit at node VA, and apply an external voltage of 5V, as a control analog signal of the switch, the bypass is done by the two switches (SW1 and SW2), each one is in the state ON-OF, and the LNA2 is disabled too.
However, when using the output of the LM211 to control the switch block, the voltage falls down to 2V and it's not able to control the switches.
In fact, each switch is in the state ON-ON (and not ON-OFF as proposed) so the RF signal is mixing (LNA2 path, plus direct path) and producing intermodulation distortion at the output.
The input impedance of the inverter block is more than 100 Kohm and consumption current is only microamperes (basically the base current of NPN transistors and control current of the switches that is 50 microamperes for each one). The output current of the comparator (LM211) is milliamperes, and both collector output and emitter output configurations were performed for the LM211 (see data sheet).
I also tried voltages followers (with opamps) at nodes VA and VB, but it's not work. Even an optocoupler at node VA. The power detector is the AD8363 from ADI.
What I realized in all cases, and I don't understand, is that voltage at node VB is always 2.291V (>2.288V of reference voltage for the comparator) for a 10 dB power varying RF signal at the input (RFin). In other words: i need to increase the power of the RF input signal up to 10dB so that the bypass is done well by the two switches. But as I said at the beginning, if I open the circuit at node VA, and increase the power of RFin in 1 dB, it's enough for the AD8363 to detect it, thus VB > 2.288V (Vref) and the LM211 outputs a voltage of 5V. But when closing the loop at VA -> output (LM211) falls down to 2V and both switches are in ON-ON state, and VB is only 2.291V for a 10dB power varying RF input...
What's happening? Is there any error in the design?
I would be very grateful if someone could help me.
Many thanks.
Jose
If I open the circuit at node VA, and apply an external voltage of 5V, as a control analog signal of the switch, the bypass is done by the two switches (SW1 and SW2), each one is in the state ON-OF, and the LNA2 is disabled too.
However, when using the output of the LM211 to control the switch block, the voltage falls down to 2V and it's not able to control the switches.
In fact, each switch is in the state ON-ON (and not ON-OFF as proposed) so the RF signal is mixing (LNA2 path, plus direct path) and producing intermodulation distortion at the output.
The input impedance of the inverter block is more than 100 Kohm and consumption current is only microamperes (basically the base current of NPN transistors and control current of the switches that is 50 microamperes for each one). The output current of the comparator (LM211) is milliamperes, and both collector output and emitter output configurations were performed for the LM211 (see data sheet).
I also tried voltages followers (with opamps) at nodes VA and VB, but it's not work. Even an optocoupler at node VA. The power detector is the AD8363 from ADI.
What I realized in all cases, and I don't understand, is that voltage at node VB is always 2.291V (>2.288V of reference voltage for the comparator) for a 10 dB power varying RF signal at the input (RFin). In other words: i need to increase the power of the RF input signal up to 10dB so that the bypass is done well by the two switches. But as I said at the beginning, if I open the circuit at node VA, and increase the power of RFin in 1 dB, it's enough for the AD8363 to detect it, thus VB > 2.288V (Vref) and the LM211 outputs a voltage of 5V. But when closing the loop at VA -> output (LM211) falls down to 2V and both switches are in ON-ON state, and VB is only 2.291V for a 10dB power varying RF input...
What's happening? Is there any error in the design?
I would be very grateful if someone could help me.
Many thanks.
Jose
Last edited:
