[SOLVED] comparative op amp circuit using RSSI of 2 RF recievers ( beginer !!! )

[benbiles]

Hi, I am trying to understand how to use an OP AMP to compare two changing DC voltages of a RSSI ( receive strength signal indicator ).
I would like the audio outputs of two RF modules to flipflop to the higher RSSI DC voltage using a quiet switch IC as in the schematic..

I would ultimately like a delay to be set between the flipflop and then eventually I would like to change the flipflop switch to a more subtle fast pan
between the audio outputs of the modules.

But !! I feel I am running way ahead of myself on the OPAMP area of this design in progress !!

Do I need to use some resistors here to set some input parameters for the two DC inputs ? do I have this completely wrong !! should I be using two or even for OPAMPS ?
should I be using two schmitt triggers somehow to smooth things out if the RSSI voltage is erratic?

The Voltage will rise and fall from 1-3 vdc on either input depending on how strong the receive signal will be. However if the two voltages become too similar then it would be useful for the flipflop to latch for a given time.. ( wait for a stronger signal difference between the two voltages )

any ideas would be welcome.. I would like to avoid using a micro controller if possible since the dual receiver will be run on battery.

The LED lights are supposed to show which module is currently being output..



Thanks,

Ben

 

[sonicsmooth]

Hi, in general you have the right idea. you can use an opamp, or you can use a comparator. The comparator is likely to have some hysteresis built-in which will reduce a little bit the amount of switching given similar noisy RSSI values. Also please don't use the term "flipflop" unless you actually mean a real flip flop! In either case, you can use resistors to add additional hysteresis to the comparator, so you don't need a schmitt trigger.

You may also want to confirm whether your analog switch is a break-before-make, so you don't accidentally short the outputs of your audio thing together. This probably isn't a big deal given what's probably a few ohms of switch resistance, but it's good to have a clean design.

Something like ADG884 may make more sense for doing an audio mux. https://www.analog.com/en/switchesmultiplexers/analog-switches/adg884/products/product.html

In either case, your audio will probably make pops and other noises when switching. Is your audio output from Rx module always between ground and 3.0v? If not, you will need to add some opamp circuit to bias it up from under ground.

Your LED2 will never light up. Currently the only way to light it up is if the pointy end (cathode) goes below ground, which it won't do. If you want LED2 to light up, then you need to change TR1 to NPN (with a ~100 ohm resistor before the base), and put LED2 in series with TR1 and R3, pointy end down. Even easier to get one LED for each state is just to let the comparator drive them directly. In this case, you delete TR1 and just put each LED (and its 10k resistor) in series from VCC to ground, then put the comparator output in between. So when the comparator outputs high, only the bottom LED+R lights up, and when the comparator outputs low, only the upper LED+R lights up.

Also 10K for an LED is too high. It'll barely make any light. Use something like 330 ohm, 100 ohm, etc. to get some brightness.



For fast panning, you may want to make your own analog switch. In this case you can get a complementary CMOS pair (nfet and pfet) for each channel, then use some RC circuit to control the gate so it's slow. Put a resistor in series with each channel so you don't short them out.

Also, as for the microcontroller, you are likely to burn more power in your comparators, audio RX, opamps, and LEDs than in the micro, if it's programmed properly to use power-down states, slow clock, etc. Anyway you're right, a micro is overkill for this application.

Hope this helps

Hi, I am trying to understand how to use an OP AMP to compare two changing DC voltages of a RSSI ( receive strength signal indicator ).

I would like the audio outputs of two RF modules to flipflop to the higher RSSI DC voltage using a quiet switch IC as in the schematic..

I would ultimately like a delay to be set between the flipflop and then eventually I would like to change the flipflop switch to a more subtle fast pan
between the audio outputs of the modules.

But !! I feel I am running way ahead of myself on the OPAMP area of this design in progress !!

Do I need to use some resistors here to set some input parameters for the two DC inputs ? do I have this completely wrong !! should I be using two or even for OPAMPS ?
should I be using two schmitt triggers somehow to smooth things out if the RSSI voltage is erratic?

The Voltage will rise and fall from 1-3 vdc on either input depending on how strong the receive signal will be. However if the two voltages become too similar then it would be useful for the flipflop to latch for a given time.. ( wait for a stronger signal difference between the two voltages )

any ideas would be welcome.. I would like to avoid using a micro controller if possible since the dual receiver will be run on battery.

The LED lights are supposed to show which module is currently being output..



Thanks,

Ben

 

[benbiles]

Hi sonicsmooth!! thanks so much for your brilliant tips !!! I have tried to correct some things in the drawing.. not sure I have any of it correct!!

Yes the output on module A + B would be line level audio +- 1v swing so I have tried to add 2 low power audio op amps to the drawing.
I Currently have no idea what resistors / capacitors to use for them to bias the audio to swing between 1.5 and 0v.

Am I correct in thinking the switches in ADG884 cannot switch +-V therefore signal needs to be above 0v?
I suppose this is also true if I can build a delayed switch panning circuit?

Am I correct in thinking I need 2 opAmps to reverse the process the other side of the switch / switch(pan)? I would need +-1v line level audio output.

Do you know of any example schematics of a delayed switch pan? I could take a look at.. I included the CMOS pair (nfet and pfet) off to the right but have no idea how to use them yet !! I better learn the difference between a transistor and them quickly.

TI have a reference design called swra394 of a digital gfsk true diversity audio RF receiver that uses micro possessor. it has two receivers but switches the received packets from each receiver based on RSSI level. very clever but I'm just trying to make a simple system that I can add to different RF receiver boards without too much trouble so I can test the usability of audio links in different RF frequencies with the redundancy (true diversity) aspect

design **broken link removed**

Your right in saying the I'll probably use more power with the comparator and LED's !!! I didn't realize comparators can burn so much power !!!
+ I think there is a MC on the audio modules I'm testing !! Haha so yes forget my power argument

anyway here's another drawing, sorry about messy drawing of the ADG884 , I couldn't find a way of bringing a spice file of it into TINA-TI..

Is there an pan controller IC I could use of some kind?

thanks again, any more ideas from anyone much welcome!!

BEN

PS sorry about calling it a flipflop!!

Hi, in general you have the right idea. you can use an opamp, or you can use a comparator. The comparator is likely to have some hysteresis built-in which will reduce a little bit the amount of switching given similar noisy RSSI values. Also please don't use the term "flipflop" unless you actually mean a real flip flop! In either case, you can use resistors to add additional hysteresis to the comparator, so you don't need a schmitt trigger.

You may also want to confirm whether your analog switch is a break-before-make, so you don't accidentally short the outputs of your audio thing together. This probably isn't a big deal given what's probably a few ohms of switch resistance, but it's good to have a clean design.

Something like ADG884 may make more sense for doing an audio mux. https://www.analog.com/en/switchesmultiplexers/analog-switches/adg884/products/product.html

In either case, your audio will probably make pops and other noises when switching. Is your audio output from Rx module always between ground and 3.0v? If not, you will need to add some opamp circuit to bias it up from under ground.

Your LED2 will never light up. Currently the only way to light it up is if the pointy end (cathode) goes below ground, which it won't do. If you want LED2 to light up, then you need to change TR1 to NPN (with a ~100 ohm resistor before the base), and put LED2 in series with TR1 and R3, pointy end down. Even easier to get one LED for each state is just to let the comparator drive them directly. In this case, you delete TR1 and just put each LED (and its 10k resistor) in series from VCC to ground, then put the comparator output in between. So when the comparator outputs high, only the bottom LED+R lights up, and when the comparator outputs low, only the upper LED+R lights up.

Also 10K for an LED is too high. It'll barely make any light. Use something like 330 ohm, 100 ohm, etc. to get some brightness.



For fast panning, you may want to make your own analog switch. In this case you can get a complementary CMOS pair (nfet and pfet) for each channel, then use some RC circuit to control the gate so it's slow. Put a resistor in series with each channel so you don't short them out.

Also, as for the microcontroller, you are likely to burn more power in your comparators, audio RX, opamps, and LEDs than in the micro, if it's programmed properly to use power-down states, slow clock, etc. Anyway you're right, a micro is overkill for this application.

Hope this helps

 

[benbiles]

OK,

Here is another drawing,

have tried to work on the audio Pan part and added some resistors to the OpAmps.

Any advice on where to go from here?

I am sure most of these resistor values are wrong !!!

An example switch controlled with some RC anyone ? I'm not sure the Audio will make it through the one I have Plagiarized from a musical instrument boss de-click pedal !!

Also if anyone has an example of how to Bias audio +-V to 0v-+3v would be great !!

Thanks,

BEN

 

[benbiles]

woops, diode and caps shouldn't be there !! update..

There must be an IC that can route one of multiple audio inputs to an output in a quiet way by panning or something ? anyone ?

 

[FvM]

It's very unlikely that the FET circuit does what you want. At least transistor DC bias and R and C values seems to be far off.

 

[benbiles]

It's very unlikely that the FET circuit does what you want. At least transistor DC bias and R and C values seems to be far off.

Yes, thanks FvM.. I think I might be going down a blind ally with these FET's at the moment anyway !!!

I'm thinking I might be better off using VCA's ( voltage controlled amps ) to control audio level with some RC delay so that you get a fast pan between module audio as the comparator switches output state...

Do you think this might work if I can setup the VCA's properly ?

There's some quite nice looking low noise VCA's about including these from ti... and I'm thinking that if they reach 0volts on Vcc then no audio would go through?

 

[FvM]

My first comment is a general one. I'm missing a requirement specification. Without it, nobody can rate the discussed design ideas.

Assuming typical audio signal quality requirements, VGA810 won't be ever used.

Secondly, the standard solution for the addressed problem is a diversity receiver which switches the IF signal.

To avoid permanent switching, the comparator circuit should have at least sufficient hysteresis, or preferably evaluate the switching criterion based on both RSSI values and time, e.g. switch immediately in case of signal dropout, delay switching to the stronger signal if the other signal has still sufficient quality. Under this conditions, hard switching may be an acceptable method.

 

[benbiles]

My first comment is a general one. I'm missing a requirement specification. Without it, nobody can rate the discussed design ideas.

Assuming typical audio signal quality requirements, VGA810 won't be ever used.

Secondly, the standard solution for the addressed problem is a diversity receiver which switches the IF signal.

To avoid permanent switching, the comparator circuit should have at least sufficient hysteresis, or preferably evaluate the switching criterion based on both RSSI values and time, e.g. switch immediately in case of signal dropout, delay switching to the stronger signal if the other signal has still sufficient quality. Under this conditions, hard switching may be an acceptable method.

Hi FvM, I guess the requirement / specification is to switch high quality audio without pops and clicks to the module with the higher rssi level. I agree hysteresis needs to be added. I think i mentioned that already before. I like your idea of switching quicker based on the speed of the RSSI level change?.. maybe I could use this as a factor in the speed of the pan to avoid audible clicks instead of hard switch. fast pan of 5ms would be better than switch. anyway I agree not much difference in RSSI should = no change.. but i think that is hysteresis..

How about this VCA instead of VCA810 ?
http://www.thatcorp.com/2162_Dual_Pre-Trimmed_Blackmer_IC_Voltage-Controlled_Amplifier.shtml

maybe they would have better audio quality?

I am sure there are better ways to have diversity but audio panning between receiver output isn't new in some high end professional broadcast radio receivers. as here **broken link removed**

I am sure this example of RF diversity by selecting packets modulated over GFSK on UHF with the higher RSSI is perhaps one of the better ways?
**broken link removed**

but my idea is to make a simple audio switcher/pan triggered by RSSI so I can test audio sent in different ways.
could be 2.4ghz packet data or UHF WBFM compressed audio with a compander circuit...

looks like early days for this project anyway !!

 

[FvM]

THAT 2162 or any similar device that has exact gain and low distortion in "fully on" state should work.

 

[benbiles]

My first comment is a general one. I'm missing a requirement specification. Without it, nobody can rate the discussed design ideas.

Assuming typical audio signal quality requirements, VGA810 won't be ever used.

Secondly, the standard solution for the addressed problem is a diversity receiver which switches the IF signal.

To avoid permanent switching, the comparator circuit should have at least sufficient hysteresis, or preferably evaluate the switching criterion based on both RSSI values and time, e.g. switch immediately in case of signal dropout, delay switching to the stronger signal if the other signal has still sufficient quality. Under this conditions, hard switching may be an acceptable method.

Hi FvM, I guess the requirement / specification is to switch high quality audio without pops and clicks to the module with the higher rssi level. I agree hysteresis needs to be added. sonicsmooth mentioned that already which is why I switched to a comparator rather than an OpAmp. I will try to add values for Hysteresis later. I like your idea of switching quicker based on the speed of the RSSI level change? Do you know any easy way of doing that? Maybe I could use this as a factor in the speed of the pan to avoid audible clicks instead of hard switch. fast pan of 5ms would be better than switch. anyway I agree not much difference in RSSI should = no change.. but i think that is hysteresis..

I was thinking of using a 555 timer to ramp up/down the VCA's to make the pan. maybe i can add some logic in there that controls speed of the pan!
if RSSI Vdc difference > 25 % then switch FAST ! but how do you calculate a fast drop in RSSI Vdc ! tricky !

How about two of these VCA's instead of VCA810 ? maybe they would have better audio quality?
http://www.thatcorp.com/2162_Dual_Pre-Trimmed_Blackmer_IC_Voltage-Controlled_Amplifier.shtml

I am sure there are better ways to have diversity but audio panning between receiver output isn't new in some high end professional broadcast radio receivers. as here **broken link removed**

I am sure this example of RF diversity by selecting packets modulated over GFSK on UHF with the higher RSSI is perhaps one of the better ways?
**broken link removed**

but my idea is to make a simple audio switcher/pan triggered by RSSI so I can test audio sent in different ways.
could be 2.4ghz packet data or UHF WBFM compressed audio with a compander circuit...

looks like early days for this project anyway !!

 

[benbiles]

Hi FvM,

do you know where i could take a look at any 'diversity receivers that switches the IF signal' schematics ? I'm also interested in that approach !!
I think putting together a WBFM receiver demodulator etc would be out of my capabilities entirely at this stage but maybe I could get some ideas from
the way the comparator is setup and the way the IF signal is switched

I've been avoiding using an MC in this project because I'm trying to pick up some basic skills with IC's and components before I take on MC's with SPI interfaces and the like !! that's way out of my depth !!

 

[benbiles]

OK, VCA's changed on drawing with some AGC on audio inputs with the help of someone called Adam on another forum !

need to work on ramping the voltages up and down next for panning audio..

 

[benbiles]

OK, VCA's changed on drawing with some AGC on audio inputs with the help of someone called Adam on another forum !

need to work on ramping the voltages up and down next for panning audio..

picture didn't make in last post

 

[benbiles]

ok, I should start this again !

Just found out the PTP modules I wanted to use for this project do not have RSSI output !!!!

OK !! so I'm thinking I should start this thread again and call it high quality radio mic receiver.

This time I should Include UHF PLL recievers that do actually have RSSI output...

I have also looked into using a PIC micro-controller and found that some have comparator(s) on board. so I'm thinking it makes more sense to use one to compare RSSI and switch RF signal from the Two RF receivers before it gets to a WBFM modulator..

I feel like i am re-inventing the wheel ( radio mic )

I tried to come up with a digital version but have come to the conclussion that most DSP's that
can encode / decode at the required compression to fit digital into 200khz RF space ( maximum allowed ) would require too much power. Therefore WBFM modulated audio compressed and expanded with a compander is the better way to go for now..

So If your looking for a continuation search for ' High Quality radio Mic Receiver '

BEN