This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Wideband 2x1 RGB Multiplexer

Other Parts Discussed in Thread: OPA3684

I am simulating the Low-Power RGB Mux/Line Driver Application Note for the OPA3684 (Figure 6 in the OPA3684 Datasheet   SBOS241C – MAY 2002 – REVISED JULY 2008).

The simulation is done in TI-TINA (Version 7.0.80.96 SF-TI    Build date: 2009 April 6, 11:48:24 AM).

 

I'm a little confuse because if I draw (build) the circuit as in the example (Figure 6) I don't get the expected voltage levels.

 

For the simulation I'm using 700 mv for Input 1 ( R1,G1,B1 = 700 mv) and 400 mv for Input 2 (R2,G2,B2 = 400 mv). Let's say that if I

disable Input 2 , I would expect an output voltage of around 1400mv (due to the gain of 2, really is around 2.18) but I get another

voltage level. I also added the 75ohm load of the RGB Line.

My question is that if the 78.7ohm limiting resistor goes in each output as in Figure 6.  Or it just goes in each V Out , I mean

 a 78.7 ohm resistor in VOut Red, 

a 78.7 ohm resistor in VOut Green, 

a 78.7 ohm resistor in VOut Blue,

so instead of the six (6) 78.7 ohm resistor I would just need three(3).

 

My second question is that the circuit uses 681 ohm resistor and 806 ohm resistor for the feedback network , but those are not necessary

commercial values. for the 681 ohm the solution is kind of simple a 680 ohm plus a 1ohm resistor, for the 806 ohm, I could use 470 ohm plus

330 ohms , for 800 ohm , ( I could use more resistors or a combination in parallel, but I think 470+330 could be a good replacement for the 806).

With 680 and 800 the gain is about 2.17 , close to the original 2.18.  Well , my real question is that in the Application note says that:

 " When one channel is shutdown, the feedback network is still present, slightly attenuating the signal and combining in parallel

with the 78.7 ohm to give a 75 ohm source impedance."

 

so what they mean is that  ( 681+806 || 78.7 ) = ( 1487 || 78.7) = 74.74 ohms (eq. 1)

or the 78.7 resistor in each output is also part of the feedback network ?

(681+806+78.7 || 78.7) = (1565.7 || 78.7 ) = 74.93 ohms (eq. 2)

The reason of my question is that if I use 680 ohms and 800 ohms resistors I guess the values of the 78.7 resistor wouldn't match correctly

the 75 ohm impedance of the load, plus 78.7 it is not a commercial value. Using eq. 1 i find a value of 79 ohms for the resistor, which would

make a 74.99 ohm in combination with the feedback network.

 

Figure 6

 R1 ----------- [ op amp ] --------- 78.7 Resistor -------|---------> Vout Red

R2 ----------- [ op amp ] --------- 78.7 Resistor---------|

 

Or I should be like this ?

 R1 ----------- [ op amp ] -------------|------- 78.7 Resistor--------> Vout Red

R2 ----------- [ op amp ] --------------|

 

Any advice, clue, suggestion, link to another application notes, the name of book, questions, everything is really appreciated and welcome.

My goal is to build a circuit to switch between the signal of a PC and a Laptop to show it in a projector, also the Monitor of the PC it's going to be working.

 

 PC ------------------------------------|----------------------------------  Monitor

LapTop -------------------------------|-----------------------------[ PC | Laptop switch] ------ Projector

 

Thanks.

Ernesto Arzabala

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  • Hi Ernesto,

    To start, could you upload the TINA schematic you're working with?

     

    Thanks,

    Nick

  •  Hi Nick,

    Here it is.

    Thanks

    vga_dist_b.TSC
  • Hi Ernesto,

    You are seeing odd behavior in your simulations because the disable function of the op amp model is not fully modeled.  Specifically, the inputs and outputs of the amp will go into a high impedance mode when it is disabled, but the model does not properly display this.  So what you are seeing is the result of one channel not "turning off" and thus loading down the output voltage.

    If you "break" the circuit by disconnecting the output of the disabled op amp, you will then see the appropriate behavior.  I made a slight modification to your schematic and re-uploaded it here as a quick example.

    Additionally, you mentioned that you're expecting to see about 1400mV at the output because of a gain of ~2V/V.  You should plan to see ~700mV at the output since the matching network (75ohms in series + 75ohm load) will cut the signal in half.  This is expected and the reason why video drivers typically have a gain of 2V/V.

    Regarding your resistor value selection, you should have no trouble finding standard surface mount resistors with the values listed in the datasheet (681, 806, 78.7 ohms).  These are common values and can be picked up easily from somewhere like Digikey.

    Regarding your question about the feedback network slightly attenuating the signal, your second calculation of (681 + 806 + 78.7 || 78.7) = 74.93 ohms is correct.  Returning to my initial point, when the output of the amp is not accurately modeled as high impedance in shutdown, it factors into this calculation and screws up the impedance in parallel to the output, and thus, also screws up the output level.

    Finally, we recommend keeping the 78.7ohm resistors at the output of each amp (as outlined in your first example and in the datasheet) for accurate matching when channels are disabled.

    I hope this helps.  Please let me know if you have any additional questions.

     

    Regards,

    Nick

    vga_dist_b[1]_modified.TSC
  •  

     Thanks a lot Nick.

    About the resistors value, I didn't look at the values available for surface mount resistor, but I will be more than happy to get the values according to the original circuit.

    But I was choosing the resistor values on the basis that I wasn't able to get those resistor values, after all that is the reason we  have/learn circuit laws, op amp equations  and simulations tools.

    =)