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.

ADC32J45EVM: Is the THS4541 working properly?

Part Number: ADC32J45EVM
Other Parts Discussed in Thread: THS4541, , OPA857, TINA-TI, ADC32J45, THS4551

Hello,

I have a question about THS4541 of ADC32J45EVM.

[Q]
My customers are going to implement OTDR using the ADC32J45EVM.
What they refer to is TI Designs Optical Front-End System Reference Design.

Therefore, ADC32J45EVM input (Bch) was remodeled to the path using THS4541.
Assuming that it is input from OPA857, the constant was referred to p30 Figure 35.
We also performed simulation using TINA-TI.
However, VIN/P which could be confirmed on actual machine was different from simulation.
VIN = 1.83V
sim: VIN/P = 1.69 V
test: VIN/P = 1 V

1882.180220_THS4541_ADC32J45EVM_Bch.TSC

Next, we compared the result of THS4541EVM with the simulation result of TINA-TI.
The result of the THS4541EVM test is the same as the simulation result.
Therefore, the THS4541EVM and TINA-TI determined that the simulation result was correct.

Subsequently, the input (Bch) of the ADC32J45EVM was modified to be the same as the THS4541EVM.
The results obtained are as follows.

(1) VIN = 1.83V
<sim>
VIP/N = 1.89V
<test>
VIP/N = 2V
1882.180220_THS4541_ADC32J45EVM_Bch_VIN=1.83V.TSC

(2)VIN = Open
<sim>
INN/P = 197mV
VIN/P = 771mV
<test>
INN/P = 500mV
VIN/P = 1.94V
4606.180220_THS4541_ADC32J45EVM_Bch_VIN=Open.TSC

Therefore, THS4541 of Bch thought that it is not working properly.

Next, I confirmed the THS4541 of ADC32J45EVM C ch.
Cch is an input that has never been remodeled, inputting a signal.
The input was Open, we confirmed the voltage of the THS4541 at the time of power-on.
<sim>
INN = 36.92mV
INP = 70.8mV
VIN/P = 316mV
VOP = 993mV
VON = 910.56mV
<test>
INN = 110mV
INP = 211mV
VIN/P = 0.94V
VOP = 0.97V
VON = 0.97V
8321.180220_THS4541_ADC32J45EVM_Cch_VIN=Open.TSC

Could you tell me why the THS 4541is like this?
Why is the THS4541 EVM operating properly, but the THS4541 of the ADC32J45EVM does not work properly?
Why does not it work properly even on THS4541 that neither remodel nor input?

There are two ADC32J45EVM, but the same phenomenon is occurring in both.

Thank you for your cooperation.

Best Regards,
Kaede Kudo

  • Kaede,

    We are looking into this.

    Regards,

    Jim
  • Jim, 

    Thank you for your cooperation
    I will wait for your feedback.

    We are using Rev. B.

    Best Regards,

    Kaede Kudo

  • Hi Kaede,

    Could you please confirm whether the Vin for the THS4541 is open or connected to 1.85V? It is a bit confusing from your post to determine exactly what it is.

    Also, could you please provide the actual schematic which is used for interfacing with the ADC32J45? It seems that the Vocm is open from the TINA schematic file you have attached which could be reason for the DC voltage shift in your test. The THS4541 Vocm should be connected to 0.95V in-order to properly interface with the ADC. Could you also confirm whether the THS4541 inputs/outputs are AC coupled ? I think it would help a lot if you could attach the actual schematic and we can work to solve your problem from there.

    Best Regards,
    Rohit
  • Hi Rohit,

    Thank you for your cooporation!
    And I'm sorry that the response was delayed.

    The schematic and simulation are attached below.
    Could you confirm it?

    Sorry to trouble you, but if you have any missing information or unclear point please contact me.

    (1)
    This is remodeled with reference to TI Designs Optical Front-End System Reference Design.

    <sim>
    VIN/P = 1.69V
    <test>
    VIN/P = 1V


    4380.2867.180220_THS4541_ADC32J45EVM_Bch.TSC

    (2)
    This is remodeled referring to THS4541EVM.
    The input is 1.83 VDC.

    <sim>
    VIP/N = 1.89V
    <test>
    VIP/N = 2V

    8422.7658.180220_THS4541_ADC32J45EVM_Bch_VIN=1.83V.TSC

    (3)
    This is remodeled referring to THS4541EVM.
    The input is Not Connect.

    <sim>
    INN/P = 0V
    VIN/P = 636mV
    <test>
    INN/P = 500mV
    VIN/P = 1.94V

    1641.180220_THS4541_ADC32J45EVM_Bch_VIN=NotConnect.TSC

    (4)
    Cch has never been remodeled.
    I have never input a signal to Cch.
    Evaluation was carried out in the state of just turning on the power to ADC32J45EVM.

    <sim>
    INN = 36.92mV
    INP = 70.8mV
    VIN/P = 316mV
    VOP = 993mV
    VON = 910.56mV
    <test>
    INN = 110mV
    INP = 211mV
    VIN/P = 0.94V
    VOP = 0.97V
    VON = 0.97V

    1145.5504.180220_THS4541_ADC32J45EVM_Cch_VIN=Open.TSC

    · Since the simulation results of THS4541 EVM and TINA-TI are consistent, I have determined that the simulation is correct.
    · I am wondering that THS4541 of Cch which does not perform remodeling or signal input will not be the same as the simulation result.
    So, I think " Is the THS4541 working properly?".

    I am grateful for your cooperation.

    Best Regards,
    Kaede

  • Hi Kaede,

    For all the scenarios you described above from 1 to 4, it seems that the DC input applied at VIN is blocked at the THS4541 input due to the ac coupling capacitors C11 and C19.
    For (1), no matter what the VIN would be, the THS4551 input (VIN/P) and output will both be set to 1V because the DC current is blocked by the ac coupling capacitors C11 and C19. I don't think you have modeled these caps in the TINA-TI simulator properly.
    Similarly for (2), the THS4541 Vocm is floating which will set its output to 2V on a single +4V supply and because of the ac coupling caps the VIN/P will also be set to 2V.
    Same scenario exists for (3) where you have the VIN open and THS4541 Vocm floating.
    So, I don't think there is an issue with the THS4541 device or its TINA-TI model.

    However, the underlying question you have is how to connect the THS4541EVM to the ADC34JxxEVM as shown in the Optical Front-end Systems Design, right? So, my recommendation would be to look at Figure 9 in the Optical front end systems design: www.ti.com/.../tiduaz1.pdf and implement your setup as it is without any DC blocking caps in-between. The THS4541 output after the two series 50-ohms will interface with the ADC32Jxx device input.

    Best Regards,
    Rohit
  • Hi, Rohit

    Thanks for your comment.

    Regarding C11 and C19, I think that it does not matter because I do not implement R217 and R21, R258, R226.
    Does C11 and C19 affect even the following circuits?

    For example (1), if I input 1.83 VDC to both BINP and BINN, I predicted that the output of THS4541 will be 0.95 V.
    And I confirmed that the prediction is right with simulation and actual measurement.
    However, I confirmed that the common mode input of THS4541 is different from simulation and actual measurement.
    I think that this is not normal behavior.

    Please let me know if there is a mistake in simulation or circuit.
    If there is missing information or unclear point, please tell me.

    My customers have not confirmed the correct movements with the four THS4541 on the two ADC32J45 EVM.
    EVM is RevB.

    Best Regards,
    Kaede

  • Hi Kaede,

    No. Based on what you are describing, I don't think the C11 and C19 will affect the following circuits if R217, R21, R258 and R226 are open.

    For example (1), I would expect the output of THS4541 as 0.95V because the device Vocm pin 9 is tied to 0.95V. However, I am puzzled as to why you are measuring 1V at the input, when you should have actually measured 1.69V - same as the TINA simulation. Did you by chance have the inputs open, when making this measurement? I am saying this because only if the input VIN is open, then the input common mode voltage of the THS4541 will measure 1V.

    Would it be possible for you to repeat example (1) as you have shown, and measure pins 1, 2, 3 & 4 of the THS4541 keeping the same setup with VIN = 1.83V? The result we are looking for is that pins 1 & 11 should be equal voltages, and also the pins 4 & 10 will be equal. Also, make sure the pin 12 or PDB pin is tied high to +VCC. It would not hurt to make sure that pin 9 stays at 0.95V as well.

    Best Regards,
    Rohit
  • Hi Rohit,

    Thanks for your comment!

    ・I confirmed that 1.83 V was input.
     This measurement was the same result no matter how many times I did, so I think accidentally input has never been Open.
    ・The PDB pin is tied high to +VCC.
     I also confirmed the operation when pulling down the PDB pins, so I think that the possibility that the THS 4541 was being powered down is low.

    I will measure pins 1, 2, 3 & 4 of the THS4541 and feed back results.

    Best Regards,
    Kaede

  • Hello Rohit,

    I apologize for losing contact.

    Since I did not have an environment to measure (1), I will tell you the results measured with other circuits (see figure below).
    The problem that is happening is the same, the common mode input is different from the simulation.
    Please let me know if you have any questions.

    (2)
    This circuit is remodeled with reference to THS4541EVM.
    Therefore, Vocm is 2V.
    VIN = 1.83V.

    1050.3113.180220_THS4541_ADC32J45EVM_Bch_VIN=1.83V.TSC

    (3)
    This circuit is not remodeling.
    I have never input a signal.

    8816.8883.180220_THS4541_ADC32J45EVM_Cch_VIN=Open.TSC

    I apologize that the measurement conditions have changed.
    However, I think the problem that is happening is the same.
    If there is an uncertain point, please tell us.
    Sorry, I appreciate your cooperation.

    Best Regards,

    Kaede Kudo

  • Hello Rohit,

    I think that the actual circuit and circuit diagram of EVM are different, can you confirm it?

    <schematic diagram>

    <Actual Circuit>

    I thought that it would be a problem for the common mode input to be different between simulation and measurement.
    However, when simulation was carried out with the actual circuit, almost the same result was obtained.
    Therefore, I think that the circuit of EVM is wrong.

    0435.6675.180307_THS4541_ADC32J45EVM_Bch_VIN=1.83V_ActualCircuit.TSC

    <sim>

    <test>

    The EVM I use is RevB.
    Could you confirm it?

    Best Regards,

    Kaede

  • Hi Kaede,

    I think you are correct that the actual circuit implemented and schematic of the EVM are different, and we are sorry for this error. It seems that the circuit connection needs to happen as shown in the EVM schematic available on the web, but unfortunately it is implemented the other way. We will work on fixing this issue on our ADC32JXX EVM in the near future.

    Best Regards,
    Rohit
  • Hi, Rohit

    Thank you for your comment.

    Is EVM rev C solved this problem?

    Best Regards,

    Kaede

  • Hi Kaede,

    I don't think we have released a rev C version of the EVM. However, let me confirm this and get back.

    Best Regards,
    Rohit
  • Hi Rohit, 

    Thank you for your comment.

    I will wait your feedback.

    Best Regards,

    Kaede

  • Hi Kaede,

    If you order Rev C of this EVM, then it should have the right schematic and correct implementation. I think the Rev C should be orderable from ti.com. Sorry for the mixup.

    Best Regards,

    Rohit

  • Kaede,

    Who did you order this board from? We would like to make sure they do not stock anymore Rev B boards. You should be able to request a customer exchange free of charge. If you need any help with this, please let us know.

    Regards,

    Jim

  • Hi Jim, Rohit,

    Thanks for your supports!

    My customer bought two EVMs.
    I will request a customer exchange free of charge.
    On this matter, should I request via sales?

    I appreciate your cooperation.

    Best Regards,

    Kaede

  • Kaede,

    Yes.

    Regards,

    Jim