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.

LMH0346: SDI data splitting circuiting

Part Number: LMH0346
Other Parts Discussed in Thread: LMH0307, LMH0344,

hello, I am trying to equalize, re-clock and split SDI signals. The input is a single ended HD-SDI signal passed via a 75Ohm track. Now, the cable equalizer converts it into a differential signal which is 100 ohms. For this I am using the LMH0344, LMH0346 and LMH0307. I came across a post in this forum regarding a similar query. The post - 


Now, the post states that LMH0307 accepts 50 Ohm signal. How so ? Could you point to the portion in the datasheet that states so ? Now, supposing I use the LMH0344 followed by LMH0346 and LMH0307. The LMH0344 will be given a single ended 75Ohm HD-SDI data. This will convert single ended data to differential (100 ohm). The LMH0346 will reclock and drive differential signals (HD-SDI) to LMH0307. This will finally split the HD-SDI to 2 different signals. Is my understanding correct ? In the forum it is mentioned that LMH03466 expects 50Ohm input. If it is differential, how can it accept 50 Ohm ? Also, how is SDI differential pair ?

  • Hi Vinod,

    1). Please note LMH0307 data sheet application circuit figure 4. In the left hand side, you see two 49.9-ohm termination resistors. It means LMH0307 accepts 50-ohm single ended or 100 ohm differential input. These are the same and they are used through the data sheet interchangeably. 

    2). Your idea of LMH0344+LMH0346+LMH0307 is correct and should work fine.

    3). LMH0346 needs external 100-ohm termination resistor as noted in the application circuit of data sheet(Figure 6). Data sheet uses 50-ohm or 100-ohm input interchangeably. 

    Regards,,nasser

  • Hi Vinod,

    You are correct that the inputs to the LMH0346 and LMH0307 input is designed for 100-Ohms differential, and therefore the traces going into these devices should also be 100-Ohms differential. Sorry for the confusion.

    We commonly refer to the impedance interchangeably for the LMH0346 and LMH0307 input as "50-Ohm Single-Ended" or "100-Ohm Differential." This is a means to differentiate between this characteristic impedance and the one used by the cable (75-Ohm single-ended).

    The reason we also say "50-Ohms" is due to the need for an external 50-Ohm resistor on each input polarity in order to terminate properly. Note that this does not necessarily imply that the differential traces, when measured single-ended, measure to be 50-Ohms. As the traces become more tightly coupled, the electromagnetic effect of the other polarity will begin to affect the single-ended impedance.

    Thanks,

    Michael
  • Hello Michael,

    Thank you for the clarification. So, just to explain my understanding - this chip (just like other similar functional chips) accepts both single and differential inputs. The circuit in fig 4 shows provision for both. In case it is 50 Ohm single ended - I un-populate the 49.9R connected to /SDI and. It is a Series-RC parallel termination. In case of 100 ohm differential signals I keep it as is.

    Now, my aim is to split HD-SDI signals. So, is it not simple to use just am LMH0307 instead of the 3 IC solution ? What are my benifits ? I understand that the LMH0344 would be quite useful if I use longer cable lengths. But I am using a cable (RG179) of less than 100cms. But on the flip side I am using passives to convert 75Ohm signal into 50 Ohm. Will this have an impact ? Is there any other lossless (theoretical) to do the same  ?

    Vinod

  • Hi Vinod,

    For SDI application cases, we expect the signals after the Cable EQ to be differential 100-Ohm, so we recommend you keep as is.

    Regarding your next question, can you clarify whether you are designing a 1:2 Distribution Amplifier? It seems to me that you are trying to take one 75-Ohm signal and split to 2 x 75-Ohm outputs.

    The Cable EQ is designed to account for insertion loss due to long cable lengths. Looking at your use case, where your cable is less than 1m, you may not need such a device. If all that is necessary is a 50-to-75 Ohm converter, you can accomplish this with passives, such as below:

    The impact of using a passive like this is that there will be attenuation due to the passive. From this type of circuit, we expect an attenuation of approximately -7.8dB. If you wish not be subject to this type of attenuation, I would recommend going with just the LMH0344 -> LMH0307 without the reclocker, given that the interfacing cable before the Cable EQ is less than 1m long.

    Thanks,

    Michael

  • Hello Naseer,

    I have finished my layout for this board. Is it possible that I send it to you for verification before production please ?
  • Hi Vinod,

    Yes, we can help double-check the board layout. Please provide to us for review.

    If you wish not to post on E2E, please provide us your work e-mail so that we can communicate off-line.

    Thanks,

    Michael 

  • email - vinodkaruvat@gmail.com