LMH0302: 3G SDI Buffer Operation and Alternatives

Hi,

Our team is looking into this and will get back to you shortly. 

Greetings,

LMH0302 is a limiting device - meaning it digitizes the signal to one or zero. So it is not a true analog buffer per say. 

Additionally LMH0302 is mainly used for 3G SDI and the bandwidth/rise/fall time is limited. The amplitude - as you noted is independent of the input signal.

Can you please let us know the application, protocol, or a block diagram of this design. If you like, you can send me a friend request to send me any confidential info you may want to exchange.

Regards,Nasser

Hi Nasser,

The application is for testing HD-SDI and 3G-SDI signals.  In this particular sub-section of the design SDI signals are routed and switched via 50 ohm microstrip traces, with amplitudes up to 800mV.  This is all done with broadband resistor networks and RF relays.  I've prototyped this part of front end with good results in terms of bandwidth/rise-time and return loss.

The selected signal needs to be sent out to a waveform monitor via 75R coaxial cable, so active impedance transformation and cable drive are required, with minimal alteration of the original signal.

Let me know if you require anything further.  Thanks.

Greetings,

Given this is SDI application to begin with, then the only thing we need to be cognizant of is related to the single ended versus differential input. LMH0302 or other SDI parts are characterized using differential input versus single ended input.  Single ended input could work but data sheet spec is based on differential input.

How many SDI signals - inputs and outputs - are we talking about?

Regards,Nasser

For this section there are 4 inputs being muxed to 1 output.  On the small scale prototype I am able to switch this output to an LMH0344 with a 50R single-ended matching/termination network at the differential input (similar to Figure 3 of the LMH0344 datasheet).  This approach seems to work quite well with the input return loss well below the SMPTE spec for 3G-SDI.

If possible, can you please send me a block diagram all the way from inputs to output? LMH0344 can be used either in 50-ohm or 75-ohm input.

Regards,Nasser

Hi Nasser,

Got your suggestion for the LMH0344 parts.  However, as mentioned in the original post, we cannot use anything with a reclocker/retime/equalizer, etc as this (to my understanding) will significantly alter the incoming signal we are trying to inspect with the waveform monitor.  I have them on the prototype board for other testing purposes.

We ideally would want an opamp buffer with sufficient GBW for a 3G-SDI signal providing 2.353 times gain (1 times gain is ok).  I asked about the LMH0302 as it is mentioned as a potential buffer in a different E2E thread.  Is there someone who is more familiar with the analog catalogue that might be able to suggest something?  Thanks

I believe you need to analyze the raw waveform. There are some test equipment manufacturer who do the same thing and use wide bandwidth op-amp to measure incoming jitter and etc. All of our SDI parts are limiting and not good for this application.

We are assigning this to the high speed amp group to comment on the availability of the high speed amp with at least 3GHz bandwidth.

Regards,,Nasser

Ok, thank you.  Is there someone I can follow up with on this request?

Greetings,

High speed amplifiers or 1:2 muxes could be applicable here.

Regards,nasser