LMH0387: 5940MHz emission failing

Rob Yost

Part Number: LMH0387
Other Parts Discussed in Thread: LMH0397

We have a design which uses the LMH0387 SDI Cable Equalizer. During pre-scans for emissions compliance I came across a very high emission at 5940 MHz. It seems to be coming directly from the chip. Has anyone experienced this. Could TI tell me what inside the chip could be running at a frequency that could cause this harmonic? Are there any known solutions to this?

over 6 years ago

0 Nasser Mohammadi over 6 years ago

TI__Guru* 93670 points

Hi Rob,
5940MHz is second harmonic of 2970MHz. This is the max data rate that LMH0387 operates. I am thinking you see this emission when there is a signal going through the device. Right? LMH0387 has not been qualified or checked for emission or radiation.

Regards,,nasser

0 Rob Yost over 6 years ago in reply to Nasser Mohammadi

Prodigy 90 points

Hi Nasser. We're working on a debug strategy to determine if it's the data or just the chip when it's powered. I'll let you know as soon as I find out. Are there any other high-speed clocks running inside that silicon that would produce 5940MHz as a harmonic?

0 Nasser Mohammadi over 6 years ago in reply to Rob Yost

TI__Guru* 93670 points

Hi Rob,
There is no clock or VCO in this data range. LMH0387 has integrated equalizer and cable driver. Maximum running frequency is under 100MHz.

0 Rob Yost over 6 years ago in reply to Nasser Mohammadi

Prodigy 90 points

Hi Nasser,

Could you answer a question for me? What is the reason for the instruction in the part's datasheet to ensure there is no metal (traces or vias) directly underneath the component? It is shown in Figure 12 on page 22 of the datasheet.

Thanks!

0 Nasser Mohammadi over 6 years ago in reply to Rob Yost

TI__Guru* 93670 points

Hi Rob,
Actually on this device, die is placed on a small pcb. This pcb traces would interact and affect one another if there is metal under the part.
Regards,,nasser

0 Rob Yost over 6 years ago in reply to Nasser Mohammadi

Prodigy 90 points

They accidentally placed ground flood directly under the part, and we're seeing a rather high emission at 5940 MHz. And I thought I saw a dramatic improvement by scraping this ground flood off the PCB and replacing the part. I'm wondering if something could be coupling to the grounded metal under the part and causing this emission to be so high.

0 Nasser Mohammadi over 6 years ago in reply to Rob Yost

TI__Guru* 93670 points

Hi Rob,
This is possible. We have high speed traces with fast slew rate and it may cause this emission.
Please note there is a new device - LMH0397 - where you can place ground or thermal pad under the device.
Regards,,nasser

Interface

Interface forum

LMH0387: 5940MHz emission failing