Rugged LVDS Driver/Receiver

Dana Monemi14

Prodigy 30 points

All,

I am using the SN65LVDS1DVBR driver and the SN65LVDT2DBVR, both in a hostile environment. I am suspecting

that there is EMI pickup on the LVDS lines that is jeapordizing the Absolute Maximum ratings of the parts. Can anyone

suggest more rugged parts by TI (with extended ratings). I don't believe it can be found in the SOT23-5 package I am

using (both parts), so I may end up having to respin.

Thanks in advance. Sorry if this question is on the wrong forum. Please advise if it belongs on a different forum.

In the meantime I will be searching through the TI parts pages.

over 9 years ago

0 Michael Peffers over 9 years ago

TI__Mastermind 20820 points

Hi Dana,

LVDS is immune to noise inherently due to the differential nature of the bus. I wrote a short blog regarding differential signals that explains why LVDS is immune to noise. You can view this here: http://e2e.ti.com/blogs_/b/analogwire/archive/2014/02/21/get-connected-differential-signaling?DCMP=awire&HQS=hpa-int-hsi-awire-20150213-getconnected-awire-20140214-en

Do you have any scope captures that you could provide of the bus so that we can examine the signal levels?

Are you seeing IC failures in your system or bit errors that lead you to believe that their is something wrong with the bus?

0 Dana Monemi14 over 9 years ago in reply to Michael Peffers

Prodigy 30 points

Michael,

Thanks for the reply. Yes we have scope captures. I need to coordinate with another team member to obain those and post. Can I post those or do I need to send them to someone in a PM? We may also need to change our trigger to show both AC and DC offset so both can be seen simultaneously.

We are seeing IC failures. We have noticed two things. On the Driver we have seen that one leg swings and the other goes flat line. There doesn't appear to be a preference for _N or _P to go flat. Equal statistics on both. On the Receiver we are seeing the 100-Ohm termination open up.

What we THINK is happening is that EMI pick up is causing the two legs to move above the absolute max or below the absolute min voltage rating.

0 Michael Peffers over 9 years ago in reply to Dana Monemi14

TI__Mastermind 20820 points

Hi Dana,

You can post the screen captures in the forum or send them to me offline: mpeffers@ti.com (whatever you prefer)

Could you also provide me the portion of the schematic dealing with the LVDS driver, receiver and any signals that you think are coupling into the devices? Also, any details that you think are pertinent like what does the bus look like (i.e. backplane or ribbon cable, etc...)

Could the high energy spike be coupling in on the driver (single ended) side of the device and not the LVDS side? Are all the IC failures on the driver side only or is the receiver failing as well?

0 Dana Monemi14 over 9 years ago in reply to Michael Peffers

Prodigy 30 points

Michael,

I though I responded to your last post but somehow it didn't take. I am in North Carolina and been snowed in for a couple days. I will send you a diagram of our setup and as soon as I coordinate with the other team members get you some scope captures that you could hopefully comment on.

One of the options, pending investigation of root cause, would be to respin our boards. Are you aware of any LVDS drivers/receivers that are rugged/robust? I'm thinking we may just have to go to a bigger package if it comes to that.

0 Michael Peffers over 9 years ago in reply to Dana Monemi14

TI__Mastermind 20820 points

Hi Dana,

I am going to close this post since we hve moved off line with our conversation.

Interface

Interface forum

Rugged LVDS Driver/Receiver