• State Answer Suggested
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 178 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

AM26LV31E: AM26LV31E + AM26LV32E

Max Pom
Intellectual 350 points
Part Number: AM26LV31E
Other Parts Discussed in Thread: AM26LV32E, THVD1450, SN65LVDS391, SN65LVDS348, SN65LVDT348

Tool/software:

Hello,

We are willing to use the couple of RS-422 devices (transmitter and receiver): AM26LV31E + AM26LV32E.

There is a restriction on the timing of the channel, the total jitter needs to be low (The signal needs to be sampled at specific time).

The jitter parameter is not mentioned in the datasheet, is it possible to get this parameter for both devices? (Maximum Jitter) 

(Maybe you can suggest some other device which is designed and dedicated to have low jitter?)

Thanks!

Max

  • 0
    Guru 311010 points

    In general, you get the lowest jitter by using a device with a lower propagation delay, i.e., with a higher maximum speed. The only 50 Mbps RS-4xx devices are 1+1 transceivers like the THVD1450.

    LVDS allows higher data rates and is commonly used for clock signals; the SN65LVDS391 and SN65LVDS348/SN65LVDT348 are more than ten times faster and have more detailed timing specifications.

  • 0
    Intellectual 350 points

    Thanks,

    Yes, I know that LVDS is faster and with lower jitter, but I have a requirement to use RS-422 for EMI/RFI immunity issues..

    Is it possible to receive the jitter parameter from TI? Maybe measurement..?

    Ia already found a device which is faster and there is an eye diagram measurement in the datasheet where I can see the jitter which is good anough for me.

    If the jitter of AML26 is not much worst, I prefer to use it.. 

    (page 15: MAX22502E - 100Mbps Full-Duplex RS-485/RS-422 Transceiver for Long Cables)

    Thanks!

  • 0 in reply to Max Pom
    Guru 311010 points

    The specified skew limits are an upper bound for the jitter. More information does not exist.

  • 0 in reply to Max Pom
    TI__Guru 67380 points

    Hi Max,

    We don't really guarantee any sort of Jitter parameter for our devices and it can be system dependent (using lower quality cables will make jitter worse, longer cables will result in higher jitter in all systems as well).

    We normally don't provide values but I just happen to have some data I collected a while back (one unit at room temp).

    Below is 10 feet cabling (CAT5e rated cable with 100 ohm termination)

    Below is 100 feet cabling (CAT5e rated cable with 100 ohm termination)

    These were tested on a breakout board so it's not the best set up but the measurements we got were in the ns range for jitter.

    -Bobby

  • 0 in reply to BOBBY
    Intellectual 350 points

    Hi Booby,

    Thanks a lot, this is really helpful! 

    is this for a single device (transmitter/receiver), or the couple AML26LV31E + AML26LV32E ?

    My cable length is much shorter, only 1 feet..

    I guess I can assume that if :

    100 feet = 5.25 nsec ,10 feet =1.5 nsec, (/3.5), than 1 feet ~ 0.5nsec

    (if its for the transmitter / receiver only, then the total jitter is ~1nsec)

    What do you think?

    Thanks!

    Max

  • 0 in reply to Max Pom
    TI__Guru 67380 points
    Max Pom said:
    is this for a single device (transmitter/receiver), or the couple AML26LV31E + AML26LV32E ?

    Yes, this test had one transmitter and one receiver. in a direct point to point connection. 

    Max Pom said:

    I guess I can assume that if :

    100 feet = 5.25 nsec ,10 feet =1.5 nsec, (/3.5), than 1 feet ~ 0.5nsec

    I'm not 100% sure if they are linear like this but this is probably a decent estimate. 

    Max Pom said:

    (if its for the transmitter / receiver only, then the total jitter is ~1nsec)

    What do you think?

    Again, it's system dependent but I believe you could get these numbers. My guess is you would probably get better numbers than me since your set up will also include impedance matched traces (mine were not impedance matched to the cable). If you use ribbon cables for example, your measurements might be worse since those cables typically aren't impedance controlled. 

    -Bobby

  • 0 in reply to BOBBY
    Intellectual 350 points

    Thanks a lot Bobby, this was verry helpful!

    Max