• State TI Thinks Resolved
  • Locked Locked
  • Replies 14 replies
  • Answers 2 answers
  • Subscribers 34 subscribers
  • Views 303 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TL16C752C: XR16L2750IM-F Cross

Marco Rojas
Intellectual 2780 points
Part Number: TL16C752C
Other Parts Discussed in Thread: TL16C752D

Tool/software:

Hi team,

I have a customer with the below inquiry. Let me know your thoughts and thanks in advance!

"We are currently utilizing the Maxliner/Exar XR16L2750IM-F dual uart in one of our products.  I was looking to see if TI had a compatible part.  I did see the TL16C752 family.  Is there any difference between the TL16C752D and the TL16C752C other than errata fixes?  If so is there documentation explaining the differences and why one may be preferred over another?

Are either of these (or some other product) a cross for the XR16L2750IM-F?  Thanks"

Marco

  • 0
    TI__Guru 73271 points

    Hi Marco,

    I will try to get back to you before end of day tomorrow on this. 

    -Bobby

  • 0 in reply to BOBBY
    TI__Guru 73271 points
    Marco Rojas said:
    I did see the TL16C752 family.  Is there any difference between the TL16C752D and the TL16C752C other than errata fixes? 

    I looked at this from a datasheet side by side comparison but I didn't see anything that stuck out between D and C version. There is a minor difference in that the D version can support 48MHz at Vcc=3.3V but other than that almost everything is the same. I checked the register maps and other datasheet parameters. (minor difference in timing for that 3.3V case).

    Internally I checked and it seems like the dies are different though. The D version is considered the newer of the two, D is made in the 300mm wafer technology that is considered modern for TI whereas the C version uses an older wafer technology that TI has been trying to move away from. While there is no clear advantages between the two devices from a datasheet perspective, it's probably still better to try the D version since it's in the newer process technology.

    Marco Rojas said:
    Are either of these (or some other product) a cross for the XR16L2750IM-F? 

    I really only recommend looking at the 752D and 750E series of devices from our UART portfolio since they are in our newer process technology...

    I took a look at 752D and the XR device. The registers aren't 100% the same between the two and the XR device has an additional revision decode register. 

    I don't think you can reuse the software/firmware from the XR to the 752 family. If you move to the 752 device, you will likely need to create software to support it. 

    -Bobby

  • 0 in reply to BOBBY
    TI__Intellectual 2780 points

    Hey Bobby,

    Thanks for the reply! Unfortunately, this needs to be a drop-in/P2P for an existing product so I don't think these are viable options. Let me know if there's something I'm missing.

    Marco

  • 0 in reply to BOBBY
    TI__Intellectual 2780 points

    Hey Bobby,

    Quick follow up from the customer:

    "Took a further look at the TL16C752D and I see that it clearly indicates it has a sleep mode, but I did not see a specification for current in sleep mode. Did I miss it? Or does it not produce a power reduction? Thank you"

    Marco

  • 0 in reply to Marco Rojas
    TI__Guru 74282 points

    Marco,

    Thanks for the follow up question, Bobby is out of office at the moment but we should be able to get a response to you by end of business 01/31/2025 CST.

    Regards,

    Eric Hackett 

  • 0 in reply to Marco Rojas
    TI__Guru 73271 points

    Hi Marco,

    It looks like your customer is correct. The device does not actually specify the sleep mode current when it's enabled. I checked one of our later/newer UARTs that used a majority of it's IP from the D version and that one also fails to mention the current during sleep mode.

    This seems like a miss from the systems engineer who spec'd the device and since we don't have the validation data for the device. We likely don't have the data to provide how much current it's actually saving. I have a guess of how this mode saves current... It likely disconnects the XTAL input (clock source) which takes away the switching voltages. This will stop shoot through current seen by the CMOS inputs through the device which look at the CMOS inputs. So I do believe the mode does lower current consumption (most of it likely from the clock input signal) but we don't have numbers on how much that is.

    -Bobby

  • 0 in reply to BOBBY
    TI__Intellectual 2780 points

    Hi Bobby,

    See below for customer response:

    "Thanks for the response.  My appraisal is that the TI TL16C752DPFBR or the NXP SC16C752BIB48 will physically drop in for the XR16L2750IM-F, but both would require a FW tweak for the enhanced mode registers.  Looking at features that can be achieved by a FW tweak, they look to be equivalent to what is utilized, but the sleep mode feature on the TI part is only words. 

    The original part at 3.3V has a maximum sleep current specification of 15 micro-amps.  The possible NXP alternative also specifies the sleep mode current and has a maximum of 50 micro-amps.  Our application is very power sensitive and the uart is idle most of the time, but needs to be able to respond to communication activity.  The sleep mode is heavily used to minimize power consumption.  The only power consumption specification from TI is that it won’t exceed 16 mA, but that is a near worst case situation.   

    The initial preference is to go with TI, but without any information on the sleep current there is always the question of whether it was intentionally left off due to being non-competitive and not wanting to publish the number. I am sure that sleep mode will provide a significant power reduction, but even an order of magnitude reduction would leave it one or two orders of magnitude larger than the competitors.  Can TI provide a typical number or the results from a test unit to know what ball park the sleep current is in?"

    This seems to be a good opportunity for us to win! Let me know what you think

    Marco

  • 0 in reply to Marco Rojas
    TI__Intellectual 2780 points

    Any response here?

  • 0 in reply to Marco Rojas
    TI__Guru 73271 points

    Hi Marco,

    We don't have this data available. Typical measurements usually require 30x units to be tested across temperature and voltage levels. We don't have the bandwidth to get these measurements unfortunately as this device is older and lower volume to support the business need to do so.

    -Bobby

  • 0 in reply to BOBBY
    TI__Intellectual 2780 points

    Hi Bobby,

    Would we be able to get even one measurement from a unit sitting on an engineer's bench?

    Marco

  • 0 in reply to Marco Rojas
    TI__Intellectual 2780 points

    Would this be possible?

  • 0 in reply to Marco Rojas
    TI__Guru 73271 points

    Hi Marco,

    I will need to see if I can get my set up going. Since this device is digital I need to use a microcontroller to put it into sleep mode so it's not quick/easy to grab. 

    Not sure if I have the files for it. Can I get back to you by Friday? 

    -Bobby

  • 0 in reply to BOBBY
    TI__Intellectual 2780 points

    Hi Bobby,

    Friday works, customer said that's fine and would be appreciated

    Marco

  • 0 in reply to Marco Rojas
    TI__Guru 73271 points

    Hi Marco,

    I just finished doing testing but it looks like the device doesn't really save much power/supply current when I enter sleep mode. 

    I was getting about 9.6mA during idle conditions (No INTs, RX held high, IER = 0h10). It didn't change much before entering this mode (sometimes I saw current go up to 10.6mA. 

    From my testing it seems like the customer's concerns about the sleep mode are valid.

    -Bobby