Re: 16C550C UART Auto Flow Control

[Sergey <skis...@gmail.com>]

Wayne,

Given that 16C550 UART in DIP packages doesn't implement auto flow
control properly, do you still want me to modify Zeta SBC to support
switching between RTS/CTS and DSR/DTR?

Since the Zeta SBC is mostly geared toward beginner builders, I'd like
to keep it as simple as possible.

Thanks,
Sergey


On Jun 16, 4:42 pm, Wayne Warthen <wwar...@gmail.com> wrote:
> So, I got my Multifunction/PIC card built today.  I don't have the PIC
> components stuffed, but the serial and parallel circuits are done.  I used
> a TL16C550C in a PLCC package for the UART so I could compare it's behavior
> with my existing TL16C550C in the DIP package.  According to the datasheet,
> it should be identical.  As I suspected, it was not!
>
> Took me about 30 seconds to determine that the TL16C550C in PLCC package is
> working exactly as the datasheet says it should with respect
> to automatic flow control.  The "bit" to turn this feature on reads "on"
> after being set and I was able to tell from my RS-232 breakout box that the
> chip is automatically controlling RTS.  RTS is turned on as the buffer is
> filled and turns back off as the buffer is emptied.  Absolutely as it
> should.
>
> I am now convinced that the TL16C550C in the DIP package is simply not
> conforming to the datasheet.  Even though the datasheet seems to indicate
> that the functionality is identical in any package, it lies.  I am betting
> that the DIP package was close to being phased out and the TL16C550C in the
> DIP package is really just the TL16C550B.
>
> This is unfortunate because it means that it is not possible to implement
> automatic flow control for all of the DIP package UARTs which are prevalent
> in the N8VEM designs.  Sigh.  At least, I can generally prove that it is
> not a code issue.
>
> If anyone thinks there is a different possible interpretation of these
> results, please do let me know.
>
> I do have another experiment that I am waiting on parts for...  There is a
> SC16C550B chip (from NXP) with a DIP package that is pin compatible with
> TL16C550's.  They are very hard to find and I have a few coming from
> UTSource so it will be a couple weeks before I can try it.  In theory, this
> chip supports auto flow control.  Unfortunately, it uses a different
> register convention for enabling it, but I can work around that easily
> enough.  If it actually works, it may be an option for some.
>
> Thanks to everyone that has been so helpful with this!
>
> Wayne