TL16C550 UART Problems

[[email protected]]

Hello,
I'm having a horrific problem communicating with TI's TL16C550 UART.
I'm using the UART and an RS232 (MAC3221) transceiver to communicate
with my PC over a serial cable. I've successfully used this exact
architecture in 3 previous designs. I'm using the exact same DSP
software that I used in my previous designs. I've scoped all of the
control signals of the UART and they're all doing what they should be
doing. I've written all of the internal registers of the UART and
read them back and they're all set to what they were set to in the
previous designs. But.... I'm still not getting anything on the SOUT
pin. It just sits high and doesn't budge. I have a baud rate signal
coming out of the UART, but I cannot get anything on SOUT. I tried
switching to a brand new board and ran my code and I got something out
on SOUT for a short time (30-60 seconds) and then nothing ever again.
Any ideas would be greatly appreciated. I'm stuck and just about out
of ideas. Please throw any ideas out there and I'll give it a try.

Thanks,
Dale

 

[mpm]

Hello,
I'm having a horrific problem communicating with TI's TL16C550 UART.
I'm using the UART and an RS232 (MAC3221) transceiver to communicate
with my PC over a serial cable. �I've successfully used this exact
architecture in 3 previous designs. �I'm using the exact same DSP
software that I used in my previous designs. �I've scoped all of the
control signals of the UART and they're all doing what they should be
doing. �I've written all of the internal registers of the UART and
read them back and they're all set to what they were set to in the
previous designs. �But.... I'm still not getting anything on the SOUT
pin. �It just sits high and doesn't budge. �I have a baudrate signal
coming out of the UART, but I cannot get anything on SOUT. �I tried
switching to a brand new board and ran my code and I got something out
on SOUT for a short time (30-60 seconds) and then nothing ever again.
Any ideas would be greatly appreciated. �I'm stuck and just aboutout
of ideas. �Please throw any ideas out there and I'll give it a try.

Thanks,
Dale

IIRC, there were lots of 550 "flavors" out there.
This device has plenty of Reset and Enables (Chip selects, etc..)
I'm wondering if there's a problem with those (or the PCB layout?)
allowing you to still read/write but inhibiting actual throughput.

I would get the datasheet handy, and then work through all of the
control pins and determine pass/fail for each.
Maybe you'll get lucky?

If this is a legacy product, that you know works, maybe you just have
a bad batch of chips, or have something weird going on with the power
supplies, (or something else unrelated to the 550's themselves).
Just a thought.
-mpm

 

[[email protected]]

Thanks for your input. You make two good points. I think there's a
very good chance that we had a bad batch of chips. We got a new batch
and it seems to be working for the time being.
Your last paragraph is also a good possibility.
As of now we're getting data out of the UART and now the RS232 chip is
not responding. So, we're still not getting data out to the PC.

 

[[email protected]]

To me, the fact that you got one to work for 30 seconds and
then it failed indicates that either the IC's are no good or you're
doing something that blows it out... such as maybe a bad PCB
layout that makes some sort of unintended connection.

Jean

www.pcb-now.com