1200 Baud needed on hp50g (Help please, JYA?)
dot
requires the use of 1200 baud rs-232. I have a serial cable for the
calc. The hp50g does not support 1200 baud, however. Is there any way
that 1200 baud can be included via a sysrpl program or a ROM patch?
Thank you.
-Dot-
Jean-Yves Avenard
They didn't provide 1200 bauds support on this chip
TW
Can't be done. I already asked him to try it last summer. He did,
spent most of a weekend trying it out, but it never worked. The
Samsung chip does not support 1200 baud. All the other baud rates are
done using the on chip controller.
Besides, he doesn't offer free support for HP anymore.
TW
dot
wrote:
Well, that's not the best news of the day. Thanks for the quick
response, anyway!
-Dot-
dot
Well, it seems like every step forward is followed by two back. I
suppose a uart speed shifter with a fifo memory would do the trick.
I'll have to see if there are any other tricks that I can use with the
equipment available.
-Dot-
TW
> suppose a uart speed shifter with a fifo memory would do the trick.
> I'll have to see if there are any other tricks that I can use with the
> equipment available.
You can always purchase one of our surveying cases/circuit boards.
Would set you back $300+ though. Too bad we aren't selling volumes
needed for mass production. :-)
TW
Al Borowski
Hi Jean-Yves,
Is there any reason why the RS232 pins couldn't be reconfigured for
GPIO, then bitbanged at such a slow speed?
cheers,
Al
Jean-Yves Avenard
> Hi Jean-Yves,
>
> Is there any reason why the RS232 pins couldn't be reconfigured for
> GPIO, then bitbanged at such a slow speed?
You would need to ask Kinpo.
As I have no will to investigate really :)
Al Borowski
Fair enough. I gave the datasheet a quick skim before, and I'm pretty
sure it's possible (the hardware serial port can be bypassed
completely). Dot/TW: How badly did you want this speed?
Cheers,
Al
jeff
We have a circuit in production (that will be installed in an e-case)
that will reduce 2400 to 1200 for those instruments that only
communicate at the lower rate. This should be available in 3 weeks or
so. Check stk4hp.com for availability.
Jeff
TW
Would be nice, but I already have it for my purposes. Our
communication board does the full range of baud rates, can provide
power to external devices, has selectable flow control, programmable
communication buffers, full NMEA GPS parsing, etc.
TW
mk
>
> Well, it seems like every step forward is followed by two back. I
> suppose a uart speed shifter with a fifo memory would do the trick.
> I'll have to see if there are any other tricks that I can use with the
> equipment available.
>
> -Dot-
I ended up buying a couple of these: http://www.bb-elec.com/product.asp?SKU=232BRC&Trail=3&TrailType=Main
($200 ea), when I needed to interface an optical pyrometer with a
PLC. I didn't realize when I ordered the pyrometer that it's baud
rate was fixed at 38400, yet the PLC's serial inface card had a max
baud of 19200. Luckily, the application didn't require high-speed
communication (it needed 1 reading per second), otherwise I might have
had to have been concerned about buffer overrun on the 19200 side.
Tim's solution sounds like a perfect fit.
mk
dot
like to find a way to get the hp50g to work at that baudrate.
There is a recent article in QST (Feb 2K7) which uses a bluetooth
modem from www.sparkfun.com called "BlueSmiRF" that I would like to
experiment with. Traditionally, it works in conjunction with a
bluetooth dongle for a pc. I want to try to use it with some data
acquisition equipment and the hp50g.
-Dot-
On 19 Apr 2007 02:45:12 -0700, Al Borowski <al.bo...@gmail.com>
wrote:
Claudio Lapilli
I just read the Samsung manual, and I don't see any reason why it
won't work at 1200 bauds, they actually have a clock division
completely configurable by the user, so the chip can even support non-
standard speeds.
I think it was more of a software decision by the Kinpo OS to provide
or not the speed.
{{ QUOTING FROM SAMSUNG MANUAL }}
Baud-Rate Generation
Each UART's baud-rate generator provides the serial clock for the
transmitter and the receiver. The source clock
for the baud-rate generator can be selected with the S3C2410X's
internal system clock or UCLK. In other words,
dividend is selectable by setting Clock Selection of UCONn. The baud-
rate clock is generated by dividing the
source clock (PCLK or UCLK) by 16 and a 16-bit divisor specified in
the UART baud-rate divisor register
(UBRDIVn). The UBRDIVn can be determined by the following expression:
UBRDIVn = (int)(PCLK/(bps x 16) ) -1
Where, the divisor should be from 1 to (2^16-1).
{{ END QUOTE }}
So, anyone using hpgcc could easily transmit at 1200 bauds. There's no
need to do it by software as Al suggested, altough that will also work
fine, since the pins can be configured as IO port H.
The standard PCLK (when running at 75 MHz) is 18.75 MHz, you get a
division value of 976, well within acceptable range. It's not an exact
division but it should be close enough to establish a communication.
Is there any reason I'm missing why it would not work?
Claudio
Al Borowski
> I just read the Samsung manual, and I don't see any reason why it
> won't work at 1200 bauds, they actually have a clock division
> completely configurable by the user, so the chip can even support non-
> standard speeds.
[...]
Whoops, you're completely right. Based on what Jean-Yves said, I
assumed that the clock division register wouldn't support such a low
baud rate - other microcontrollers I've used have had this issue.
Serves me right for making assumptions
Cheers,
Al
Jean-Yves Avenard
> Whoops, you're completely right. Based on what Jean-Yves said, I
> assumed that the clock division register wouldn't support such a low
> baud rate - other microcontrollers I've used have had this issue.
> Serves me right for making assumptions
tried to get 1200 bps in the early days, but removed it from the list of
supported baud rate after they had issue with the chip.
JY