problems related to ST7538

yongfiona · ‎2004-02-24

Posted on February 24, 2004 at 09:23

yongfiona · ‎2004-02-19

Posted on February 20, 2004 at 05:40

hi! i'm using ST7538 for DC power line communication and i am facing a lot of problems. i would like to know why the crystal oscillator isnt working at all. there is nothing at Xin and Xout when i use the CRO to see the waveform and also i cant see anything from the pins ATOP1 and ATOP2. i am using the circuit shown in the application note AN1714 (page 11). can i use it on DC power line???

jatin · ‎2004-02-23

Posted on February 23, 2004 at 09:00

I am not able to find AN17 Could you exactly tell the AN number? Which ST7 device are you using????

sjo · ‎2004-02-23

Posted on February 23, 2004 at 13:05

AN1714:

http://us.st.com/stonline/books/pdf/docs/9546.pdf

The only thing I can suggest is double check your layout/voltages etc.

Is the crystal grounded ok ?

Regards

sjo

yongfiona · ‎2004-02-23

Posted on February 24, 2004 at 03:18

acutally the application note states that it's not possible to use a probe to see the waveform at xin and xout. then how can i know that it's working?? i'm using AT90s2313 as the mcu and the mcu keeps sending 8bit signals to the Tx pin of St7538. i can see the 8-bit signals at the Tx pin and according to the datasheet, there shoudl be a sinusoidal waveform at the pins Atop1 and atop2 if asynchronous mode is selected(in fact asychronuous is the default mode). however what i can see at atop1 is a 5V signal and there is nothing at atop2. i dont know why and what to do next.

sjo · ‎2004-02-23

Posted on February 24, 2004 at 05:54

Have you tried writing to the control register, remember the watchdog needs refreshing by your micro - this can be turned off in the control register.

Do you get a waveform on pin 11, this will be clock/4 and is enabled by default.

Also check RSTO and PG pins to make sure 7538 is resetting ok.

Regards

sjo

dix · ‎2004-02-24

Posted on February 24, 2004 at 09:23

hi fiona,

I didn't respond before, as I hadn't picked up your e-mail. I have replied now, but I hadn't looked at this message string.

Okay if your trying to use Asynchronous mode, then you have a problem. The Control Register in the 7538 can only be configured in synchronous mode. (see page 13 of the 7538 datasheet.)

So you need the CLR/T clock line from the SPI master. Remember that in this application the Modem is the master device in the SPI comms system.

Why are you using asynchronous mode any way? In this mode the host MCU has to recover the clock ref. in Rx mode and control the bit times in Tx mode. Its far simpler to use synchronous mode and let the modem recover / generate the SPI clock.

If you're trying to work in asychronous mode then you probably haven't configured the 7538's control register. As you just can't do it without the CLR/T clock.

How are you trying to look at the ATOP1 and ATOP2 signals. The chip is a single supply device, with a differential sinewave output. Because the of the single supply the buffer can't swing +/- with respect to ground. To overcome this limitation STM have applied a DC bias to the ATOP1 and ATOP2 outputs. Now if you're using a scope to look for the signals, then you need to remember that the scope ground is an earth. So if you simply connect the scope across the ATOP1 and ATOP2 lines you will short-circuit one of the outputs down to ground. This will force the chip into thermal shut-down, (as you're shunting current down the earth lead). You need to use the scope in differential mode, (i.e. CH1 + CH2 inverted) and no ground lead.

Regards

Steve.

[ This message was edited by: Steveboy on 24-02-2004 13:55 ]