#140 APF28 v0.E datasheet error (PWM3/PWM4 reversed)

[brett.muir]

Hi, I believe there is an error in the APF28 v0.E datasheet on the table on the Connectors table.

Refer to J1 pins 45 and pins 46. The schematic indicates that pin 45 connects to processor ball G7 and pin 46 connects to ball G6, which I believe to be correct.

According to the processor datasheet, the functions available on ball:
G7 should be Pin 46 / SAIF0_MCLK / PWM_3 / AUART4_CTS / GPIO_3_20
G6 should be Pin 45 / SAIF0_LRCLK / PWM_4 / AUART4_RTS / GPIO_3_21

However the datasheet indicates that the following functions are available:
Pin 46 / SAIF0_MCLK / PWM_4 / AUART4_RTS / GPIO_3_20
Pin 45 / SAIF0_LRCLK / PWM_3 / AUART4_CTS / GPIO_3_21

ie, it has PWM_3 and PWM_4 reversed (ie, assigned to the wrong pins) and AUART4_CTS and AUART4_RTS reversed.

I was able to confirm my theory that these were incorrect by configuring PINID_SAIF0_MCLK as PIN_FUN2 and activating PWM3. I could then see that signal on my APF28DEV board on J9, pin 8. If the datasheet were correct, it would have shown at pin 7 (which it did not).

Note that there is a note on the schematic at J9 pin 8 reading PWM3 / AUART4_CTS / GPIO3_20. This is correct.

Regards
Brett

[jorasse]

Thx Brett for your feedback,

datasheet apf28 V0.E page 20 - here is the correct specification:
pin 46 SAIF0_LRCLK PWM_4 AUART4_RTS GPIO_3_21
pin 45 SAIF0_MCLK PWM_3 AUART4_CTS GPIO_3_20

datasheet apf28dev V0.C page 17 - here is the correct specification:
pin 7 SAIF0_LRCLK PWM_4 AUART4_RTS GPIO_3_21
pin 8 SAIF0_MCLK PWM_3 AUART4_CTS GPIO_3_20

Fix already applied on the APF28dev wiki page
Datasheets will be updated accordingly for the next revision of the document.

BR
Eric

[brett.muir]

Hi Eric, thanks for your prompt response.

I am not certain that the following is quite right:
> pin 46 SAIF0_LRCLK PWM_4 AUART4_RTS GPIO_3_21
> pin 45 SAIF0_MCLK PWM_3 AUART4_CTS GPIO_3_20

I think that the pin numbers are actually the other way around in this statement (ie, 45 and 46 are back to front). I believe that my testing indicates this. Would you mind checking which way around these pins should be?

My testing is as follows:
- I set pin SAIF0_MCLK to have funtionality PWM3
- I can see that signal on APF28DEV, J9, pin8.
- I believe that this confirms that {APF28DEV, J9, pin8} connects to {iMX286 ball G7} via {J1 pin 45}, not 46.
- This is also consistent with the "G7" ball assignment (page 1 of the APF28 schematic), connecting to J1 pin 45 (page 5 of the APF28 schematic)

I agree that the pin functionalities are now grouped together correctly.

Let me know if you still believe that your pin assignments are correct, or if you made a typographical error?

Regards
Brett

[jorasse]

Hi Brett,

could you please clarify this pin issue?
It seems we are fully agree with the same description: G7 ball (SAIF0_MCLK/PWM_3) is wired to apf28/J1 pin 45 then to apf28dev/J9 pin 8 according to the apf28 schematics (pages 1, 5) and apf28dev schematics (page 2, 7).
Right?

Best regards,
Eric

[brett.muir]

Hi Eric,

Apologies, yes we do agree. I was confused by a typographical error in my original bug report:
> G7 should be Pin 46 / SAIF0_MCLK / PWM_3 / AUART4_CTS / GPIO_3_20
> G6 should be Pin 45 / SAIF0_LRCLK / PWM_4 / AUART4_RTS / GPIO_3_21

In that statement, 45 and 46 are reversed!

So I agree with your correction 100%, sorry about the confusion.

Regards
Brett

[Nicolas]

A new datasheet (version 1.2) is available on the Armadeus Systems Web site:
http://www.armadeus.com/_downloads/apf28/documentation/datasheet_apf28.pdf
Thanks for the feedback !

Have a nice day
Nicolas