DRA821U: J7200

Hi,

May I know is there any update?

Thanks and regards,

Cherry

Cherry,

In the PADCONFIG, there are bits that are prepended with DS (e.g. DS_PULLTYPE_SEL or DSOUT_DIS). These bits control the pad when in deep-sleep mode. There are equivalent bits that control the pad when NOT in deep-sleep mode (e.g. PULLTYPESEL or TX_DIS to select the complementary bits).

In Deep-Sleep mode, the output driver is weakened so that may account for your current not being as expected. 

One thing you might try is this:

Clear DSOUT_DIS = 0

Set DSOUT_VAL = 1

and this is just for testing -- Set Force_DS_EN = 1

This last bit is an override for this pad, and it would make sure that you are really in Deep-Sleep at least from the perspective of this pad.

-2-

0x0004_0000 --> receiver is active, transmit is active if the muxmode has set it, There is no defined pull enabled

0x0006_0000 --> receiver is active, transmit is active if the muxmode has set it, There is no defined pull enabled 

0x1304_0000 --> receiver is active, transmit is NOT active, There is no defined pull enabled. 

0x0306_0000 --> receiver is active, transmit is NOT active, There is no defined pull enabled 

0x0D04_0000 --> receiver is active, output high and pull low.

0x0906_0000 --> receiver is active, output low and pull low

Kevin

Hi Kevin,

Thanks for your help.

1/0x0D04_0000 --> receiver is active, output high and pull low.

normal mode : receiver is active and pull low.

deepsleep mode : output high and pullup/pulldown is disable

In deepsleep mode this setting should be output high, they've tested his drive current is only 0.1ma,Indicates that this setting is incorrect

2/Set Force_DS_EN = 1   Setting this bit has no effect on the port;

They want to output high level in deepsleep mode, how to configure this padconfig?

Thanks and regards,

Cherry

Hi,

May I know is there any update?

Thanks and regards,

Cherry

Cherry,

C14 is WKUP_GPIO0_6 and is controlled by WKUP_PADCONFIG_54
0x0D04_0000 sets bits 27, 26, 24, and 18
bit 28 configures the deep sleep pull type; cleared it sets a pull-down
bit 27 configures the deep sleep pull enable; set it enables the pull (-down)
bit 26 sets the output value in deep sleep; set it enables a high value
bit 25 configures the pin to allow / disallow it to drive; cleared, the pin may drive an output in deep sleep
bit 24 sets the deep sleep enable for this pin qualified by deep sleep being properly entered in the rest of the device
-- notice that bits 28 - 24 are in contradiction driving high and pulling down in deep sleep does not make sense to me
bit 21 configures the transmit portion of the pin; cleared it allows the pin to drive
bit 18 configures the pin to receive; set the pin will be in a receive mode when not in deep sleep
bits 3:0 configure the mux mode; cleared to all 0s, the mode is WKUP_UART0_CTSn

Can you try the following:
-1- 0x04048000 -- again forcing deep sleep but removing the conflict between drive high and pull low.
-2- 0x00040007 -- not deep sleep; drive the pin with the GPIO0
WKUP_GPIO0_GPIO_DIR01 &= ~0x40
WKUP_GPIO0_GPIO_SET_DATA01 |= 0x40
read value on WKUP_GPIO0_GPIO_IN_DATA01

I have run through these options in the attached gel file. On my board, I measure a valid HIGH in both Deep Sleep and functional mode.

Kevinrandom_stuff.gel

Hi Kevin,

They've tried, but it didn't work regarding the config of 0x04048000. Still pullup the input.

Testing: port is shorted to ground after sleep with a 10k resistor, test current is 0.1ma, assuming port state is input pull up.

Thanks and regards,

Cherry

Cherry,

I gave you a second setting as well -- what current do you measure there? The reason for my question is that a 3.3V signal through 10K should be about 0.3mA. You are not off by a huge amount. 

Kevin

Hi Kevin,

-2- 0x00040007 -- not deep sleep; drive the pin with the GPIO0
WKUP_GPIO0_GPIO_DIR01 &= ~0x40
WKUP_GPIO0_GPIO_SET_DATA01 |= 0x40
read value on WKUP_GPIO0_GPIO_IN_DATA01

-Before sleep:C14

output 3.3v;current is 0.33ma

-After sleep:

output 3.3v;current is 0.1ma

0x00040007 This configuration is useless.

The customer's suspect is as follows:

Why is 0.1ma：The pull-up resistor for the input pull-up is 23k，3.3v/(23k+10k)=0.1ma

Thanks and regards,

Cherry

Cherry,

Thanks for running the experiment. I will look for an answer to the question.

Kevin

Hi Kevin,

Thanks and expect the response.

Cherry

In Deep Sleep, the driver becomes a pull-up through 20K. The resulting 100uA is exactly right.

Obviously, a 20K pull up will not be able to pull a 10K resistor to ground to a logic rail.

The results may not meet the needs of the customer, but they are consistent with what is occurring in the device.

Kevin

Hello Kevin,

Thank you so much.

So does that mean there is no way to output high but only pull up? 

Thanks and regards,

Cherry

I am not sure... The only way to possibly work around this is to NOT put this pin into Deep Sleep.

Kevin

Hi Kevin,

Thanks. So may I know how to not put this pin into enter deep sleep? The customer want to know the configuration parameters of this pin. Could you help with this?

Thanks and regards,

Cherry