AM3352 TSC

Hi Sithara,

What software are you using? What do you mean by idle state?

Dear Biser,

Linux os is used. I mean the idle step configuration of TSC module...

Is the TSC driver initialized?

Yes

If you look at Figure 12-2 in the AM335X TRM you will see some signals that control the TSC switches (YPPSW, XNPSW, XPPSW, WPNSW, YNNSW, YPNSW, XNNSW). The IDLECONFIG Register has bitfields that can control these switches. Opposing switches on the same AIN line must never be turned on at the same time.

If XP is high at IDLE step, hope it is possible to change it to low and make YP high?

Also, what is the state of HHV(VDDA domain) is the same figure?

Any idea?

uhtis said:

If XP is high at IDLE step, hope it is possible to change it to low and make YP high?

You can see from the figure that there is no low-side switch for XP. I don't have information about HHV.

I don't understand what is the purpose of these questions. Can you explain please?

It is mentioned that the TSC switches can be configured. But whatever state we have configured, it depends on the state of the control signal, HHV, which is unknown. The NAND and AND gates in the figure 12-2 ic controlled by this signal. If HHV is fixed then how does the changes we do in software gets reflected?

I didn't say HHV is fixed. I said I don't have information about this signal - where it comes from and how it's controlled.

We have changed the SW register values and probed the touch signals . Whatever changes we have made in IDLE state, the level is not getting changed. ???

:(

The HHV signal is high while PWRONRSTn is held low.  The purpose of this signal is insure the FETs driving the AIN signals do not turn on until TSC is reset.

The pen-up/pen-down interrupt which is generated from the AIN0 input while operating in 4-wire mode assumes a fixed polarity so can not change the idle state.

Regards,
Paul

I am not clear about the TSC module working. So i was trying to understand the behaviour of TSC module by probing the 4-wire signals in the idle state. 

1. We have used HW synchronized, one shot mode in every step configuration registers. Is it possible to control the FET switches in this mode?
2. Usually in 4-wire touchscreen, first we apply supply in YU, ground the YD and read the y coordinates from XR (XL is floating). Then we apply supply in XR, ground XL and read the X coordinates from YU (YD is floating). Could you please explain how the logic works in AM3352 TSC?

The configuration values for X readout is 0x40192 and Y readout is 0x100072 in the driver. With this i have traced back the working. Hope it is as per the normal touch screen working. 

Please reply to query 1?

Anyone please reply to query 1????

We appreciate your patience, our apps team will be replying to your post soon.

My reply below is from the hardware perspective.  I cannot comment how this is done in various software drivers.

The FETs are controlled by their respective bits in each STEPCONFIG register.

Bit 11 is used to control the VSSA FET connected to AIN4.
Bit 10 is used to control the VSSA FET connected to AIN2.
Bit 9 is used to control the VDDA FET connected to AIN1.
Bit 8 is used to control the VSSA FET connected to AIN3.
Bit 7 is used to control the VDDA FET connected to AIN2.
Bit 6 is used to control the VSSA FET connected to AIN1.
Bit 5 is used to control the VDDA FET connected to AIN0.

When set to '1', the respective FET is on.
When set to '0', the respective FET is off.

Regards,
Paul

Dear All,

Thanks for your response.

In the patch from TI (with TS_CHARGE_STEPCONFIG same as IDLE_CONFIG) , the values was..

X readout ---->0x40192

Y readout----->0x100072

Z1 ------>0x400D2

Z2 ------>0x1C00D2

The Z1 and Z2 values seems to be wrong in the patch. Please have a look into that.

We have changed the Z1 and Z2 readout step register values as 0x100132 and Z2 as 0x80132. 

Now we have to check the pressure threshold value and the ADC delay values (open and sample). 

How to set the threshold value ?

Now while we do the draw test, we miss some of the coordinates. During this, the pressure data is getting zero. 

So i doubt the pressure data [Z] calculation. How is it implemented in the code from the Z1 and Z2 readings?

Anyone please let me know the pressure threshold calculation?

uhtis said:

Dear All,

Thanks for your response.

In the patch from TI (with TS_CHARGE_STEPCONFIG same as IDLE_CONFIG) , the values was..

X readout ---->0x40192

Y readout----->0x100072

Z1 ------>0x400D2

Z2 ------>0x1C00D2

The Z1 and Z2 values seems to be wrong in the patch. Please have a look into that.

We have changed the Z1 and Z2 readout step register values as 0x100132 and Z2 as 0x80132. 

Now we have to check the pressure threshold value and the ADC delay values (open and sample). 

How to set the threshold value ?

Can you be more specific about "the patch from TI"?  Are you referring to the patch set for SDK 7.00 that I posted here:

http://e2e.ti.com/support/arm/sitara_arm/f/791/p/319613/1298838.aspx#1298838

What Linux SDK are you using and which patch are you referring to?

In any case, my recommendation would be to get SDK 7.00 and apply the patch from my post above.  I spent two weeks going through every bit and every calculation to try and get everything perfect.  I know of at least one customer that has successfully integrated the patches.

My recollection is that the Z calculation may have been wrong in SDK 6.00.  I'm pretty sure that was one of the items updated in the base SDK 7.00 release. 

Oops -- I pointed to my original patch set instead of the v2 patch set.  Here's the right link (further down in the same thread):

http://e2e.ti.com/support/arm/sitara_arm/f/791/p/319613/1308705.aspx#1308705

Im using the V2 patch set for SDK 7.00.

But the Z1 and Z2 values of 0x400D2 and 0x1C00D2 seems to be wrong. Could you please confirm?

uhtis said:

But the Z1 and Z2 values of 0x400D2 and 0x1C00D2 seems to be wrong. Could you please confirm?

Those are correct.  Do you have your hardware hooked up properly:

• AIN0 = X+
• AIN1 = X-
• AIN2 = Y+
• AIN3 = Y-

At some point the device tree introduced configurability of these inputs, but that should not have been done.  This configuration is required by the hardware.  You must use the configuration above in conjunction with ti,wire-config = <0x00 0x11 0x22 0x33> in the device tree file.

Hardware connections are same as in starter kit. 

AIN0 = XL

AIN1 = XR

AIN2 = YU

AIN3 = YD

Hope this is correct as per your patch. Please confirm.

Yes that is fine. 

thanks Brad..

My pressure data is getting 0 sometimes. This needs to be verified . Hope this can be corrected by setting the delay values.