Theory of using shield electrode

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
jade_4415151
Level 1
Level 1

Hi! Sorry for my english, I can be mistaken in the selection of terms and in terminology in principle, but I hope you will understand me.

I have some theoretical questions about using shield electrode in my project.

I have a slider with ground hatching *from user guide manual* 7 mil trace width and 70 mil width of cell on the back of PCB and 7 mil trace width and 45 mil width of cell in the front. And total area of hatching is 50 cm * 6.4 cm = 320 cm2 on the back and 50 cm * 1.4 cm = 70 cm2 on the front (if i'm right and if i don't confuse anything).

And here is the question: is 1 shield electrode enough to this area?

And another theoretical question: if 1 electrode is not enough and it needs to use 2 electrodes (to increase the power of the shielding signal, so to speak) - can they be closed in one circuit, because I have front and back hatching in one circuit or different shield electrodes must be used only in different circuits?

I asking because I use CY8CKIT-049-4200 and I do not have KitProg or something like that to use SenseTuner and debug at all, so I would like to know for sure what can be done and what cannot with it.

Thanks!

0 Likes
1 Solution
Hari
Moderator
Moderator
Moderator
750 replies posted 500 replies posted 250 solutions authored

Hi jade_4415151

One shield electrode would be sufficient. You must however, stitch the hatched surface at multiple places to avoid an increase in resistance due to the traces. This will also make sure that the shield signal is uniform throughout.

Also, the shield waveform needs to be checked (using an oscilloscope) to make sure that the shield is getting charged and discharged completely. If not, the sense clock frequency needs to be lowered to ensure the same.

You can also use the tank capacitor (Csh_tank, a 10 nF capacitor) if the parasitic capacitance of the shield is greater than 100 pF. In most cases this would not be required, however, it is better to check the parasitic capacitance of the shield.

You do not need a KitProg, you can use a MiniProg to use the CapSense Tuner as this is a handy tool for tuning CapSense sensors.

Best regards

Hari

View solution in original post

0 Likes
6 Replies
Hari
Moderator
Moderator
Moderator
750 replies posted 500 replies posted 250 solutions authored

Hi jade_4415151

One shield electrode would be sufficient. You must however, stitch the hatched surface at multiple places to avoid an increase in resistance due to the traces. This will also make sure that the shield signal is uniform throughout.

Also, the shield waveform needs to be checked (using an oscilloscope) to make sure that the shield is getting charged and discharged completely. If not, the sense clock frequency needs to be lowered to ensure the same.

You can also use the tank capacitor (Csh_tank, a 10 nF capacitor) if the parasitic capacitance of the shield is greater than 100 pF. In most cases this would not be required, however, it is better to check the parasitic capacitance of the shield.

You do not need a KitProg, you can use a MiniProg to use the CapSense Tuner as this is a handy tool for tuning CapSense sensors.

Best regards

Hari

0 Likes

Thank you for the answer!

It's completely suits me.

Also, the shield waveform needs to be checked (using an oscilloscope) to make sure that the shield is getting charged and discharged completely. If not, the sense clock frequency needs to be lowered to ensure the same.

You can also use the tank capacitor (Csh_tank, a 10 nF capacitor) if the parasitic capacitance of the shield is greater than 100 pF. In most cases this would not be required, however, it is better to check the parasitic capacitance of the shield.

I'll consider your prompts.

You do not need a KitProg, you can use a MiniProg to use the CapSense Tuner as this is a handy tool for tuning CapSense sensors.

I meant that if I do not have even a KitProg - I absolutely do not have and a MiniProg too, because MiniProg in 15 times more expensive, then most cheap solution with KitProg, so.. ))

And I have another question now, I cannot understand nothing from user manual.

What does Cmod do?

Manual says:

Ensure that the CMOD is not adjacent to any switching/communication pin. Since there is an analog signal on CMOD, it should be surrounded by the CapSense (analog) signal rather than being surrounded by the switching/communication (digital) signal. Ground of CMOD should have a least possible path to device ground.

The decoupling capacitors and CMOD capacitor must be connected as close as possible to the chip to keep impedance of the ground and supply traces as low as possible. 

CSD requires one external capacitor called CMOD for operation

But what it does - a totally cannot understand. Why and where it's need?

Thanks!

0 Likes
Hari
Moderator
Moderator
Moderator
750 replies posted 500 replies posted 250 solutions authored

Hi jade_4415151

Cmod is an external capacitor that is required for CapSense operation. For detecting a touch, the CapSense component detects the change in capacitance of the sensor. The CapSense sensors are charged and discharged periodically (depending on the sense clock frequency). The charge that is needed to charge the capacitor to Vref will be different in case the sensor is touched (since capacitance is different).

Initially, before each scan, the Cmod is charged to Vref. Then, once the scan begins, the sensors are charged and discharged from the Cmod capacitor (with the help of switches that switch between Cmod and ground). The block inside detects the small voltage variations in Cmod cap and this change is used to determine the capacitance of the sensor.

This is the reason an external cap is needed to CapSense design. It is also the reason for adding a limit on the trace length. An increased trace length will result in variation in potential from the cap and the IC and this may cause erroneous measurements.

pastedImage_1.png

Csd block diagram.

Best regards

Hari

0 Likes

AH_96 написал(а):

Hi jade_4415151

Cmod is an external capacitor that is required for CapSense operation. For detecting a touch, the CapSense component detects the change in capacitance of the sensor. The CapSense sensors are charged and discharged periodically (depending on the sense clock frequency). The charge that is needed to charge the capacitor to Vref will be different in case the sensor is touched (since capacitance is different).

Initially, before each scan, the Cmod is charged to Vref. Then, once the scan begins, the sensors are charged and discharged from the Cmod capacitor (with the help of switches that switch between Cmod and ground). The block inside detects the small voltage variations in Cmod cap and this change is used to determine the capacitance of the sensor.

This is the reason an external cap is needed to CapSense design. It is also the reason for adding a limit on the trace length. An increased trace length will result in variation in potential from the cap and the IC and this may cause erroneous measurements.

pastedImage_1.png

Csd block diagram.

Best regards

Hari

AH_96 написал(а):

Hi jade_4415151

Cmod is an external capacitor that is required for CapSense operation. For detecting a touch, the CapSense component detects the change in capacitance of the sensor. The CapSense sensors are charged and discharged periodically (depending on the sense clock frequency). The charge that is needed to charge the capacitor to Vref will be different in case the sensor is touched (since capacitance is different).

Initially, before each scan, the Cmod is charged to Vref. Then, once the scan begins, the sensors are charged and discharged from the Cmod capacitor (with the help of switches that switch between Cmod and ground). The block inside detects the small voltage variations in Cmod cap and this change is used to determine the capacitance of the sensor.

This is the reason an external cap is needed to CapSense design. It is also the reason for adding a limit on the trace length. An increased trace length will result in variation in potential from the cap and the IC and this may cause erroneous measurements.

pastedImage_1.png

Csd block diagram.

Best regards

Hari

Thanks for explain!

So.. Should 2,2nF capacitor be necessarily present in system or only in some cases?

Because now I use 30 CapSense pins + (maybe) will use Shield. trace length ~10 cm and everything now without that capacitor, seems work pretty well. So how should I understand this?

0 Likes
Hari
Moderator
Moderator
Moderator
750 replies posted 500 replies posted 250 solutions authored

Hi jade_4415151

So.. Should 2,2nF capacitor be necessarily present in system or only in some cases?

Because now I use 30 CapSense pins + (maybe) will use Shield. trace length ~10 cm and everything now without that capacitor, seems work pretty well. So how should I understand this?

-> The 2.2nF capacitor is mandatory for any CapSense design. If you are using a kit, it is possible that the kit already has a 2.2 nF capacitor populated in the designated pin.

For shield, a tank capacitor (Ctank) is required only if the parasitic capacitance (Cp) of the shield is greater than 100pF.

Please probe the sensors and shield pin and make sure that the capacitors are getting charged and discharged completely. You will observe a waveform similar to this:

pastedImage_5.png

Best regards

Hari

0 Likes

Thank you very much! Now I got it!

0 Likes