- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello,
I have a question about actual acquisition time for PSoC4100S SAR ADC.
Please see "PSoC® 4 SAR ADC Input Impedance – KBA224710".
PSoC® 4 SAR ADC Input Impedance – KBA224710
According to this,
RSW2+RSW1 ≈1/(sample rate * 6.4 pF)
If the clock frequency is 12MHz, the sample rate is 750000sps.
Then, RSW2+RSW1 =1/(750000*6.4*10^-12)=208.333kohm.
The acquisition time = 9 × (RSRC + RSW2 + RSW1) × CSHOLD.
Assuming RSRC is 0, Acquisition time = 18749nsec.
What do you think?
Best regards,
Yocchi
Solved! Go to Solution.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Yocchi san,
It is actually the input resistance seen from the source point where is signal source is connected.
Now the image you have attached is also not right. The resistor is similar to a switched capacitor resistance. Due to sampling sudden bursts of charges will be drawn into the system. If you take the average current flown over time, this can be modeled as a resistance connected.
Kindly check online references for switched cap resistances.
Best Regards,
Vasanth
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Yocchi-san,
Unfortunately your inference is not right. The input impedance of the ADC includes the RSW2+RSW1 + 1/(sample rate * 10 pF) ,the resistance due to the switched capacitor equivalent resistance formed by sampling circuit. As the third part is dominant, it defines the total input resistance of the circuit. It is calculated based on the periodic bursts of charge going into the capacitor.
Now regarding the acquisition time, the sampling switch is closed.
So acquisition time is the time to charge the internal hold capacitor with 9 time constants, when the switch is closed.
tACQ = 9 × (RSRC + RSW2 + RSW1) × CSHOLD should be used for calculation of the acquisition time.
Best Regards,
Vasanth
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Vasanth-san,
I know what you mean. I agree with you.
Then, What does "An approximate value of this resistance can be found using 1/(sample rate * 6.4 pF)" in KBA224710 mean?
Best regards,
Yocchi
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Yochi san,
Rinput = RSW2+RSW1+RSAMPLING .
RSW2+RSW1 = 2.2 K Ohms
RSAMPLING = 1/(sample rate * 10 pF).
Normally this will be the dominating factor in the resistance calculated. The KBA tries to tell that the acutal input impedance is not just RSW2+RSW1 ohms, but along with that the equivalent resistance due to sampling, RSAMPLING should be added. I hope that clears your doubt.
Best Regards,
Vasanth
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Vasanth-san,
"Rinput = RSW2+RSW1+RSAMPLING" means the total resistances that are seen from analog input pin, right?
Best regards,
Yocchi
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Yocchi san,
It is actually the input resistance seen from the source point where is signal source is connected.
Now the image you have attached is also not right. The resistor is similar to a switched capacitor resistance. Due to sampling sudden bursts of charges will be drawn into the system. If you take the average current flown over time, this can be modeled as a resistance connected.
Kindly check online references for switched cap resistances.
Best Regards,
Vasanth
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Vasanth-san,
I am sorry. Certainly my drawing was wrong.
The other side of SWACQ appears as a switched capacitor.
Rinput = RSW2+RSW1+RSAMPLING .
Best regards,
Yocchi
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Vasanth-san,
The CSHOLD is A_INCAP(10pF) in the datasheet, right?
Best regards,
Yocchi
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Yocchi san,
A_INCAP is 10pF. Thanks for pointing out. We will do corrective actions to reflect the changes both in KBA and thread.
Best Regards,
Vasanth
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Vasanth-san,
May I correct the equation like below?
RSAMPLING = 1/(sample rate * 10 pF)
Best regards,
Yocchi
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Yocchi san,
I have done necessary steps to update the equation in the KBA. The responses the thread are corrected accordingly.
The resistance is, RSAMPLING = 1/(sample rate * 10 pF)
Best Regards,
Vasanth
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello Vasanth-san,
I really appreciate your help.
Thank you very much.
Best regards,
Yocchi