TPS8802EVM: How to EVM interface with MSPM0L1306 launchpad using dual smoke ray example test

Pratik Panchal

Part Number: TPS8802EVM
Other Parts Discussed in Thread: MSPM0L1306, , TPS8802, SYSCONFIG

Tool/software:

Hello TI team,

I am currently using the MSPM0L1306 -Launchpad along with the TPS8802EVM (Evaluation Module for the TPS8802 Residential Photoelectric Smoke/CO Alarm) sensor. My goal is to configure the TPS8802EVM to read real-time smoke and CO data via I2C using the dual ray smoke detector example.

as per given instruction in TI documentation, I did below MSPM0L1306 - launchpad interfacing with TPS8802EVM.

So, my Question here is,
when I activate the P14 switch on the launchpad, it triggers the alarm and plots the Infrared LED and Reflection Delta graph. However, I am not observing the expected variations in the Infrared LED, Blue LED, and Reflection Delta, as outlined in the Application Note of Dual-Ray Smoke Detector with the TPS8802 and MSPM0 MCUs [DEMO GUI - Page-10].

Could you please guide me on how to properly interface the MSPM0L1306 Launchpad with the TPS8802EVM to achieve the expected output for smoke on the GUI graph plot? Additionally, I would appreciate guidance on CO detection as well.

Best Regards,

Pratik Panchal

Team Lead - Embedded Firmware Engineer

Mantra Smart Identity Private Limited

(A Subsidiary of Mantra Softech India Pvt Ltd)

B -1003, Shapath Hexa, S. G. Highway, Sola, Ahmedabad -380060, India

Email: pratik.panchal@mantraidentity.com Web: https://www.mantraidentity.com/

9 months ago

0 David Stout 9 months ago

TI__Prodigy 690 points

The example code you are using from the SDK was developed for use with a different board which is a TI internal design. It was not meant to be used with the TPS8802EVM. While we have not attempted to use the example code and launchpad with the TPS8802EVM, it should be possible to replicate the correct signaling and connections to get measurement results in the DEMO GUI as you are attempting, albeit maybe with some component changes.

Just a few notes on the connections you have shown: 1) You have LEDEN connected to 3.3V, LEDEN should be connected to PA.26 and used as a GPIO for control of the Blue and IR LED's for smoke measurements. 2) VBAT input to the TPS8802_EVM should be the input power source for the system. The supply pin for the MCU should be connected to VMCU on the EVM. The 5V and 3.3V jumpers between the emulator and DUT side of the launchpad need to be open. 3) You have AMUX connected to PA.25. This analog input is used by the temp sensor on the demo board which the TPS8802_EVM does not have. PA.16 is used for the AMUX signal for Photo and CO reads in the firmware. CSA and CSB are two additional ADC inputs used in the code as feedback for setting the LED drive currents. 4) Vref+ and Vref- are hardwired to the demo board through the launchpad connector which need to be manually connected in your setup for correct ADC reads.

While the above is just a start of the corrections needed based on what you've shown, there are other differences in the hardware that need to be accounted for in order for what you are trying to will actually work.

Are you using an optical chamber of any kind with the TPS8802_EVM? By default, the LED's and photodiode are mounted pointing straight up. One major issue with trying to measure smoke with this is that there is no scattering angle for the light from the smoke particles to either reflect or scatter forward onto the Photodiode. Secondly, without a chamber, the photodiode will be saturated by ambient light.

I would suggest contacting your local sales support and direct them to reach out to me directly if this is something you would like to pursue further. If this is not possible, let me know and I can find out how to provide my direct contact information.

Regards,

David

0 Pratik Panchal 9 months ago in reply to David Stout

Prodigy 40 points

Hi David,

Thank you again for your detailed guidance. Please find my consolidated responses and queries below:

1. Yes, we are aware that the example code was developed for a different internal TI board and not specifically for the TPS8802EVM.
2. We have already routed LEDEN to PA.26 and configure it as a GPIO for Blue/IR LED control.
3.We're currently powering the EVM by providing 3.3V to VBAT via the launchpad and Jumper is connected on VBAT side for J3 connector on TPS8802EVM. 
  Since Vref+ and Vref- are internally connected through J3, no manual connection is required in our setup, please clarify?

4. You mentioned PA.16 is used for AMUX in the firmware, but in the dual-ray example code, PA.16 doesn’t appear to be configured as an ADC pin — could you clarify?


Also, could you please confirm which pins are used for CSA and CSB in the example code for LED current feedback?

5. During the smoke and CO testing, we did not observe any logs like debug, warning and error messages on the UART. 
Is there anything we need to do on MSPM0 launchpad's UART pin, or do we need to enable a specific configuration in the Dual Ray smoke example?

6. For initial validation, we’re using a black box setup placed over the diodes to minimize ambient light. For future tests, 
we plan to use a 3-diode cavity enclosure (3D printed) from TIDA-010941.

Please share your direct contact information on my earlier shared email ID or let me know the best way to continue this discussion in more detail.

Best regards,
Pratik Panchal

0 Pratik Panchal 9 months ago in reply to Pratik Panchal

Prodigy 40 points

Hello David,

Just a quick follow-up—could you please respond to this when you get a chance?

Thanks & Regards,
Pratik

+1 David Stout 8 months ago in reply to Pratik Panchal

TI__Prodigy 690 points

Pratik,

I have sent a complete list for configuration of the EVM/Launchpad to match the demo board setup described in the app note previously referenced. For completeness, please find my answers to your questions below. As we are discussing this topic directly outside of this forum, please close this thread, as I will not be sharing any additional information for this setup in the forum.

Regards,

David

3.We're currently powering the EVM by providing 3.3V to VBAT via the launchpad and Jumper is connected on VBAT side for J3 connector on TPS8802EVM. 
  Since Vref+ and Vref- are internally connected through J3, no manual connection is required in our setup, please clarify?

Vref+ and Vref- are the connections for an external ADC reference for the M0L.  This is how the firmware is configured.  We chose to use the external reference instead
of the MCU VDD supply because the AMUX on the TPS8802 uses the VINT rail which is 2.3V nominal.  Using Vint as the ADC Vref maximizes the dynamic range of the CO and
Smoke readings.

4. You mentioned PA.16 is used for AMUX in the firmware, but in the dual-ray example code, PA.16 doesn’t appear to be configured as an ADC pin — could you clarify?

The firmware was ported to M0L from a previous MSP430 project which had no support for Sysconfig. As such only one ADC input was used to configure the ADC peripheral in sysconfig for the demo code.


Also, could you please confirm which pins are used for CSA and CSB in the example code for LED current feedback?

Please refer to the information I email you and your team directly

5. During the smoke and CO testing, we did not observe any logs like debug, warning and error messages on the UART. 
Is there anything we need to do on MSPM0 launchpad's UART pin, or do we need to enable a specific configuration in the Dual Ray smoke example?

Referring to Figure 4-1 and 4-2 in the app note, you should see at the bottom of the screen a linked indication, The COM port connected
along with Baud rate followed by "Hardware Connected".  If you don't see this information, the board is not connected to the PC / GUI for display.
If you are connected, there should be data being displayed.  It is possible that without all the correct connections to the launchpad
that the firmware could be getting hung waiting for signals that are not present.  Please refer to the information I provided by
email and try this again.

6. For initial validation, we’re using a black box setup placed over the diodes to minimize ambient light. For future tests, 
we plan to use a 3-diode cavity enclosure (3D printed) from TIDA-010941.

I discussed this in my email reply. While blocking the ambient light, your scattering angle is still 0 degrees.  This will not work for
detecting scattered light due to smoke.  A proper chamber which does both (eg., blocking ambient light and holding the diodes at the correct
angles) is needed.

Sensors

Sensors forum

TPS8802EVM: How to EVM interface with MSPM0L1306 launchpad using dual smoke ray example test