EVM430-F6736: Communication loss - EMDC Stability

Update ! ******************************************************************

Test 7:
Following the success of the previous test, I removed the external power source, kept the TTL converter, but this time powered the system directly from the mains and also connected the load.
I launched Serial Port Utility and started receiving data. However, after about a minute, the transmission stopped.
I noticed that the TTL, which usually blinked when sending data, had stopped blinking.
This led me to suspect that the mains power supply could be the source of the problem.

Test 8:
To verify this hypothesis, I replaced the mains power with an external DC power supply, swapped the TTL for the HID, and launched the test—not via Serial Port Utility, but using the EMDC tool instead.
My reasoning was: if there were no data interruptions this time, it would confirm that neither EMDC, HID, nor the PC were responsible for the issue.
Therefore, the root cause would be the instability of the mains power supply.
The test lasted  without interruption, till I manually stopped it, after 35 minutes, which seems to confirm the hypothesis.

It is now necessary to further investigate the specific impact of mains power instability on the EVM430.

But I'd like to ask you if you already have an explanation for the problem

Hi Fathi,

Let me take a look into your description and check any recommendation here.

B.R.

Sal

Hi Fathi,

So, according to your test, the main power will imapct the stability between the EVM430 and PC.

Fathi Chelly said:

I launched Serial Port Utility and started receiving data. However, after about a minute, the transmission stopped.
I noticed that the TTL, which usually blinked when sending data, had stopped blinking.

When it suddenly stop, what is the last data sent by EVM430. I want to figure out whether the device reset, or just some noise make the communication interruptted.

Fathi Chelly said:

, I replaced the mains power with an external DC power supply

So, the main power is for the EVM430 correct? Changes it from AC generated 3V3 to external DC?

As for the communication stop, what I think is MCU reset, or the noise make the communication data wrong, maybe you can capture the signal on the RX/TX to see any difference here.

B.R.

Sal

Here are the latest data received:

You said :So, the main power is for the EVM430 correct? Changes it from AC generated 3V3 to external DC?

Yes exactly ! I changed the jumper JP4 (from 1-2 to 2-3 as mentioned in the photo you uploaded) and I powered the EVM430 from JP3 pins (Which are used to connect an external power supply). 

btw I made another test. I used the stable external power supply again, but this time I plugged the load (because in the last successful test I tested the system without connecting anything to the Line and Neutral pins) .

I did the test to make sure that my last deduction is right. And the result was confirming that! The system worked well without any problems for 30mn till I decided to stop it myslef. Which again means that the problem is due to the power supply.

Here I'd love to ask a question: Could the instability of my country's mains be the reason? Infact, the Nominal voltage should be 230V but I measured,using a multimeter, the voltage during different phases of the day in different places, I found out the the voltage variates between values from 220V to 240V. The signal isn't pure sinusoidal as even the third harmonic is present in the mains.

Might that cause a problem or the board designers should have previously taken care of that?

Hi Fathi,

Fathi Chelly said:

Here are the latest data received:

Looks like the a whole package sent is finished. What I suspect is that the MCU reset.

Fathi Chelly said:

I then tried to send the following command:
0x55, 0xAA, 0x06, 0x04, 0x01, 0x01, 0x01, 0x07, 0x00
to prompt the EVM to send me data.

After it suspend, then if you resend the command you list here, will it re-send the information again?

Fathi Chelly said:

Might that cause a problem or the board designers should have previously taken care of that?

Please do check its voltage on PL_VCC, and compared with EXT_VCC, see what difference here.

Actually this design is quite old, so i am not sure on the hardware design part. We are developing new version based on MSPM0 platform and maintain its source code and hardware design.

You can check below for the new version: https://www.ti.com/tool/TIDA-010243

B.R.

Sal

However, You inspired me to test PL_VCC values and Vsupply values, when powering the system, not from external source, but from the PL_VCC.

(By Vsupply i mean the Probe point that does exist in the supply circuit before the Buck converter as mentioned here:)

I measured the voltage when the EVM430 was sending data and when it stopped I remeasured again and the result was quite interesting:

When the system worked well:

* PL_VCC = 3.358V and it was stable.

* Vsupply = 37-38 V and it was stable.

However When the system stopped sending data:

* PL_VCC was fluctuating.When I placed the probe it kept showing me different values : 0.9V then it gets up to 3.3 then suddently drops down to 1.4V then to another value and so on and so fourth...

So this might explain why the MCU stopped working. So I asked myself the following question : Is there a problem maybe with the buck converter ? If I measure the V supply and it is stable then the Buck is the cause of the problem.

I put the probe and it turned out that :

* Vsupply was also fluctuating sometimes 14V, sometimes 20V, sometimes 16V and so on and so fourth..., fluctuating each second to different values.

Thus, the Buck converter isn't the problem.

At this time, I activated AC mode in the multimeter and measured the voltage between the line and Neutral pins of the EVM (A.K.A measuring the mains).

It was stable showing 242V.

Hi Fathi,

See below design document: https://www.ti.com/lit/an/slva491/slva491.pdf 

It limited the LINE max current, so what is the input current? If it is larger than expected, which might make the circuit not work;

B.R.

Sal