MTQN-MNG1-B02 Powering Down After 20 Seconds

[cshea@updesigns.com]

I am working with an MTQN-MNG1-B02. I have it plugged into MTUDK-ST development board. I have have all of the jumpers on JP98 of the MTUDK-ST unpopulated. I am skywiring the UART signals from a NUCLEO-L476RG development board. I also am controlling the Reset signal to the MTQN-MNGQ-B02 from the Nucleo board.

On the Nucleo board, I have an application that starts up, resets the Dragonfly and starts off by issuing some AT commands. After 20 seconds, almost precisely, the Ublox modem stops communicating and the UART transmit line that the modem drives becomes an undriven signal. It’s as if the modem has powered down or gone into a deep sleep.

I can find no explanation for why this would happen. I changed my application to first query the modem to verify that the Power Saving Mode is disabled (using the +CPSMS command). The response indicated that mode was equal to 0, which means that Power Saving is, indeed, disabled. So, the only conclusion that I can think of is that the 1.8V power supply on the Ublox modem is being turned off.

On the MTQN-MNG1-B01 (this is the version with the Cortex M4 populated), two of the pins on the Cortex M4 are connected to RADIO_RESET and RADIO_ONOFF. These nets are connected to the Ublox modem and allow the application on the Cortex M4 to enable/disable power to the modem and to reset the modem.

ON the MTQN-MNG1-B02 (the version without the Cortex M4 and the version I am using), the Reset pin on the Ublox modem appears to be directly connected to the Reset pin on the 40-pin interface header. However, the RADIO_ONOFF signal is not accessible via the interface header.

When looking at the signals RADIO_ONOFF and RADIO_RESET (pins PE4 and PF3 of the Cortex M4 footprint, respectively) I see that the RADIO_RESET pin stays high for the entirety of the 20 seconds that the Ublox modem is on. However, the RADIO_ONOFF signal turns on for only the first 400 ms of operation. Afterwards, it stays low. I then connected PC5 of the Cortex M4 footprint to my scope. This pin is defined in the -01 version as the 1.8V monitor. However, this signal was up at 3.3V for the entirety of the 20 seconds.

I have a couple of questions:

1) Has anybody experienced this type of behavior and, if so, what was the cause?

2) Is there any difference in the board layout between the -01 and -02 versions of the Dragonfly? I ask this because I am assuming that they are identical apart from the population of the Cortex M4 processor. If they are, then I can continue to use the BGA pads to gain access to the Ublox signals of interest, specifically, the RADIO_ONOFF and RADIO_RESET signals. I have some doubt that this assumption is true because of the fact that I saw 3.3V on the signal that I expected 1.8V on.

Thanks!

[Leon Lindenfelser]

On the MTQN-MNG1-B02, the RADIO_ONOFF and RADIO_RESET are controlled by an ATtiny9 microcontroller. You should not try to control these. The beginning of chapter 7 covers powering down and resetting the radio. It can be done by AT command or the reset line. The reset line is an input to the ATtiny9.
https://www.multitech.com/documents/publications/manuals/s000702.pdf
Note: Signal 1V8_MON operates at 3.3 V not 1.8V.

On the MTQN_MNG1-B01, RADIO_ONOFF and RADIO_RESET are controlled by the STM32L471QG microcontroller and the ATtiny9 is omitted.

[cshea@updesigns.com]

Thank you for the quick response. To clarify, I am not trying to control RADIO_ONOFF or RADIO_RESET. I was just monitoring them with a scope to see if they could indicate for me why the U-blox radio is not responding.

I was able to locate the ATtiny9 and verify the path between the reset line on the connector and ATtiny9. Not that I doubted it but was more interested than anything else.

Most importantly, I am still having the issue in which the modem only stays on for 20 seconds. Any thoughts on why this may be occurring? Power is good, there are no glitches on the reset line, a command is not being sent to power off the unit, and the unit does not have Power Saving Mode enabled. In fact, even if I just reset the unit and don’t issue any commands at all, the same behavior is observed.

[Leon Lindenfelser]

I took an MTQN-MNG1-B02 and measured the voltage on RADIO_RESET and RADIO_ONOFF. I am not seeing the same as you see. RADIO_RESET is low. FYI, both signals get inverted before they reach the radio. Please open a support portal ticket at https://support.multitech.com/support.

At the end of the 20s do you see 1V8_MON go low?

[cshea@updesigns.com]

Hi Leon,

First, thank you for the help. Your first response led me in the right direction. As I mentioned, I had the reset line connected to my Nucleo board. The line was configured as a general purpose output and was not being properly initialized to the logic level high. I didn’t realize that as an issue until you pointed out that there is an intermediate micro (ATtiny9) handling reset and power based on the reset signal on the interface header. When I read the section of the document that you pointed me to, I realized that I initiating a power down of the Ublox radio when holding the reset line for longer than 1 second. I updated my application to properly control the reset line and all is working properly. I apologize for the user error.

Thanks again!

Casey

[Leon Lindenfelser]

Hi Casey,

You’re welcome. I’m glad you were able to get it working.

Kind regards,
Leon

[Author]

Posts