Strange behaviour of HL7692

Hi,

I am just trying to get familiar with the HL7692 LTE-Modul. Before this, I worked with the HL6528RD and the HL8548. Both Modules worked to my full satisfcation. My application makes a data exhange with a HTTP server via POST .
Integrating the HL7692 into my project seems to be an easy game, due to the similiarity to the other HL-Series modules. But the HL7692 shows a strange behaviour: after the first connect and transmission of data it seems to ‘stall’ and ignores any AT command.

The CTS signal after the transmission is continously high (inactiv). Only after a reset, the module restarts working.

This is the log (a little bit shorted):

// Explanations
//
// ‘>’ sent to modem
// ‘<’ modem repsonse
//
// Some CRLF removed

// look for modem

AT

  • Time out +

AT

  • Time out +

AT

  • Time out +

// modem found

AT
< AT
OK

// echo off

ATE0
< ATE0
OK

// ask for modem type

ATI
< HL7692
OK

// explicit activation of hardware handshake

AT&K3
< OK

// ask for firmware version

AT+GMR
< AHL769x.2.10.154401.201609141850.x7120m_2
OK

// report errors numerical

AT+CMEE=1
< OK

// start with full funcionality

AT+CFUN=1
< OK

// Enter SIM PIN

AT+CPAS
AT+CPAS
AT+CPIN?
< +CPIN: SIM PIN
OK

  • Time out +

AT+CPIN=“5314”
< OK

AT+CSQ
< +CSQ: 99,99
OK

< +CSQ: 25,0
OK

// start registration

AT+CREG?
< +CREG: 0,0
OK

AT+CREG?
< +CREG: 0,0
OK

// succesfully registrated to network

AT+CREG?
< +CREG: 0,1
OK

AT+CNMI=2,0,0,0,1
< OK

// setup connection profile

AT+KCNXCFG=1,“GPRS”,“web.vodafone.de”,"","",“IPV4”,“0.0.0.0”,“139.7.30.125”,“0.0.0.0”
< OK

AT+KCNXTIMER=1,90,3,60
< OK

// increase buffer size

AT+KIPOPT=0,“HTTP”,1,1460
< OK

AT+KCNXPROFILE=1
< OK

// connect to network

AT+CGATT=1
< OK

// re-check field strength

AT+CSQ
< +CSQ: 23,99
OK

AT+CSQ
< +CSQ: 23,99
OK

// initiate http connection

AT+KHTTPCFG=1,“http://www.myserver.net”,80,0,"",""
< +KHTTPCFG: 1
OK

+KCNX_IND: 1,4,1

< +KHTTPCFG: 1
OK

+KCNX_IND: 1,4,1

+KCNX_IND: 1,1,0

// connect succesfull
+KHTTP_IND: 1,1

AT+KHTTPHEADER=1
< CONNECT

Accept: text/plain
Content-Type: application/x-www-form-urlencoded
Content-Length: 82
< OK

AT+KHTTPPOST=1,"/myurl/myphpscript.php",0
< CONNECT

< XMITRESULT=1000–EOF–Pattern–
OK

+KHTTP_IND: 1,3,82,200,“OK”

// anser from HTTP server: XMITRESULT=1000 -> success !
< XMITRESULT=1000–EOF–Pattern–
OK

+KHTTP_IND: 1,3,82,200,“OK”

// close connection

AT+KHTTPCLOSE=1
< OK

+KCNX_IND: 1,5,30

// delete HTTP profile

AT+KHTTPDEL=1
< OK

// next connect starts with field strength check, but modem does not respond
// and CTS is always ‘high’.

AT+CSQ

  • Time out +

AT+CSQ

  • Time out +

AT+CSQ

  • Time out +

AT+CSQ

— END Log —

Any idea, what’s going wrong ?

Regards,

moardy

Moardy,

HTTP commands are very high level and not really used that extensively by many people (not to say they should not be properly supported). First thing you need to push for/ask is for the latest firmware 2.10 is getting a little old now, we are up to 2.16 right now.

You will need to get hold of this through your disti/Sierra contact.

Regards

Matt

Matt,

thanks for your information. Yesterday I tried the BHL769x.2.15.170200.201701201640.x7120m_1
with same result. Perhaps the 2.16 will work a little bit (better) :slight_smile:

Regards,

Moardy

Moady,

I suspect not, ticket will need to be created by the distributor for us to recreate and fix.

Regards

Matt

Hi Matt,

thanks for this very important information. I sent it to my distributor.
Can you give me an idea of the time period, after such error is fixed ?

Regards,

Moardy

Moardy,

That is a very difficult question to answer, like all companies we are commercially driven so the larger the project the harder it can be pushed internally.

Regards

Matt

Hello,

thanks to technical support of my distributor, the problem is solved.
The reason was very simple:
The HL6528RD and the HL8548 have a disabled sleep mode of the USART by default:
But the HL7692 has an enabled sleep mode :unamused: !!!
The firmware used was the BHL769x.2.16.170400.201702221601.x7120m_1
I think, it would be very time-saving, to have a list of deviations
inside the HL-family - or better: no deviations.

Regards,

Moardy