WLAN - Pic 18F46K80 - Flowcode V5 - Resolved

[Sefi]

Firstly apologies for this appearing in the V4 section but I currently do not have access to posting new messages in the V5 section.

I have tried to use the WLAN code that Benj has very kindly provided. This works fine with various chips that I have tested such as the 16F877A, 18F4431 but there is an issue with the 18F46K80 chip which is the actual chip I need to use for my project.

The problem is that the page that is served by the pic to the WIZ610 chip does not appear on the browser. I have various checks in place that show that the page request is seen by the PIC inside the main loop and it apparently responds correctly but no data appears on the browser.

I have attached the file. Can anybody help me please.

many thanks
Sefi

WLAN-18F46K80.fcf

(19.62 KiB) Downloaded 310 times

[Benj]

Hello Sefi,

Have you done things like a 1 second LED flasher test to confirm your settings like your clock speed and configuration are working as expected on the hardware? If this is not flashing at the correct rate then the baud rate for the serial communications channel will be incorrect and would explain why you aren't getting any data on your browser.

[Sefi]

Hi Ben

Thank for your reply. I have checked the timings with a single LED and it appears to be functioning correctly. In fact at the beginning of the flowcode chart you will notice that I toggle an LED just to show that the microprocessor is working. I have changed the on and off times of the LED to 100ms on and 900ms off to aid with the timing check of the LED flashing.

This appears not to be the problem. The WIZ610 chip appears to initialise correctly. Once initialised it appears as a normal wifi access point. I check for page requests in the flowcode chart inside a loop. I check for the return status and it appears to see the page request. But, nothing appears in the browser. I also check to see the connection status between the PIC and the WIZ610 and that also appears as "online".

Any other thoughts would be greatly appreciated.

many thanks
Sefi

[Benj]

Hi Sefi,

Which version of v5 are you using? It's strange that most devices are working but the 18F46K80 has issues.

The device's UART must be receiving data of some kind or the startup loop would never end.

Have you got a packet sniffer or a scope you can use to check the UART signal data?

I have a 18F26K80 here but no 3V3 board to run it on so finding it hard to try to replicate the problem.

Maybe worth having a go with a RS232 component and streaming data to a known working device. This will confirm if the baud rate calculation holds true for the K80 devices.

[Sefi]

Hi Ben,

I am running Flowcode V5.2.

I have tried the program using all Matrix hardware today, (up until today, I was using my own design hardware as I didnt have an 18F46K80 in a suitable package for the EB006), So an EB006, EB069 etc. Exactly the same thing happens. The EB069 initialises. Creates an Access Point which can be accessed in the normal manner. The "Page requests" are seen by the PIC but still nothing appears on the browser.

I also used an RS232 component and connected to the WIZ610 via Telnet. With this setup I had 2 way comms with the PIC. So it appears that the baud rates etc are accurate but using the WLAN component it appears one way.

Did I understand correctly from your previous reply that you dont have access to an EB069? If that is the case I am happy to send you mine for testing.

many thanks
Sefi

[Sefi]

Hi Ben,

Just to let you know that the issue has now been resolved.

The problem was that the Trigger and Reset pins should have been 1 & 3 respectively. These are the pins that are connected to the WIZ610 chip on the E069. In the FCF file that I had these were set to 0 & 1 and I assumed they were the right settings.

What does still surprise me is that the program was actually getting through the initialization routine, even with incorrect Trigger and Reset pin assignments. Anyway it is all sorted now and thank you for all the time you spent on this. While writing I should also say that the time for initialisation is now lightning fast where as before it would have taken about a minute or so for the WIZ610 to come up.

Many thanks
Sefi

[Benj]

Hello Sefi,

That's great news glad you got it running correctly and thanks for letting us know how you did it.