I am evaluating Wiz820io module for my companies project with NXP microcontroller. I program W5200 using SPI CLK @24Mhz. The system work fine and I so far I have am using 6 sockets for various TCP and UDP client and servers.
However I am experiancing below reliability issues with the Wiz820io/W5200.
1.) W5200 reset using the SW Reset (bit 7 of MR) simply does not work. the only option is to use nRESET pin.
2.) After few hours (not constant) of working perfactly the w5200 stops responding to PING requests. Also I cannot do any socket commuication after that. I beleive the ICMP and ARP are handled in hardware so as long as the W5200 has SHAR,SIPR, SUBR and GWAY registered programmed, it should respond to ping requests.
To confirm that my software/microcontroller is not doing anything wrong, I did a simple test
- Reset W5200 using nRESET
- Programme SHAR,SIPR, SUBR and GWAY with valid values
After the above test steps I can ping the W5200 but again after few hours it stops responding to ping.
Can you please help ? My project is suppose to be a high reliable network remote control/monitor system.
Let me know your data code on W5200 chip, purchase date and dealer.
Data code is 4 digits number on the bottom of the w5200 chip.
How many WIZ820io modules did you test? all of the modules are same?
I think the W5200 chip damaged by unknown problem.
(The ping reply is implemented in the H/W logic. If you set the valid MAC / IP address, you should be able to get the ping reply.)
If possible, please RMA request to distributor ASAP.
note that, W5200 SW reset operation needs defined delay time in datasheet. please check again.
The data code on wiznet chip is 1243.
Purchase date is 09 January 2014
Dealer = TME POLAND
I have only tested 1 module because I only have one. Ok I will contact the distributor for the replacement. Also I will give sw reset another try.
As hkjung said, you should be request RMA to distributor.
I bought in february 2014 three wiz820io and I have the same described problem.
I used the last arduino library, a 2560 Mega arduino with a separate 3.3V power supply for wiz820io module and a common ground with arduino.
Each time I install a new module on my testboard, power on, reset the arduino and the program start.
The ping test is OK. My application is running perfecly.
After a few hours all the lights on wiz820io are off and the module seems to be frozen.
After it’s very difficult to restart the 820io and the module seems to fail at power on.
All the module are R1.0 and the w5200 chips are labeled HP2785 1213. The dealer is Lextronic in France.
Can you take a picture of the bottom side of WIZ820io and post it onto this forum?
I may give an advice to you with that picture.
Please find hereafter a picture of the Wiz820io module.
Thanks for help.
Yes, you have an old version of WIZ820io as I worried.
We, WIZnet, already announced PCN of WIZ820io in Feb., 2013 and replaced old ones which our distributors and customers have with new ones.
But it seemed that there were some modules which TME, your distributor, missed to replace.
You can contact to TME and request to replace it with new one.
If you look at the bottom of WIZ820io, you can find which one is new or not.
On new one, we changed two ceramic capacitors to two tantalum.
I attached a document explaining the change.
Please refer to it and contact to TME to fix your problem.
I hope you to solve your problem as soon as possible.
Improvement of WIZ820IO.pdf (76.3 KB)
Thanks for your information, I will contact my french distributor to replace my modules.
I’m not sure if I should be reviving this thread or starting a new one, but I am having a similar problem. I am using a Teensy 3.1, Wiz820 and SD card adapter board and Wiz820io, all unmodified. With USB power and an Ethernet connection, I tapped the Ethernet connector of the Wiz820io with scissors and UDP traffic from the Teensy stopped. Also, I am unable to ping the board. I am able to repeat this fairly regularly, but not every time.
I noticed this bug while running the system for a long period of time - I believe I have seen traffic stop up without touching it, but I suppose it could be a different problem.
The Wiz820io board I used has the tantalum caps, and the board is R1.0. Is there any new errata with the Wiz820io?
We do not have an adequate workaround for this bug - the system must stay online and send data rapidly, and a half-second or more of delay would be noticeable to the user.