FP-1600 Communication Errors

[Prashant]

I have three "banks" of modules, each "bank" consists of a FP-1600 ethernet module, followed by eight other i/o modules. Of the three "banks", the first bank seems to lose communciation after a couple hours. It transmits and receives more rapidly, upon startup, than the other two banks, then begins to deteriorate with a series of 7 status error flashes, and finally it experiences total loss of transmission to the module. Even though the devices are all powered communication is lost. I have upgraded the firmware to 411, I have isolated the ethernet cables (for EMI), and I have replaced more than two FP-1600 modules on this particular bank (using suggested values for connection to each module). The bank of modules, consists of the followi
ng I/O's in order.... AO-200, AO-200, AO-210, AO-200, AI-111, AI-110, AI-111, RLY-420. The relay terminals are connected to a number of solenoids. This bank is the only bank that I lose communication with. Is there any other suggestions, than what I have already done? These modules are all linked to the host via a switch.

[Ben]

Hi Prashant,

Provided your relay module is not sharing power with the logic of the 1600, you should give NI support a call.

I have heard some nasty rumors about the 1600 dropping off-line. I have a 1600 on a brand new issolated LAN segment that does the same thing you described. From what I understand, the 1600 does not recover from the "7-blink" comm error without power cycling.

I will keep you posted re:what I find if you do the same.

Ben

[steven@ni]

Please call in to National Instruments and talk with an Applications Engineer.

7 status flashes usually is unrecoverable.

[Prashant]

HI Ben,

I have contacted NI support regarding this problem, and they have suggested electro-magnetic interference. I am in the process of isolating the module away from all other power sources, but question this experiment, because I have no other transmission loss on the other two network modules. I have tried a different Cat 5 cable, and also have switched terminals on the ethernet switch. I am running out of options, and NI is limited in their responses to my problem.

[swag]

Prashant

I was having difficulties with FP-1600 communication errors as well. In the end, I had to upgrade to FP-2000. Before you come to that conclusion, I would add error trapping to your program to see what sort of error you are getting. After each FP Read, I unbundle the error out (unbundle by name.vi) to check status. If status is true I use the general error handler.vi with the type of dialog set to 0 (to prevent a message box dialog) to record the error code. In my system, I was receiving error 33163. Using this method I could recover from the error. Upgrading to the FP-2000 did not eliminate this error completely, but made it managable with error trapping and recovery code built into the program. Again, this may not be the same problem you are dealing with,
but I wanted to post what I found as others have had similar problems.

[Ben]

Exchange the 1600 that is failing with one of the other 1600's.

If the probelem follows the 1600 then scream that the unit is bad and you want a new one.

If the 1600 works fine in the new location but the previously good 1600 starts to fail, then you have to keep looking.

Ben

[jhylton]

NIs suggestion that EMI is responsible is intolerable. FieldPoint is supposed to be an "industrial", distributed I/O system. If it won't work in and industrial loacation - that is typically high in EMI then it shouldn't be advertised/sold as such.

BTW, I have had virtually no problems with any of the FP1000/1001s that I've been using over 5+ years. I've had nothing but problems with the FP1600s.

[Prashant]

I have had no problems with the 1000's and 1001's either. I contacted NI, and they told me they just released a new 1601 module, because of problems customers were experiencing on the 1600's. Temporarily, I have placed a 2010 module on that bank, and have successfully run for 3 straight days.

[Green_Rives]

A small percentage of our customers have reported modules that give them 7 status flashes. If you are receiving this problem, please call technical support. We now have a resolution to this problem and they should be able to get you a new module that does not have this issue.

Also, as Prashant stated, we just released the FP-1601. However, this module was not built due to the problems customers were experiencing on the FP-1600s. It was in response to customers requesting additional features on the FP-1600.

This module is based off of the FP-2000 architecture. It has the same functionality as the FP-1600, but with several added benefits. It has better immunity to electromagnetic noise, a more efficient power supply a
nd processor (4.5 Watts vs 7 Watts), a wider temperature range (-25 to 55 C), 7 user accessible DIP switches, 1 reset DIP switch located on the front of the module and user controllable LEDs.

Green Rives
National Instruments
FieldPoint Product Support Engineer