DP83640

Hello Wesley,

I have a few questions to get more insight about the system;

Can you send the values of Register 0x00, 0x01 and 0x10?
Is the PHY working in MII or RMII mode? Is it strapped?
What is the strap resistor value used on RX_ER (FX_EN_Z) pin?

-Regards,
Aniruddha

Hi Aniruddha,

We have the strap resistor at 2.2K (R46 on the development board)for RX_ER per datasheet specification. we also have jumpers on PMDIO_MII, PMDC_MII, MII_MODE, RMII_MAS as well as 3v3_USB. We are operating in RMII mode.

I tried to add some images to help you further, but it isn't working.
I asked our software guy if he did anything to the registers such as changing values etc. His replay is as follows...


No. Whatever the standard PIC32 83640 driver does at initialization. The RMII bit is set in the Harmony configurator, since we have it wired to use RMII (otherwise the copper wouldn’t be able to pass data either). That fact should also be mentioned. We:

1) Used default TI 83640 eval board config to test copper using RMII, since that it how PIC32 Ethernet Starter Kit is wired when using it’s daughtercard
2) Can successfully pass data.
3) Strapped hardware for fiber mode, full duplex according to eval board datasheet (list exactly what things are set up for fiber on the 83640 Eval board)
4) Wired up optical transceiver using this circuit (provide circuit diagram, maybe a JPEG picture?)

5) Link LED is on solid, activity blinks, no response (pings always fail)
6) When hooked to media converter board (copper to fiber), can pass data
7) IF we go back to copper, can successfully pass data again.

Hi Wesley,

When you open the 'Reply' window, please click on the 'Use Rich formatting' at the bottom right. This will open a larger reply window with more text formatting option. Try pasting the pictures there. If this does not work, we can take the conversation offline on email. Block diagrams and circuit schematics will surely help to better undestand the system.

When I was asking for the register values, I meant to ask for the values that you read when you have a succesful data transmission and when you don't have data transmission. In other words, please read the values of the above mentioned registers in both functioning system and non-functioning system. My apologies for being unclear.

Can you check if the PIC has internal resistor on the pins connected to the PHY? If yes, what is the nature of the pull resistor (pull-up or pull-down) and what is the value? Sometimes the MAC has strong internal resistors which can cause the PHY to be strapped in a undesired mode.

Also, are you power cycling the PHY when changing from copper to fiber interface?

-Regards,

Aniruddha

Hi Aniruddha,

I have posted some images that might be helpful to you below. To recap-

Note** we removed the Daughter board that contained the default PHY for the PIC32 and we placed solid core wire in the places of the needed pins and connected them to jumper wires to make the connection between the PIC and the PHY dev board.

1).first we tested communication between the desktop and the PIC via copper in Fig A (the PHY development board came set up for copper by default)

2). we then pinged the PIC 32 from the desktop and received replies from the PIC.This verified that everything was functional

3). Strapped the PHY dev board for Fiber mode by instructions in the datasheet

4). Shorted components in fig C (shown by red lines) and left resistor values open so we could populate on an external breadboard incase resistor values needed to be changed.

5).after populating the bread board the PHY and transceiver were connected to the breadboard to complete the communication path between the desktop computer and the PIC 32.

6). Then we attempted to ping the PIC 32 again with the optical fiber setup and were not successful

After this we noticed that only the “link” led was on and not blinking and the “activity” light was blinking. If we pull the fibers out of the transceiver, both lights turn off. We checked the SD signal out of the transceiver and got a “high” output around 2.5V when the fibers are plugged into the transceiver. Once the fibers are pulled out of the transceiver the signal goes low with an output a little over 1V. We also tried changing the schematic between the PHY and the transceiver per the transceiver example circuit in its datasheet (part number AFBR-59F1Z. I looked through the PHY’s datasheet and checked the jumpers we had in place with the “forced mode” table on page 10. We are running in 100BASE-Fx, Full-Duplex mode which requires FX_EN_Z=0 (which is pulled low via the 2.2k resistor) and AN0=1 (which is pulled high through the jumper to the right of the indicator LEDS). AN_EN and AN1 are “don’t care” values according to the datasheet. This is where we are now stuck because everything is properly connected but is non-functional. we have tried a combination of shutting off the pic and resetting the PHY without any luck.

I checked to see if the PIC had any resistor pins connected to the PHY and it seems that it does not.

Fig A

                                                                                                            

Fig B

 

Fig C

The Pictures do not seem to be showing up after I post, is it possible to send them to you via email?

Thanks,

Wes

Hello Wesley,

Yes, we can take this offline. I have sent you a 'Connect' request over E2E. Please email me directly.

-Regards,
Aniruddha

Hi Aniruddha,

I sent you a message about a week ago with all of the info as to what we have tested, have you received it?

Hi Wesley,

My apologies, I was expecting an email. I can see your message on E2E. I will reply to you shortly.

-Regards,
Aniruddha