CDCE62005EVM: Register values invalid or out of range

Hi Toby,

Let us check to see if we will see the same issue here. Stay tuned for an update. Have a nice weekend.

Hi Noel,

I am heading on annual leave soon and I was wondering if you might have an update Re. the status of the GUI or board.

Regards,

Toby.

Hi Toby,

Sorry for late reply.
Unfortunately, we have no luck to duplicate the problem, so we don't have a cue what the root cause/solution could be.

Maybe they can try uninstall everything and then use clean disk software to remove relevant registry in the PC. Install the GUI again, see if that works.

Hi Noel,

Since my initial post I have tried using the EVM and GUI on multiple PCs with no success.

I now beleive there to be something faulty with the EVM. As all I have done with the EVM is plug it into a computer via the supplied USB cable and play around with the "Frequency Planner" on the GUI before the EVM started operating incorrectly, this indicates to me that the EVM was supplied in an unsuitable condition.

What would the next steps be? Would it be possible to send the EVM off for testing to confirm this? Or would it be possible to get a replacement EVM sent? I would like to get on with my testing using this peice of equipment as soon as possible.

Regards,

Toby.

Hi Toby,

When you were using different PCs to connect to the EVM, did you get three green lights from the GUI? In the EVM, did you get any LED on?

Hi Noel,

As before, the same issue occurs with the registers.

I get all three LEDs on the GUI and only a single LED on the underside of the EVM. As shown below:

Regards,

Toby.

Hi Toby,

LED D26 is a 3.3V power indicator, it proved that your board has power.

Let's share my board's characteristic, hope it can help you debug.

First of all, I have JP_3_3 selected USB Power. (use USB power instead of external power)

- Plug in USB cable.

> D26 is ON

> in the PC, Device Manager recognized the board. (see picture at the bottom)

> I measured a 12MHz, 1.7Vpp clock on the right hand side (close to the TI logo) of the 12MHz crystal.

- Open GUI

> D25 (DEVICE COMM) is ON. 

> I got output clock from the CDCE62005

> I got the same message asking me to load register data from EVM

- I click No. 

> D33 (PLL LOCK) is ON.

> I got three green lights in the GUI

- From the menu bar, Tools --> Default

- From the EVM Status box, click Write to load all the registers

Hi Noel,

Using solely USB power I have followed your process:

- Plugging in the USB cable D26 turns ON

- The PC recognises the board and is shown in the Device Manager as per your image

- I measure a 12 MHz signal from the right hand side of the 12 MHz crystal, but the Vpp is 460 mV

- I open the GUI but D25 does not turn on

- I get no clock output from the CDCE62005

- I click no on the load register da5ta from the EVM BUT D33 does not turn on

- I get all three green lights in the GUI

- I click on Tools --> Go Default, but nothing appears to happen

- I click on Write in the EVM Status box, but again nothing appears to happen

At no point do I have a signal produced by the EVM from any of the output channels.

Is the low voltage of the 12 MHz crystal output an important issue?

Regards,

Toby.

Hi Toby,

Can you try external power supply and observe the current? You can check / uncheck "Enable EVM power" and see if there's change in supply current. Also enable / disable an output buffer and observe current change. If the current doesn't change with output buffer enabled / disabled then the chip is not being configured at all. You can check if SPI is working properly. Otherwise the PLL may fail to lock due to faulty reference or configuration. You can check if there's anything abnormal about the crystal other than the 460mV Vpp. It is possible that lack of input power can cause lock failure.

Regards,
Hao

Hi Hao,

I have tried running the board with external power, but no change.

With an external power of 3.3 V the current is 0.57 A. The GUI does not allow me to check/uncheck  "Enable EVM Power" and there is therefore no change in current. I am not able to enable/disable an output buffer, they all read as disabled. I get all three lights on the GUI on including the "SPI Bus Status". I cannot lock the PLL. The crystal now outputs 12 MHz at Vpp of 500 mV.

Also I am now getting a 1.12 Vpp at 90 MHz signal from Output 1 and a 320 mVpp at 120 MHz from Output 0 despite when clicking "Read" in the EVM Status the GUI showing that all outputs are disabled. There is no output from AUX Out.

This board has not been functioning for some time now and it is important for my research that it works properly soon. What are the next steps from this?

Regards,

Toby

Hi Toby,

It is not clear if the SPI is working properly. Have you attempted to monitor the SPI bus?

I wonder if you installed the correct driver. This is what i see in the Device Manager. Does it match your system?

Kind regards,
Lane

Hi Lane,

What I see in device manager matches what you've shown.

How would you go about monitoring the SPI bus?

Regards,

Toby.

Hi Toby,

Thanks for sharing. It appears that you have the correct driver selected. Nonetheless, I will try to locate the correct driver and EVM firmware, I'll get back to you if I find anything.

Based on the information you have provided, I am not certain whether SPI is functioning correctly or at all. Does the SPI appear to be functioning differently when you are using external power vs. USB power? Do register reads, register writes, or soft button presses appear to function properly/differently for either mode?

As a debug step, you could try monitoring the SPI bus with oscillocope probes on R138, R129, R112, R111 on the EVM to confirm the SPI Bus Timings adhere to datasheet Table 6.6.

Kind regards,
Lane

Hi Lane,

Sorry for the late reply, I have been absent from work.

None of the register read or writes or any other function based on communication to the EVM work,  whether or not it is run with USB or external power.

I get the message:

When monitoring the SPI bus as you described, independent to USB or external power, all I get is noise. No clock signal is measured, and therefore neither are the other two signals.

Regards,

Toby.