Doubts about power supply of EVAL-AD5780 board when operated by a microcontroller

Hi Sapta ,

Our Bipolar DAC expert Justlap is currently checking this query,

Best Regards,

Den

Hello,

Please power VDD (analog supply) before powering VCC (digital supply). This is the recommended way of powering up the AD5780.

If this is a problem, then add a Schottky diode between VDD and VCC as shown in the figure below.

Regards

HI Justlap, Thanks for responding. I am indeed familiar with the paragraph from the AD5780 datasheet. However, as I mentioned, if anything remains connected at VCC and DGND, the chip doesn't work i.e. across V_out_Buff, you don't get any response regardless of what you send over SPI. When you disconnect with VCC and DGND and keep it floating, the chip starts operating as usual. I have not tried with the Schottky diode. I will post the updates next week. 

But my original question remains the same. Why the chip works only with analog supply but not with digital supply?

The AD57809 needs of both supplies, powering them up in sequence, first VDD and then VCC. If my understanding is correct, that's the way you operate the AD5780 without any problem.

If only VDD is supplied to the part and is still working, that mode of operation must not be used since it can stress unnecessarily internal circuit, current is flowing from VDD to VCC internally instead of being supplied externally via VCC pin.

Hi! Sorry for the delayed reply. I understand the standard mode of operation and my original post was about that only. The board doesn't work in that standard mode of operation. Let me elaborate with this example (Simply sending 0 and 262143 with a delay i.e. square pulse) and pictures. Let's Assume Case 1: VDD/VSS supplied with +-15V and VCC is floating. It works Perfectly. Case 2:  VDD/VSS supplied with +-15V and VCC is +5V supplied from the microcontroller, Does not work. Case 3: Disconnect the VCC from +5V supply works again. Connect VCC with a separate power source (Analog Discovery 2) with +5V. Doesn't work again. See the configuration of the Links and the board in the attached picture. 

With this, please let me know how to address the situation. With the standard mode of operation, the board doesn't work.
(Sorry your pic upload resolution limit is quite bad. Please find the images at this Photo Link)

Two questions:

Q1: When the AD5780 is powered from the microcontroller or the Analog Discovery, can you please measure the supplies VCC and VDD?

Q2: Is is possible to capture (using the Analog Discovery analog channels and not the digital ones) the SPI interface, for example clock and SDIN?

Hello! Yes. For both of these, I have uploaded the pictures in the same Google Photos folder as before. There is a  comment on the lower right where you can find relevant descriptions. I think it shows that indeed the data is passing over but the chip is not responding when both supplies are connected. 

And I have the solution to the problem. I will post it in a fresh reply for any future reader.

The solution to this issue is to connect 3.3 V to the digital supply of the board. The instruction in the user manual on Rev. A | Page 3 of 20 is wrong. It mentions "5 V between the VCC and DGND inputs for the digital supply of the AD5780". This is wrong. If you go above ~3.5/3.6 or below ~2.6/2.7 in digital supply VCC, The circuit will stop working, even if the analog supply VDD&VSS are connected.  

On the same note and on the same page, in the Connector J3 Pin Descriptions, Pin 5 is mentioned as DGND. It actually labeled as IOVCC on the board. Leave it disconnected. If you give power on it, the circuit stops working. 
Justlap  I suggest a update for the user manual of the evaluation board for cases when a SDP board is not used.