DAC3482EVM: FIFO Loading for Spec Measurements

Thanks Jim!

No need for factory test data. I am tasked to do an environmental test of this part and measure any parametric changes.  Therefore other data may not be of much value.

Config45 looks to be the key. I'm still new to this EVM, so thanks for your patience.  I believe the following registers should be loaded with the following values:  

config45 (0x2D) bit 0 =1

config48 ((0x30) bits15:0 = sifdac(15:0) = desired data to the DACs (in offset binary, based on config2 bit 2)

I'm not getting any output (AC or DC) from the EVM.

Could you give me direction on how to get the EVM running without the TSW1400?

In SLAU432A, I followed section 3.3 & 3.4.

In the DAC3482EVM Software Control these values can be changed directly, as follows:

Tab "Input", the block "Input Format" and select

"Format": Offset Binary ... 

"Enable Constant Input" (selected) and

"Constant Input" = (enter a value between zero & 32767)

This updates the Registers & their values, as seen on the right side of the screen.

x02 x7000 (means input = offset binary

x2D x0005 (means config 45 has constant input from sif register)

x30 x7FFF (means config 48, sif register = 3FFF=32767)/

Rick,

You will not see a DC output until you remove both transformers in the path and short the connections with 0 Ohm resistors.

Regards,

Jim 

I don't see AC either.

Jim,

Could you present a procedure, and maybe a register file (__.txt) to load to achieve measurement of these parameters?

I'm running out of schedule.

I tried what you wrote, but no output changes.  1: Transformer removed. 2: Common mode chokes unchanged and shorted.  3: Shorted the 3 resistors (0 ohm, DNI) around the transformer.  More details below.

Outputs are always:

A2=0Vdc, 0Vac

B2=1.55Vdc, 0Vac

C2=1.40Vdc, 0Vac

D2=0Vdc, 0Vac

A2-A1 = 0Vdc

B2-B1 = 1.0358Vdc

C2-C1 = 0.6994Vdc

D2-D1 = 0Vdc

Values measured after I followed the procedure in the EVM User's Guide section 3.3 and 3.4, except no TSW1400.  Did the following:

3.3 Test Setup Connection

DAC348xEVM

1. Connect the EVM-supplied 18-AWG wires to the DC plug cable (Tensility 10-01776) to a qualified lab bench power supply. The 18-AWG red wire is the 6-V wire while the 18-AWG black wire is the ground wire.

2. Connect J13 connector of DAC348xEVM to J4 connector of TSW1400 EVM.

3. Connect a 6-V power supply cable to J18, the Power In jack of the DAC3484 EVM.

4. Connect PC’s USB port to J14 USB port of the DAC348x EVM.

5. Provide a 1.3Vp, 1.5GHz max clock at J9, CLKIN SMA port of DAC348x EVM.

6. Connect the IF output port of J2, J3, J6, or J7 to the spectrum analyzer. oscilloscope

Jumpers confirmed/moved as show in the table.

3.4 DAC348x Example Setup Procedure

1. Provide the clock input 1228.8 MHz at 1.5Vrms at J9 SMA Connector of the DAC348x EVM.

2. Turn on power to the board and press the reset SW1 button on the EVM

3. Press the “Reset USB Port” button in GUI and verify USB communication.

4. Select the appropriate EVM platform on the software menu.

5. Click “LOAD REGS”, browse to the installation folder and load example file

“DAC3482_FDAC_1228p8MHz_4xint_NCO_30MHz_QMCon.txt”.

This file contains settings for 4x interpolation with the DAC3484 running at 1228.8MSPS. Load this file and wait a couple of seconds for the settings to go into effect.

I used at J9: 1.2Vp 1.2288 MHz (1000x lower freq) sine wave.  (and 1.5Vp 25MHz sine; can do other waveshapes, scope is 300MHz). Measured outputs.

Then on the GUI, tab "Input" selected "Enable Constant Input" and entered 0, 8191(1FFF) 16383 (3FFF) 32767 (7FFF).  All output measurements are identical.

Rick,

I am having the same problems with our setup in the lab. There may be an issue with the GUI. I am looking into this. What exactly are you trying to do with this test? Maybe there is some data from our test group I can pass on. Is this something that is not in the data sheet?

Regards,

Jim

Rick,

To do this test, you will  need to provide a data clock to the DAC EVM, which is required to clock in the constant value .I was able to get the setup working when the DAC EVM was connected to a TSW1400EVM, which was providing this clock.If you do not want to purchase a TSW1400EVM, you could purchase a TSW1406EVM, which is much cheaper and would allow you to run this test. Another option would be to send you a loaner TSW1400EVM. Let me know if one of these options will work for you.

Regards,

Jim  

Jim,
Sounds good. You were able to get DC out?

If the TSW1400EVM is only providing one clock, would you be able to ID what the clock must be and which connector & pins? Maybe I could just feed an oscillator to the pins.

Or if the TSW1400EVM must be used, I'll likely need extension cables between the EVMs (TSW and the DAC). Maybe 12" or more to protect the TSW1400. Any special requirements for the cable? ( flat cable, twisted pairs, length limits, etc.)

One more item, to help me with selecting which TSW140x.  From SLWU079D:

 I also must test the ADS4149, and its EVM's User Guide calls out the TSW1200EVM.  TSW1200 is obsolete.  Would the TSW1405EVM work with both the DAC3482EVM and the ADS4149EVM?

Rick

Jim,

Could you describe how to obtain the DC measurements?

I'm still unable to obtain DC measurements to confirm spec values.  I received the TSW1406EVM & installed the GUI, connected the DAC3482EVM (its GUI previously installed), followed the DAC'EVM User's Guide section 3. Next I changed in the DAC GUI tab "Input", changed "Input format" to "Enable Constant Input", and put in various values.  However, there was no change in the dc output.

Also, the DAC3482EVM was/is modified with transformers removed, and installed the 3 zero ohm resistors around the transformers.

Thanks,

Rick