This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

THS5671EVM: Suggestions for DSP/FPGA eval board

Part Number: THS5671EVM

Hi,

I'm looking to evaluate the THS56X1 DACs for an application that requires a low noise pattern generator (100 MSPS, with a modest memory depth of a few hundred points played in a loop). I'm looking for suggestions/recommendations for a DSP/FPGA eval board to use to provide the digital inputs to the EVM? I know the datasheet suggests a C542 board with an adapter board, but it seems like this solution is obsolete. I'm really only interested in quickly evaluating whether the DAC works for us, so the simplest solution that minimizes design or programming time would be best (so ideally an eval board with a known compatibility with the CMOS inputs on the THS56X1EVMs, and good documentation).


Thanks!

Naresh

  • Naresh,

    You could probably test this with a TI TSW1400EVM. You would just have to jumper across the headers on both boards with some simple jumper wires. If you decided to pursue this approach, we can assist you with the required pin to pin mapping you will need to do for these wires.

    Regards,

    Jim

  • Jim,

    Thanks! That sounds like it will work, at least as a place to begin. Would you please send me whatever you can on pin mapping, and I will familiarize myself with the board and place orders in the meantime.

    Thanks
    Naresh
  • Hi Jim,

    Also, just to make sure - the description for the TSW1400EVM mentions a future firmware upgrade to support CMOS signals, can you confirm that this upgrade is available right now? Otherwise I'll have to ask you for another recommendation.

    Naresh
  • Naresh,

    That is a very old comment as the TSW1400EVM has supported CMOS interface for several years now. One of the connectors on the board is dedicated to CMOS interface only.

    Regards,
    Jim
  • Naresh,

    Connect them as follows:

    TSW1400 Pin #                   THS5671 Pin #

    1                                                              33

    3 (Not used)                                                      

    5                                                              31

    7                                                              29

    9                                                              27

    11                                                           25          

    13                                                           23

    15                                                           21

    17                                                           19

    19                                                           17

    21                                                           15

    23                                                           13

    25                                                           11

    27                                                           9

    29                                                           7

    31                                                           5             

    Connect a few GNDs between any even pins on both connectors.

    Regards,

    Jim

  • Thanks, Jim. The boards should be here tomorrow, and hopefully everything will work as expected.

    Regards,

    Naresh

  • Jim,

    Are you sure the pin assignments you gave me are correct? On the THS5671 EVM board, pins 29 and 31 on the connector J1 are not used. The 14 data bits on the THS5671EVM are on pins 27 (LSB) through 1 (MSB). According to your wiring connections, the MSB on the DAC will not be connected to anything. Should I be connecting  pins 1-27 on the THS5671 to pins 31-5 on the TSW1400, and pin 33 (THS565671 CLK) to pin 1 on the TSW1400? Perhaps I can work it out myself if I know what the output pins on the TSW1400 CMOS connector are programmed to be. Please let me know!

    Thanks,

    Naresh

  • Naresh,

    Sorry about that. The mapping should be as follows:

    TSW1400 Pin #                             THS5671 Pin #

    1                                                            33   (CLK)

    3 (Not used)                                                                                                             

     5                                                           27 (D0, LSB)

     7                                                           25          

     9                                                           23

    11                                                           21

    13                                                           19

    15                                                           17

    17                                                           15

    19                                                           13

    21                                                           11

    23                                                           9

    25                                                           7

    27                                                           5 

    29                                                           3

    31                                                           1   (D13, MSB)      

    Regards,

    Jim

  • Also, I think I will need further documentation/help on how to configure the HSDC Pro software to be used with the THS5671EVM. In particular:

    1. Is there an onboard clock on the TSW1400? How do I set the frequency of the clock? Or do I need to provide a separate clock signal to either EVM? If an external clock needs to be provided -- where, what frequency and what voltage levels?
    2. Do you have an example program or test pattern file I can use to start off before I create my own patterns? Or a list of settings on the HSDC Pro software? (For example, should I use the settings from a different CMOS DAC eval board that has better documentation?)

    Sorry about the multiple questions, but having spent a bit of time on this, I don't think there is any way for me to work out these issues by myself given the documentation I have. I'd appreciate help in getting me to a stage where I can get the DAC to play out a simple pattern, say a sine wave. I can take it from there.

    Regards,
    Naresh
  • Naresh,

    I was able to setup a THS5671EVM with a TSW1400EVM and capture a sinewave out of the DAC. The attached document explains the setup and should help you get up and running. Ignore the previous posts about how to connect the data lines as they were wrong. I was off by a bit or two and the data coming out of the TSW1400 is reversed with the  test file I used.

    Regards,

    Jim

    THS5671_TSW1400.pptx

  • Thanks, Jim, this is very helpful.

    I will give this a try, hopefully no further questions. I have the same minicircuits splitter you're using, and I can use a high-end arbitrary waveform generator to generate a 3.3V clock for now. However, that is a bit of overkill just to supply clock signals, and I will probably also purchase a TI clock generator EVM to supply the clock. Pretty much any clock generator should work, and I see that several ICs have EVMs available, but please let me know if you have a recommendation you like for a clock generator chip.

    Regards,

    Naresh

  • Naresh,

    I would post your clocking question on our clock forum. Since this is an old CMOS device, I do not know what would be the best low cost clocking solution for this application but they should.

    Regards,

    Jim

  • Hi Jim,

    Not to resurrect an old thread -- your replies were very helpful, and we were able to make a lot of progress. I also found the clocking solution I was after. One further request for pointers: I use the TSW1400EVM to load arbitrary waveforms to the THS56X1 DAC from an external csv file. This waveform is played back in a loop. Do you know if the TSW1400EVM has the ability to output a trigger/sync signal each time the waveform starts? It seems to me that the eval board clearly has the hardware to be able to do this, but the HSDC Pro software doesn't have the capability? Is there (or are there any plans in the future) for an API etc for the board?

    Right now, I'm using the complementary current output from the DAC and conditioning it to form a digital trigger -- apart from being a clumsy solution with too many extra parts, there is some concern that I'm affecting the noise properties of the circuit. I would like to leave the analog part of the circuit untouched. Please let me know if you have any suggestions, short of having to build our own FPGA/DSP circuit.

    Thanks,
    Naresh
  • Naresh,

    Yes, a pulse is being sent on SYNC4 (J17) SMA, when the pattern starts reading from DDR for every iteration.

     

    Regards,

     

    Jim

  • JIm,

    That would be just perfect, and save me a lot of trouble. However, it does not seem to be happening on my board. None of the SYNC outputs send any pulses. Is there a switch/jumper/software setting that needs to be enabled? All my switches/jumpers on that board are at factory defaults. If it matters, I'm using HSDC pro v4.70 on Windows 10, using the "cmos" firmware for the DAC board. I have screenshots and pictures of oscilloscope traces if you want me to send them to you to verify this, but they don't add anything new. The DAC plays out exactly the waveform that I desire, but there is no SYNC pulse.

    Thanks,
    Naresh
  • Naresh,

    I am checking with the design team but this. What are you using for a data file for the TSW1400EVM?

    Regards,

    Jim

  • Jim,

    Thanks for checking, this would really be a big thing for us. Here is a test file that I'm using to generate a simple 2.56us period triangle wave with a 50MHz clock (I don't think I can attach files on this forum, I'm just pasting the contents of the file named "test.csv" below the dashed line, it is a file with 128 lines, each line corresponding to one data point). In the HSDC Pro software I set the clock to 50M, and the DAC option to 2's complement (don't ask me why. The software seems to scale /bit shift things - I don't care. This works).

    With these settings, the DAC output is a 10 mA p-p triangle wave (I have analog signal conditioning at the DAC output, not important here).

    Look forward to what the design team has to say. Also if you have a different version of the HSDC software, and you can get a SYNC output, please let me know. I'm fine with downgrading to a previous version.

    Naresh

    ------------------

    0
    512
    1024
    1536
    2048
    2560
    3072
    3584
    4096
    4608
    5120
    5632
    6144
    6656
    7168
    7680
    8192
    8704
    9216
    9728
    10240
    10752
    11264
    11776
    12288
    12800
    13312
    13824
    14336
    14848
    15360
    15872
    16384
    16896
    17408
    17920
    18432
    18944
    19456
    19968
    20480
    20992
    21504
    22016
    22528
    23040
    23552
    24064
    24576
    25088
    25600
    26112
    26624
    27136
    27648
    28160
    28672
    29184
    29696
    30208
    30720
    31232
    31744
    32256
    32768
    32256
    31744
    31232
    30720
    30208
    29696
    29184
    28672
    28160
    27648
    27136
    26624
    26112
    25600
    25088
    24576
    24064
    23552
    23040
    22528
    22016
    21504
    20992
    20480
    19968
    19456
    18944
    18432
    17920
    17408
    16896
    16384
    15872
    15360
    14848
    14336
    13824
    13312
    12800
    12288
    11776
    11264
    10752
    10240
    9728
    9216
    8704
    8192
    7680
    7168
    6656
    6144
    5632
    5120
    4608
    4096
    3584
    3072
    2560
    2048
    1536
    1024
    512

  • Naresh,

    The design team has informed me that a pulse for every cycle start is sent to SMA (J17) in the TSW1400 DAC LVDS firmware but not when using DAC CMOS firmware (from his previous post).  

     

    This pulse they think seems to be sent to CMOS header to Pin_33 (CMOS_port0_16), and Pin_79 (CMOS_port1_17) but has not been tested.

     

    Regards,

     

    Jim

  • Jim,

    Thanks. I can confirm after a quick check that there is a synchronization pulse present on these two pins. This is good enough for me, I will jumper/solder to these pins and get my trigger. I would suggest sending this pulse to one of the SYNC outputs in a future firmware upgrade.

    Regards,

    Naresh