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.

DAC3482EVM question regarding HSMC data interface

Other Parts Discussed in Thread: DAC3482, ADS4249

Hi,

I am looking at the DAC3482EVM for a solution to interface with an Altera development board via HSMC.  On the DAC3482EVM webpage there is a reference to the schematic/layout/BOM files of the DAC348xEVM (slrr002.zip).  In that layout design I am looking at the HSMC mounting holes (the two holes on either end of J13) and notice that they are set with a 2.7" gap line in-between (referenced to the hole centers [ø 0.112"]).  The standard mounting hole gap for HSMC is 2.80" in-line (ø 0.125").  Are these two mounting holes intended to provide a mechanical mount for the HSMC connector or is this data interface connector (J13) for something else?  It's signals seem to map to the HSMC LVDS signals correctly so I am unsure what to make of the mounting hole mechanicals.

Thanks,

Pete

  • Pete,

    The DAC3482 EVM is not directly compatible with the Altera development board. The 2.7" gap is for the alignment pins, and the alignment pins may block the mechanical connection to the host HSMC connector.

    The alignment pins are used to allow better mechanical connection of our TSW1400 EVM to the DAC3482 EVM. Since the alignment pins does not allow direct connections to your HSMC host board, we can rework your existing DAC3482 board to replace the connector with the one without the alignment pins. I will private message your offline for your information.

    The new compatible connector would be the QTH-060-01-L-D_DP-A. This connector eliminates the single-ended pins in between the differential pairs. This will prevent shorting the power supplies on the HSMC host board to the ground connections on the DAC3482.  The omission of the single-ended pins also creates an air gap between the differential pairs for iolsation, hence improving signal integrity. By default, this form of connector only comes with alignment pins installed. We will cut out the alignment pins during replacment so it will be compatible with your board.

    -KH

  • Case Summary:

    After replacing the locking connector with the QTH-060-01-L-D_DP-A connector, the DAC3482EVM can now mate with the HSMC host board. The new QTH-060-01-L-D_DP-A does not have locking connector pins, therefore it is not necessary to cut off the alignment pins as described above. 

    Pete was also able to run the DAC3482 to ADS4249 FPGA repeater project correctly (http://www.ti.com/analog/docs/litabsmultiplefilelist.tsp?literatureNumber=slaa545&docCategoryId=1&familyId=2023)

    See the following post for more detail:

    http://e2e.ti.com/support/data_converters/high_speed_data_converters/f/68/p/265521/934006.aspx#934006

    -KH

  • Kang,

    Thanks for that answer regarding the QTH connectors.  I did not realize the standard DAC3482EVM QTH family connector had locking pins built into it.  I ask because I would like to build a 2nd setup of the ADC & DAC configuration described above.  I am always using a QTH connector family adapter cable (Samtec can make specific versions of these) in my situation as I am connecting two TI EVMs to an Altera development board where the 2 HSMC connectors are placed immediately side-by-side.  I cannot directly connect the two EVMs to the FPGA board without an extension cable to prevent mechanical interference.

    To make this easier I thought I could potentially purchase a standard DAC3482EVM ( with the QTH -D connector [not -DP]) and then utilize an interfacing cable with a -DP QTH family connector on one side (this would prevent the DC power pins in the -D pins [every 3rd pin] from shorting to the GND pins of the TI EVM).  So is there another mechanical issue with the QTH family connector (the "locking pins" you describe) that prevents one from connecting a HQDP family connector to the DAC3482EVM connector (QTH-090-01-L-D-A)?

    Mating a HQDP twinax cable connector directly to the "factory" DAC3482EVM connector may be an easier option for my goal.

    Thanks,

    Pete

  • Pete,

    There is a plan in place to update the BOM and layout to reflect the compatibility with Altera. We will update the BOM with the correct -DP version (without the locking connector). 

    This has been placed on our project list. In the mean time, if you need to use the EVM to interface with Altera FPGA, we can help you rework the EVM's HSMC connector to QTH-060-01-L-D_DP-A free of charge. We can then swap out your EVM with the reworked EVM.

    -KH

  • Kang,

    Thanks for your response.  It is great TI is working towards supporting a -DP connector to directly interface with Altera development boards.

    But, my question was a slight variant of your topic above.  I am under the gun schedule wise and I have already ordered a stock DAC3482EVM.  My question is "will the Samtec QTH-090-01-L-D-A on the stock DAC3482EVM mechanically interface with a HQDP family connector"?  I understand I need a -DP signal translation somewhere in the chain between the TI and Altera boards to prevent the Altera power supply pins (Q series "3rd pin" sets) from shorting to the grounded pins on the DAC348xEVM design.  I am using a Samtec twinax adapter cable between the Altera board and the DAC348xEVM because my Altera board has components immediately near the HSMC connector that prevents the TI from a direct interface.

    I think I am OK buying a "factory build" DAC348xEVM since I am using a cable with a -DP option in it.  I don't think there is a "mechanical interference" issue to a HQDP family connector.  Just want to make sure of that.

    Pete

  • Hi Pete,

    I am not too familiar with the HQDP family of cables. The current DAC3482EVM in stock has the Samtec connector with retention option

    http://www.samtec.com/documents/webfiles/cpdf/QTH-XXX-XX-X-D-DP-XXX-MKT.pdf (look for page 4 of the document for RT1 - retention option). 

    I am afraid the two retention clips may interfere with your HQDP cable. I looked at the HQDP cables briefly, it seems like you will need the -060 option without the screw-down option in order for the retention clip to fit. I do not see the mechanical for the -060 with the screw down option, so it is hard to tell if the screw-down version of the cable  will work with the QTH connector with the retention clip.

    If you can help me find the mechanical of your HQDP connector (or your HQDP part number), I can double check.

    -KH

  • Kang,

    OK, thanks.  I was not aware of the RT1 option (this is the 1st time I am seeing the complete QTH datasheet), the BOM PN did not describe this.  I am using HQDP-060-STR (without the -F screw down option).  My take away from the HQCD family datasheet I have is that I will encounter about 0.018" of mechanical interference along the width portion of my HQDP connector, maybe a few mills more as the RT pin diameter is shown to widen at its base.  It looks like I can file the edges of my HQDP connector down to accommodate this mechanical retention pin (you described a filing process similar to this in other posts if I recall).

    Not ideal, but I already ordered the DAC3482EVM so I think I am stuck at this point...

    I have attached a datasheet I have for the HQCD family (same as HQDP) from Samtec.

    Thanks,

    Pete

    1157.Samtec_HQCD-XXX-XX.XX-XXX-XXX-X-X-MKT.pdf