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.

ADS8556EVM lower eight bits are not changing

Other Parts Discussed in Thread: ADS8556, SN74AHC138, TMS320F28335, TMDSDOCK28335, TMS320LF2407A

Hi all,

I am using an ADS8556EVM in the hardware mode . Lower 8 bits are not changing for any value of analog inputs. Iam using a  single strand wire connector to connect analog input from the signal generator.The parallel data output  is connected to 16 LEDs to see the converter output.Kindly help me.


Sham M.V.

  • Hi Sham,

    The ADS8556 has two modes of operation when using the hardware parallel configuration.  By setting the device in word mode (pin 29 low), the device will use DB[15:0].  When pin 29 is high, the data is transferred in byte  mode using  DB[15:8], with the byte order controlled by the HB EN pin with two accesses are required for a complete 16-bit transfer.




    Please let me know if this solves the problem.

    Thank you and Kind Regards,


  • Hi Luis Chioye,

    Thanks for your immediate response.I intend to use  ADS8556EVM for numerical distance relay application. I need 3voltages and 3 currents to be sampled simultaneously at 4k samples/sec.The input to the ADC is from Doble power system simulator and a second order butterworth filter with 400Hz cut off frequency is used as a filter. At present i am not using any controller.I am  testing the ADC8556EVM in hardware mode with the signals supplied by the an Agilent signal generator. And power supply is provided as given below.

    Hvdd ----->  +12v,  Hvss-----> -12V   Avdd-----> +5v   Bvdd------>+5v. Same analog signal is applied to all the channels.  Simple wire connectors are used to connect the analog signals from the signal generator to the analog inputs J2 on the EVM board. Please let me know if specific cables are needed to connect these inputs.

    The device is set to work in word mode by switching the SW1.1 to the right position. It shows correct output only for +10V(0111 1111 1111 1111) and -10V(1000 0000 0000 0000) analog inputs. For all other analog inputs lower 8 leds will continosly remain in low state. Kindly help me to  solve this problem .

  • Hi Sham MV,

    Can you please confirm if the internal reference and if the internal reference buffer enabled? You should be able to confirm/measure the reference voltage on Test point TP10 (REFIO pin).

    Please let me know all the switch configuration positions on the EVM (SW1), and if you are able to measure the reference voltage at TP10. 



  • Hi Luis ,

    I am furnishing the details as required by you.

    Measured voltage at TP10 i s 2.5V My input range is -10v to +10v . SW1 configured as follows.

    Switch           position

    sw1.1------     Right

    sw1.2------      Right

    sw1.3------      Right

    sw1.4------     Right

    sw1.5------      Right

    sw1.6------      Right

    sw1.7------     Right

    sw1.8------     Right

    sw1.9------     Right

    sw1.10------   Left. 

    Which type of connectors should be used to connect analog signals. EVM  can be placed openly kindly let me know

    Thanking  you ,




  • Hi Sham,

    Let's get a little detail on how you are connecting the hardware to your Agilent signal generator.  What model are you using and how do you have that tied into the EVM?  What are you controlling with it?  The SW1 conditions you mention above should be fine for hardware/parallel/word mode of operation.  There are a few additional signals on the board which must be configured properly in order to get data from the ADS8556. 

    First there is chip select (/CS) - the /CS input must be low.  There is a pull up on the board (R18) that connects into JP3.  The jumper on JP3 is loaded by default across pins 7-8.  To set /CS low permanently, you can add a shunt jumper across pins 9-10 of J1.

    You also have to ensure the standby (/STBY) input is high.  The /STBY signal is controlled through a 3-8 address decoder SN74AHC138.  By default,  /STDY is low - there are pull up resistors (R15, R34 and R35) on the select inputs and pull up/down resistors (R36, R33) on the enable inputs.  On page 2 of the data sheet for the SN74AHC138 (Function Table), you will see that this combination of enable/select inputs places a logic low on Y7 which is the /STBY input to the ADS8556.  Adding a shunt jumper to the DC_A0, DC_A1 or DC_A2 input on J4 should allow the /STBY input to be driven high.  You only need one, so for instance placing a shunt on J4 pins 7-8 will bring the A select input low which would send Y6 low and Y7 high.

    Finally there are the /RD and RESET inputs.  The /RD input is tied high through a pull up resistor (R40).  This signal has to toggle low for each channel to read data (see Figure 2, page 15 of the ADS8556 data sheet), otherwise the data bus will look static.  Assuming you have something toggling the CONVST input(s), you should see the BUSY signal toggling on J4.19.  After CONVST toggles high to low, you need to toggle the /RD input high to low for each active.  

    The RESET input is tied low through a pull down resistor (R39).  If everything mentioned above is implemented and you still don't see data changing, you might try toggling RESET high by momentarily shorting J1.1 to your DVdd supply with a jumper wire. 

  • Hi Tom Hendrick,

    Thanks for nice and detailed explanation on the EVM.I am furnishing the details that you have asked. 

    For power supply an Agilent E3631A Triple power supply is used. It is set to provide +5V,+12V and -12V for the ADS8556 EVM
    Bvdd and Avdd are connected to same +5V.

    We are using Agilent33250A 80Mhz functional arbitrary waveform generator to generate 800HZ 50% duty cycle which is used
    as Start of conversion signal(DC_TOUT).

    To set /CS low  a shunt jumper across pins 9-10 of J1 is used.Y7(/STBY) is made high by making DC_CNTL pin high ON J4.1.
    5V 8051 Micro controller is used to generate RD signal which is six times the frequency of Start of conversion signal.

    All six analog channels are connected to +5V (Avdd) using a single strand connectors.LED's are connected to D0-D15 of J5 to see the converted output.But the lower 10 LEDs are continuosly variying. For +10V,0V & -10V we are getting the exact output as mentioned in the data sheet[page no 30 ADS8556 SBAS404A-oct2006-Revised Aug 2009]. We have ordered 320f28335 and  it will be used as controller for numerical distance relay application.

    All the cables used for connecting are ordinary single strand wires.

     I Kindly request you to help us in resolving the issue.
    Thanking you,


  • Hi Sham,

    I am happy to hear that the lower eight bits of the ADS8556 data bus are toggling, that's good!  Using single strand wire should not be an issue for this testing that you are doing.  Your jumper configuration for STBY sounds fine.  One thing to keep in mind is that the ADS8556 does not 'latch' the output state of the data bus.  The bus will go tri-state when your /RD signal returns high which might cause the LEDs to flicker.  Saturating the inputs at +/-10V will also saturate the output code at 0x7FFF and 0x8000, so this sounds normal to me. 

    With the +5V input and the 4xVref Range, this would put the conversion results ideally at the 0x3FFF code.  The conversion results could be swinging a few bits above/below this ideal (say 0x3FFC to 0x4002 - see Figure 18) value, so that would also cause the LEDs to flicker as well. The better debug tool here would be a true logic analyzer, it would be interesting to see what your actual code variation is.

    Are you getting one of the TMS320F28335 eZdsp kits or are you building your own board?  If it is the TMDSEZ28335, that board as well as the ADS8556 will plug onto the HPA-MCU Interface Board which might make your formal debugging a little easier.

  • Hi  Tom,

    Thanks very much  for your guidance.I have checked with +5V input and the 4xVref Range, I am getting 0x3FFF code as output. An Agilent 54622D Mixed signal 100MHZ scope with 16 digital channels is used to observe the output.

    We recieved TMDSDOCK28335 - TMS320F28335 Experimenter Kit today.  We also have eZdsp2407A and 2812 with us, but for the presnet application I need a floating point processor therfore we bought 28335.I need to convert 6 analog channels simultaneously  at 800Hz .Kindly advice me on  use of HPA-MCU Interface Board for interfacing TMDSDOCK28335 - TMS320F28335 and ADS8556 which will reduce wiring requirements.If it is benificial I will buy HPA-MCU Interface Board immediately.


    Sham M.V.



  • Hi Sham,

    Since the TMDSDOCK28335 won't be able to plug directly onto the HPA-MCU interface card like the eZdsp would, I'm not sire how useful it would be.  SAMTEC has extension headers that we use all the time that might be a better option in this case.  The TMDSDOCK28335 does have a SIP strip of in-line pins which breaks out the parallel data bus.  You might try ordering samples of something like the SSW-116-23-F-D to wire into the parallel data bus and the SSW-110-23-F-D for the parallel control interface.  If you have someone familiar with 'wire-wrapping' at your shop, this might be the fastest/easiest way to connect the two board setups.

  • Hi Tom,

    Thanks for your suggestions and timely help.

    Thanking you,

    Sham M.V.

  • Hi Tom Hendrick,

    I have interfaced ADS8556EVM with TMS320LF2407A successfully.All the results obtained are as expected.

    Next I've interfaced ADS8556EVM with the TMDSDOCK28335. The control signals are same as that of generated in the 2407 but,unfotunately  i am not getting the results.
    Input port always remains at high state.

    I've checked port A configured as input by giving test inputs seperately it works perfect.I am getting this problem only when I interface ADS8556EVM.kindly help me in resolving this issue.

    Thanking you,
    Sham M.V.

  • Hi Sham,

    Can you possibly post schematics showing the delta in the wiring from the 2407 to the 28335?

  • Hi Tom,

    Thanks once again for immediate response. I think that i did't put my question right. I've interfaced ADS 8556  with lf2407 and 28335 control card seperately no  delta connection between them.With 2407 i had no problem ads8556 works fine. But when ads8556 interfaced with 28335 control card the input port always remains high(it will not read the ADC output. )






  • Hi Sham,

    Can you provide the schematic or a description of the connections between the two?  I'm not clear on how you could just flip from the F2407 to the F28335 without any changes.  The pin out on the control cards is different.

  •   Hi Tom,

    The 16 bits of ADS8556 are connected to GPIO 0  - GPIO 15 respectively of 28335. control signals Start of conversion and read signals are assigned to GPIO 32 ,33 respectively.End of conversion is monitored on GPIO 65.I have seen the control signals and end of conversion on the logic analyser they are perfect.But the ADC output is not read by the input port of 28335.Please help me.


  • Hi Sham,

    So are you saying that the data is actually there, you just can't read it via the GPIO's?  Can you post your setup code and maybe the LA traces?

  • Hi Tom,

    Thanks for your excellent support. I have interfaced ADS8556EVM with DSP 28335. We have generated the control signals and read all the channels successfully and implemented our  algorithm too. But some  times ADS output will be erratic , if  different signals are connected to  individual 8556 input channels. The variation will be huge and cannot be correlated. If i connect all input channels of 8556  to a Single source/signal, this problem will not be there, all the channels will read same value.Kindly correct us.

    Thanking you,


    Ananth naik & Sham MV


  • Hi Anath and Sham,

    Can you define 'erratic' for me as you see it? 

    Are you getting all zeros as if the conversion never took place on CH_x or are you getting what looks like two sets of CH_A data without CH_B or CH_C?  Can we get screen shots/LA traces of your control lines, maybe a schematic of the input?  If all three pairs are connected to the same input source and there are no problems, I would suspect there to be an issue somewhere in the input stage - source 2 and 3 are not referenced the same as source 1...

  • Hi Tom,

    I am furnishing the schematic of input connections and control lines observed on the logic analyser for your kind reference.
    I have connected a 5v(peak-peak) 50HZ signal to all the channels and obtained 16samples/cycle by appling SOC at 1.25ms interval
    and stored these values in 16 bit signed integer format. I have plotted all the six chaneels data  i got perfect values.
    I also applied DFT and obtained 4.988V on all channels.

    Later on I have connected different voltage level signals(1v,2v,3v,4v,5v,6v respectively) of 50Hz frequency to individual channels of 8556 and
    with the same program i see upto 30% change in the magnitude on either side after applying DFT.But some times it will show the exact magnitudes
    (i.e.1,2,3,4,5,6 on respective channels).We are unable to solve this problem kindly help us.



    Thanking you,

    Sham and Ananth.


  • Hi Sham and Anath,

    Can you show us what's inside the LPF block?

  • Dear Tom,

    There was a loose connection in  the two MSBs which was causing the problem. Now everythin is alright. Thanks for your support.

    Thanking you,



  • Great!  Thanks for letting us know.

  • Hi Tom,

     My hardware is working properly.Now I need nine analog signals to be sampled simultanneously to implement relaying algorithm..Can you please let us know how it can be done using ADS8556.At  present i am using  all six channels of ADS8556EVM in parallel data transfer in hardware mode.

    Thanking you,




  • Hi Sham,

    You could measure 9 analog inputs using two ADS8556 devices in parallel.  Simply use the first device as you are now, and on the second device you could use the A0, A1 and B0 channels.  You would need to tie the CONVSTs of both parts together and equalize the trace lengths to minimize skewing. 

  • Hi Tom Hendrick,

    My hardware setup was working properly, But today morning unfortunately first channel pair stopped working (A0,A1). These two channels read 16391 (in signed int format ) irrespective of any voltage level input. Kindly help to trouble shoot this. Remaining four channels are working properly.


    Thanking you,



  • Hi Sham,

    I'm not sure what I can do to help.  If you get a constant code out from pair A, but the others are correct, perhaps you physically damaged the silicon.  Do you have another board to compare the results too?  Another possibility is that you've got a broken wire to one of the CONVST pins.  Are you still working with 9 channels in all?

  • Hello Tom,

    Can I use computer SMPS to power up the ADS8556?  Please clarify.

    With regards,

    sham M.V