• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 45 subscribers
  • Views 485 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

ADS124S08EVM: Isolated USB interface to EVM

Leslie Green
Prodigy 20 points
Part Number: ADS124S08EVM
Other Parts Discussed in Thread: ADS124S08

The EVM has been working satisfactorily under Windows 10 Home (10.0.18362 build 18362), but it is grounded through the USB port, which is causing excessive noise. I bought a USB isolator, SEEED studio 114991949 (which actually came in a box marked 52Pi) This works correctly on a Windows XP machine for a flash memory stick, and also on a Picoscope 3206 USB oscilloscope. It also works correctly for the same flash memory stick on the Windows 10 machine.

 When the EVM is plugged in via the isolator you get the Windows connecting sounds, but no connection to the TI Delta-Sigma ADC evaluation software. The green LEDs on the eval board light up. The TivaWare Bulk devices report in the Device Manager is:

“This device cannot start. (Code 10)

 Insufficient system resources exist to complete the API.”

 The driver is Texas Instruments 2.1.4.0 from 18/10/2016.

 When the EVM is working correctly, the driver says it is working properly, but I wouldn’t say it was “happy”.

 ------------------------------------

Information box:

Device USB\VID_1CBE&PID_0003\12345678 was not migrated due to partial or ambiguous match.

Last Device Instance Id: USB\VID_0D8C&PID_0005\201306

Class Guid: {36fc9e60-c465-11cf-8056-444553540000}

Location Path: PCIROOT(0)#PCI(1001)#USBROOT(0)#USB(3)

Migration Rank: 0xF000FFFFFFFF0023

Present: false

Status: 0xC0000719

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

I have deleted and reinstalled the driver, and re-installed the delta-sigma ADC software.

I found     https://www.repairwin.com/fix-usb-device-cannot-start-code-10-error/

and disabled power saving in the USB driver (method 1). There were no upper or lower filters (Method 2). My machine has an AMD A10-7800 Radeon R7 processor so intel drivers (Method 4) seem inappropriate.

Is there a better driver available? Or any other suggestions?

  • 0
    TI__Guru** 113765 points

    Hi Leslie,

    Welcome to the E2E forum!  This is not a driver issue.  The EVM uses the standard WINUSB driver from MicroSoft.  The issue you are seeing is in communication speed.  The EVM has a high-speed USB port connection (480Mbps) and the the USB isolator can only operate with low (1.5Mbps) and full-speed (12Mbps) rates. So this isolator will not work with the EVM.

    The EVM was designed to communicate and power from the USB port.  At times we have heard customers stating that various computers will have higher noise.  In some cases customers have used a battery powered laptop computer to remove the AC powered connection.  But these cases have been isolated and rare.

    It is possible to power the ADS124S08 portion of the EVM board using an external low-noise bench supply.  The EVM itself uses the USB power source and steps it up to about 5.5V then uses an LDO to lower the supply for 5V analog and 3.3V digital.

    You say you are seeing excessive noise.  Can you elaborate on this and how it was determined?

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    Prodigy 20 points

    Bob, thanks for a very fast and definitive answer. The isolator says USB 2 “full speed”, which to my analog brain seemed fine. The distinction to “high speed” was totally lost on me. :-(

    I have a battery powered buffer circuit to give a differential drive to the EVM. It gives 126 lsb rms input, as displayed by the Analysis Engine readout. The input is coupled via an audio transformer (Hammond 142B) to break the ground loop from my FY6800 DDS generator. Connecting just the ground of the DDS to the (shorted input) of the transformer makes the rms noise (std dev) jump to 4740 lsb rms (at 1000 S/s), which for me is excessive.

    Incidentally, I have run the EVM at -10 dBFS and better than -100 dBc and the Analysis Engine still insists the ENOB is less than something like 3 bits. (I stopped using that readout when I realised how useless it is.)

    (I have got 12.8 lsb rms out of the hardware, but only when both inputs were internally shorted to the internal Vbias point.)

  • 0 in reply to Leslie Green
    TI__Guru** 113765 points

    Hi Leslie,

    The ADS124S08 is primarily meant for more DC type measurements such as slow moving input signals for temperature measurement, weigh scales, etc. The ENOB from the Data Analysis is really showing effective resolution in this case.  So if you have a varying input signal, the signal will appear as noise and not as a carrier.  Effective resolution as defined in the ADS124S08 datasheet is:

    Effective Resolution = ln[(2 · VREF / Gain) / VRMS-Noise] / ln(2)

    This is with respect to inputs shorted and 512 consecutive samples for the datasheet tables.  See section 8.1 in the datasheet. 

    Also note that the passband of the ADS124S08 is not flat.  The ADS124S08 is an oversampling device with a digital low-pass filter.

    Best regards,

    Bob B