• State Resolved
  • Locked Locked
  • Replies 51 replies
  • Answers 3 answers
  • Subscribers 31 subscribers
  • Views 27617 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.

EK-TM4C1294XL: ADC0 locks up from channel input spike when GPIO external trigger

Genatco
Guru 55913 points
Part Number: EK-TM4C1294XL
Other Parts Discussed in Thread: LM3S8971, INA240, TIDA-00778, DRV8305, , TL431, TM4C1294NCPDT, INA303, UCC27714, UCC21520, TIDA-00195

After removing an intermediate dongle PCB 3" jumper wires between inverter HVIC PCB and 3" jumpers from EK, ADC channels get EMF spiked locking up the entire ADC0.

The ADC channel input typical voltage divider 3-474k(1.4 megohm) in series 6.8k parallel 200pf tied to ground, 3v3 TVS diode, series ferrite into ADC channel all re-located to ground plane of HVIC. At launch pad added 100pf caps tied ground at each X11 header pin for  6" jumper wire over to HVIC inverter PCB. Previously we used 2-3" jumpers between the dongle versus 1-6" or HVIC 3" to dongle 3" to EK, dongle PCB was in the middle and worked well even with spikes.  Oddly 4-5v spikes exceeding VREFP/VDDA did not effect the ADC with the dongle PCB decoupling the signals and a single wire jumper into EK.

The dongle PCB had only 1-474k divider no TVS diodes, no ferrites and only randomly tripped HVIC faults and never locked up ADC0. Reducing ADC0 channel amplitude below 500mv has not helped to arrest the offending spikes that inevitably appear to now lock up the TM4C ADC0 at a specific level of EMF. We never had ADC0 lock up from EMF spikes with LM3S8971, said to simply chop any voltage over the ADC0 +3v0 VREF. Adding more capacitance for signal roll off effects the signal being sampled distorting the required zero crossing points. Perhaps TM4C1294 single ended channel mode can not tolerate signals jotting below ground without crashing the ADC0? 

Please suggest a way or IC to allow 500mv inductive EMF into ADC channels and block +/- voltages exceeding VREFP in a single ended channel signal.

5/4/2017 Later in this post discovered GPIO external trigger once again to be very sensitive to EMF spikes. A leaky 0.1uf/200v ceramic capacitor seemed to elevate the EMF spikes.  

  • 0 in reply to cb1_mobile
    Guru 55913 points

    cb1_mobile said:
    This Isolation greatly reduces "Ground Noise,Bounce,"Chatter" (all highly suspect as contributing to your (many) design woes)!      

    Couldn't have been the bad capacitor causing HV spikes mentioned throughout this post. A fact stated TIDUA15A–June 2015–Revised August 2015 (6.2/Pg-9. inverter stage) TIDA-00195 proper placement decoupling capacitor (0.22uf) is critical for stopping switch node dv/dt voltage spikes at the source. Imagine the frequency TIDUA15A being 40Khz high frequency has some thing to do with the value of the decoupling capacitor?

    cb1_mobile said:
    Yet you (claimed) such Isolated grounds "could not work!"  Your past post w/that claim was "frozen in time" (for your acknowledgement, consideration & review).

    That would be a misquote on your part. You are suggesting a complete PCB redesign to stop dv/dt spikes when a simple TVS and a few capacitors have achieved the same desired results. Consult TIDUA15A (Pg.9) for proper HV isolated DC bus voltage monitoring, requires several additional parts and separate +5v isolated supply to achieve MCU ground isolation. Perhaps at 22KW power level (full) supply isolation is a necessity especially with high power IGBT modules in the mix. Note also the gate driver uses 10 meter twisted pairs out to the 400v motor IGBT module, has elevated the need for isolated supply. 

    cb1_mobile said:
    For FIVE YEARS (perhaps even longer) you have been trying to develop a BLDC Controller - and yet (most always) challenge the suggestions of others!  

    The point is we have to deal with the existing design not change it completely at this juncture. Finishing a custom PCB design started last year also checking SW pitfalls TM4C1294 and other past (half witted) BLDC control issues Stellaris professionals left for morbid has caused numerous delays along the way.

    Not one word captures posted above do seem to indicate TVS diodes also snub HV switching spikes blocking them from entering the ADC channel.