• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 31 subscribers
  • Views 846 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TM4C129ENCPDT: Over Current Protection for TM4C129ENCPDT

RAGHUL N M
Intellectual 490 points
Part Number: TM4C129ENCPDT
Other Parts Discussed in Thread: TPS2041B

Hi

We are developing a industrial product for switching many relays using TM4C129ENCPDT. The TM4C129ENCPDT uses SPI to communicate with some ULN driver ICs that drives the relays.

The TM4C129ENCPDT uC IC is powered by stepping down a 24V regulated dc input supply to 3.3V dc using LM1085ISX-3.3 LDO voltage regulator. For Over Current protection we have used TPS2041B IC in between the LDO and TM4C129ENCPDT uC IC. Please look into the circuit diagram below. The TPS2041B IC limits current as 500mA continuous.

The 3.3V supply of TM4C129ENCPDT IC and the 24V of relay coils  have same  ground points, since 3.3V is stepped down from the 24V using LDO. The uC IC is programmed to use SPI, UART, Ethernet and GPIO peripheral.

The issue is while operating more relays, the TM4C129ENCPDT got burned. 

1. The TPS2041B IC limits current as 500mA continuous. Does 500mA current is enough to burn the uC IC?

2. The 24V input supply to relay coils and 3.3V to uC IC shares the same Ground point. Does this cause the issue?

3. The TM4C129EXL Evaluation board uses TPS73733 IC for Over Current Protection in its schematic. The Current Limit is mentioned as 1A in the data sheet. Is that correct? 

4. Is there any other reasons that might have caused the issue?

Noted points:

1. The uC IC consumes 180 mA under normal operating condition.

2. Our application cannot isolate the ground points of the relays and micro controller IC.

3. Since we are using LDO to convert 24V to 3.3V. So 20.7V gets dropped at LDO. And 180 mA is consumed by the uC. So 3.7 Watts of heat will be produced. So the LDO IC is always hot.

Thank You for your time.

 

  • 0
    TI__Guru**** 174346 points

    Hi,

      - What is the percentage of the chips that are burned?

      - I think  you experience a EOS (electric over-stress) to the MCU. You will need to find out if you ever exceeded the absolute maximum ratings of the MCU on the potential between the VDD and ground which is 4V. See below excerpt from the datasheet.

      - Monitor and compare your supply to the MCU wrt different number of relays? What did you notice? Do you have stable supply to the MCU?

  • 0 in reply to Charles Tsai
    Guru 117855 points

    Hi also,

    We are "in doubt" of your diagnosis of, "Over-Current" being the "originating cause" of your MCU failure.   With the presence of multiple, 24V Relays - large & proven destructive voltage transients (even ground shifts) are to be expected.    Were proper transient Safeguards - along w/an effective "Design Architecture" implemented & in effect?

    In addition to vendor Charles' guidance - may we note:

    • your use of a linear regulator - rather than a switcher - to reduce 24V to 3V3 - raises my group's temperature.   (yet not as much as that of that hapless V-reg!)   You note it as "hot" - in fact it is (likely)  Burning-UP!
    • sharing the ground between your MCU & multiple relays is unwise.    Just as you've proven!    Why cannot those grounds be independent - you've clearly recognized the "weakness" of your existing (overly convenient) design?   Convenience should appear WAY DOWN your design list!
    • have you properly installed, "Anti-Kickback Diodes" across ALL Relay Coils - to absorb the (deadly) voltage spikes?
    • your (nicely drawn) schematic reveals the power supply - yet "Nothing" regarding the MCU's connection to your "ULN family Darlington Drivers."   Certain ULN/UDN drivers include series input resistors - which (may) provide (some) protection for those devices which, "drive the ULN drivers."   That level of connection detail may prove most useful.
    • your use of the MCU's SPI to drive multiple Relays (almost) insures that a "SIPO" (Serial Input, Parallel Output) IC receives your SPI outputs (data & clock, for sure - possibly a latch, too.)   Further detail (possibly schematic) assists.
    • there is no mention of the separation distance between the Relays & your MCU.   Should relay wiring (especially the high current "Contacts - if they carry power" and the coil) come close to (especially parallel to) any of the MCU signal lines - damage. may result.
    • unnoted is whether Relay Contact Closure - or Opening - proves most damaging!   It is when the relay coil is "opened" that the inductance generates damaging transient voltages (which may vastly exceed your 24VDC!)

    During Relay Opening and Closing - you must (carefully and accurately) monitor: (via a decent scope - not logic anal.)

    • the 24V supply
    • the 3V3 supply
    • any all SPI signal lines which drive your *unmentioned yet assumed SIPO IC"
    • any all MCU pins which have the misfortune of running parallel to - or come close to - high current (likely) relay (coil or power switched wires or pcb traces)

    Rather than (further) "Risk your MCU" - you may drive those ULN inputs via an inexpensive counter IC - such "repeated on/off signalling" will "Speed, Ease, Enhance" your troubleshooting - while "preventing" the loss of (multiple) replacement MCUs.

    You may consider Power "N-FETs" - which can deliver 100+ Amps - and (avoid) most all of the destructive "side-effects" - which are "normal/customary" for the (passé) relay...   And which eliminate that "Power Hungry" relay coil - & its (proven) Semi-Killing tendency!