• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 965 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.

THVD1510: Surge transients protection IEC 61000-4-5

NIKOLAS ST
Prodigy 60 points
Part Number: THVD1510
Other Parts Discussed in Thread: THVD1500, , TVS1401

Hello all,

in the TVHD1510 datasheets (THVD15xx devices RS485 transceivers) as in other devices and application notes

a suggestion for single stage surge protection is by using the SM712 (Bourns).

My question is how the SM712 device can provide surge protection at 1KV (8/20usec) as defined in IEC 61000-4-5.

According to IEC 61000-4-5 the 1KV surge voltage will discharge through a resistor of 40+2 Ohm, producing a current of almost 24A (23.8A).

In the specs of SM712 (3 manufacturers) the max clamping current is lower than 24A,

and with some degradation due to temperature (assuming a working temperature of 50oC),

the clamping current is reduced to 80%.

25oC 50oC
Bourns 17A 13.6A
Littelfuse 19A 15.2A
Semtech 21A 16.8A

So, it seems to me, that the above devices would not withstand the surge current.

In the case where we assume the bus lines as symmetrical lines, we could split the surge current with two 40 Ohm resistors,

which then would be (1000V/40R)/2 = 12.5A.

In this case the above devices would withstand the surge current,but

the clamping voltage will still be higher than the absolute maximum voltage on the bus pins (THVD15xx : +/- 18V)

Neg. Pulse Pos. Pulse
Bourns -13.1V 23.8V
Littelfuse -16V 26V
Semtech -11V 21V

(note: I am not fan of this approach, as in real industrial world the bus lines could be anything but symmetrical)

The above clamping voltages are extracted by the graphs in the data sheets (Semtech,Littelfuse),

and extrapolated (Bourns) by the data given. Neg. Pulse : (1A,11V),(1A,19V) and Pos. Pulse (1A,19V),(17A,26V).

 

The way I see it, in any case the bus pins on the transceivers are unprotected and the devices will be destroyed by the surge pulse.

The 10 Ohm series resistors will definitely limit the current flowing in the pins,

and the energy dissipated on the transceiver due to surge pulse, but there is no information in the data sheets regarding this.

Apparently so many wise engineers in TI and in the community now better,

but I am trying to understand the way the SM712 protect the transceivers and at the end how I can choose the proper TVS device.

Please enlighten me and comment at will.

TI Transceivers: THVD15xx

Bourns : SM712

Littelfuse : SM712

Semtech : SM712

PS : Having read a number of application notes, blog articles and datasheets, I have to thank all these TI engineers provide this knowledge.

  • 0
    TI__Mastermind 25705 points

    Nikolas,

    Your concerns are valid. Ideally we would want to have an TVS diode with Ipp >24A and Vpp less than the abs max of the bus pins of RS-485. Since the energy is dissipated through TVS diode and IC (internal ESD), the current path and response voltage could be different case by case. The abs max value of the bus pins are specified by DC. With transient stress, it could have some margin. Putting all the non-ideal factors together, the only way to verify the design might be the lab tests. If the surge protection is a real concern to you, it would definitely be better to put a 10Ohm series resistor on bus.

    You may have read all these, but here are the links of some technical notes. Please let me know if you have more questions.

    How to choose TVS diodes

    TI design for IEC, EFT and surge protection

    Regards,

    Hao

  • 0 in reply to Hao L
    Prodigy 60 points

    Dear Hao,

    many thanks for the reply.

    The article you mention above "How to choose TVS diodes " is the one that confused me initially.

    Although the text was clear, the example with the SM712 and the THVD1500 was not.

    Of course, according to your explanation, we accept the pairing (SM712 with THVD1510) as the lab tests were successful.

    Going through to the next proposed article TI design for IEC, EFT and surge protection , and regarding the IEC 61000-4-5 tests described,

    I see the tables with the test voltages but I do not see the relevant current tables (Ip).

    What is the internal resistance used during the surge tests, and what is the current level (12A, 12.5A or 24A) ?

    The same question applies to the datasheet of THVD1510 where the 1KV surge immunity is mentioned (what is the Ip surge tested?)

    I have also seen the excelent Flat-Clamp TVS series from TI,

    but it may not fit on RS485 bus protection due to leakage current

    as the slew rate will be higher than 2.5V/us (115.2Kbs, 250Kbs).

    Is that correct ? or we may use a device like TVS1401 for RS485 bus pins protection?

    Regards,

    Nikolas

  • 0 in reply to NIKOLAS ST
    TI__Mastermind 25705 points
    Nikolas,

    In the TI design, the impedance is 40Ohm. Same as that in THVD1510 datasheet. TVS1401 may not be a good option since it's standoff voltage is -14V to 14V while general RS-485 transceivers are -7V to 12V. I'll address more about your confusion.

    Regards,
    Hao
  • 0 in reply to Hao L
    Prodigy 60 points

    Dear Hao,

    the RS485 typical voltage range is -7V to +12V,

    but as the THVD1510 transceiver bus pins can go up to -/+18V (extending the typical RS485 voltage range),

    I think we could use a symmetrical TVS device.

    (In the case of the excellent TVS1401, we have to use two devices, one per bus pin)

    Regards

    Nikolas

  • 0 in reply to NIKOLAS ST
    TI__Mastermind 25705 points

    Nikolas,

    Actually THVD1510 works with symmetrical common mode -15V to 15V. In this sense, you can use TVS1401. But please be aware of the slew rate limit of it. THVD1510 rise/fall time could be more than 2.5V/us in normal operation depending on your application.

    Let me try to clarify the clamping voltage a little bit more. The difference you see from the datasheet of Bourns, Littelfuse or Semtech is due to the power rating. If you want to have lower clamping voltage, you may want to choose a TVS with lower power rating. Back to IC spec, usually you can leave a little margin on the abs max. In the real system you may need to find a balance on the cost and performance. Please let me know if you have more questions.

    Regards,
    Hao