Shop OBEX P1 Docs P2 Docs Learn Events
On the topic of CIRCUIT damage prevention, please advise. — Parallax Forums

On the topic of CIRCUIT damage prevention, please advise.

Clock LoopClock Loop Posts: 2,069
edited 2020-09-03 22:59 in Accessories
This is an extensive topic, but I would like your advice on protection of circuits.

Links to parts like TVS diodes, fuses, or other protection circuitry methods.
Schematic examples, etc..
I have used fuses (the more obvious) and TVS diodes, and also have used resistor dividers, etc.

Any ideas on the topic?

Do you have any neat methods using voltage dividers with zener diodes, or the like?
It seems today, that entirely too many companies don't provide ANY protection AT ALL.
Not even FUSES! (I think that is grounds for lawsuits using LEMON laws or neglect, sure you won't win, but any tech can easily make the argument .)

Over DECADES, I have burnt handfulls of fuses in my MULTIMETER due to stupidity...
(like leaving the current mode measurement on, and probes plugged into the current measurement plugs)

Dead%20Fuse.jpg

If they didn't design for morons like myself, I would have a DEAD 200$ multimeter, but since they DID, I have had this multimeter for 20+ YEARS, thanks radioshack!
Radioshack # 22-168A That is just PROPER engineering of a product.



Please suggest even the most basic things. Fuses, TVS, resettable fuses...
It would be nice to have suggestions to parts (links to digikey) and stuff for zener diodes, thermal fuses, tvs's, movs' etc etc etc for use with our MICROCONTROLLERS.

This means circuit suggestions for 5v and 3.3v (and with the Propeller2, even 1.8v)
How about some other common circuit voltages like 12v, and in model railroading 14v, and 24v is used alot for industrial electronics, etc...

No suggestion will be refused or considered too basic.
If you suggest a fuse, I will say, give me a link to the digikey part.

For instance:
I would accept a fuse suggestion for the Propeller 1, specifically to its current, the package can handle
Max. current out of Vss pins 300 mA
Max. current into Vdd pins 300 mA


And for pin protection one could even suggest:
Max. allowable current per I/O pin 40 mA

It would be nice to have other suggestions here for convenience for the Basic Stamp, BS2, SX chip, Propeller chip, Propeller2 chip, etc...
Sure one can look up each one and do it themselves, by why not help nooobs get it all in one thread?
If you don't want to do the work for this, then move on. :) I will contribute to this thread as I find the time.
Thank you!

P.S. Advice on protection from LIGHTNING would also be very welcome here, due to many of us putting sensors on our roof tops, outside, etc.
(like I will have a g-scale brass railroad track running around my yard, and it WILL be vulnerable to lightning)


«13

Comments

  • Clock LoopClock Loop Posts: 2,069
    edited 2020-08-31 23:56
    Here is a training module on TVS diodes.
    Suggestions on tutorials, or the like are welcome also.

    https://www.digikey.com/en/ptm/l/littelfuse-inc/transient-voltage-suppression-tvs-diodes/tutorial
    
    The VR (Reverse Standoff Voltage) parameter on the V-I curve is taken from zener diode terminology. 
    It is used to indicate the maximum continuous voltage that can be applied to the TVS diode that will not cause it to conduct. 
    It is measured by running a 1µA test current, called IR through the device and measuring the resulting voltage. 
    VR should be a higher value than the highest working voltage in the protected system.
    

    THIS IS A WORK IN PROGRESS


    Using this information, I have come up with these various TVS diodes for:
    These are selected using filters: ACTIVE, CUT TAPE, IN-STOCK with a WORKABLE sized package in smt. (SOT) (NO tsop tiny spec of a pepper)
    And the lowest price available with those filters.

    5v and above, the VR was raised due to regular differences in higher voltage power supplies, as you get higher, they tend to be OVER voltage a bit.

    NO CONSIDERATION IS MADE FOR CAPACITANCE, if you are that smart, DIY.


    TVS - Propeller2 power supply & IO protection - 1.8v - 100 mW ((ESD) only) (this will pop if above 1.8v @ 55ma ) 100 for 0.11040 each.
    SOT-553 = https://www.digikey.com/product-detail/en/toshiba-semiconductor-and-storage/DF5A3-6CJE-LM/DF5A3-6CJELMCT-ND/5765603
    ONLY OPTION THIS VOLTAGE with cut tape.

    Too close to 1.8?, next up is 2.5v
    TVS - Propeller2 power supply & IO protection - 2.5v - 100 mW ((ESD) only) (this will pop if above 2.5v @ 40ma ) 100 for 0.10120 each.
    SOD-323 = https://www.digikey.com/product-detail/en/toshiba-semiconductor-and-storage/DF5A5-6JE-LM/DF5A5-6JELMCT-ND/5403466


    TVS - Propeller1 & 2 - power supply & I/O protection - 3.3v - 1A peak pulse - (this will pop if it goes above 3.3v @ 1A) 100 for 0.12570 each.
    SOT23 = https://www.digikey.com/product-detail/en/diodes-incorporated/D1213A-01SO-7/D1213A-01SO-7DICT-ND/3340430
    I rarely ever see any 3.3v regulator put out OVER 3.3v

    TVS - Propeller1 & 2 - power supply & I/O protection - 3.3v - 200A SUPER LOL peak pulse (bustafuse) (you use fuses right?) - 100 for 0.22440 each.
    DO-214 larger package https://www.digikey.com/product-detail/en/vishay-semiconductor-diodes-division/SMBJ3V3-E3-52/SMBJ3V3-E3-52GICT-ND/1147834

    You may feel better with 3.6v since 3.3 is VERY close to the output of 3.3v regulators..

    TVS - Propeller1 & 2 - power supply & I/O protection - 3.6v - 20A peak pulse (bustafuse) (you use fuses right?) - 100 for 0.13150 each.
    SOD-323 = https://www.digikey.com/product-detail/en/stmicroelectronics/ESDA041-1JY/497-ESDA041-1JYCT-ND/11687263


    TVS - Basic STAMP, BS2, SX chip - Power supply & I/O protection - 5.5v - (no peak pulse info, assuming 1A for sot23 package) - 100 for 0.13460 each.
    (and to protect the 5v before your 3.3/1.8v regulators, also for I/O of 5v coming INTO the propeller)
    SOT-23-3 = https://www.digikey.com/product-detail/en/on-semiconductor/PACDN042Y3R/PACDN042Y3ROSCT-ND/3487662
    I did not choose 5v exactly due to many power supplies for 5v being over, many are 5.25v, so 5.5v was selected.

    TVS - Basic STAMP, BS2, SX chip - Power supply & I/O protection - 5.5v - 80A LOL (Mc Bustafuse) - 100 for 0.14630 each.
    SOD-323 (couldn't find larger package) https://www.digikey.com/product-detail/en/toshiba-semiconductor-and-storage/DF2S6P2FU-H3F/DF2S6P2FUH3FCT-ND/10131125

    5.5v too close?

    TVS - Basic STAMP, BS2, SX chip - Power supply & I/O protection - 6v - 58A LOL (Mc Bustafuse) - 100 at 0.18000 each.
    DO-214AA larger package - https://www.digikey.com/product-detail/en/littelfuse-inc/SMBJ6-0CA/SMBJ6-0CALFCT-ND/286102



    TVS - 12v protection. 13v tvs was selected. 1.7A 100 for 0.12440 each.
    SOT-23-3 = https://www.digikey.com/product-detail/en/on-semiconductor/SZMMBZ16VALT1G/SZMMBZ16VALT1GOSCT-ND/9087880

    TVS - 12v protection. 13v tvs was selected. - 28A (Mc Bustafuse) - 100 at 0.18000 each.
    DO-214AA (larger package) - https://www.digikey.com/product-detail/en/littelfuse-inc/SMBJ13CA/SMBJ13CALFCT-ND/285977


    TVS - 14v protection. 15v tvs was selected. - 20A (Mc Bustafuse) - 100 for 0.08130 each.
    SOT-23 - https://www.digikey.com/product-detail/en/smc-diode-solutions/S23LC15TR/1655-2112-1-ND/9566425


    TVS - 14v protection. 15v tvs was selected. - 24A (Mc Bustafuse) - 100 for 0.18180 each.
    DO-214AA Larger package - https://www.digikey.com/product-detail/en/diodes-incorporated/SMBJ15CA-13-F/SMBJ15CA-FDICT-ND/816059


    TVS - 24v protection. 26v was selected. 14A (Mc Bustafuse) - 100 for 0.18000 each.
    DO-214 https://www.digikey.com/product-detail/en/littelfuse-inc/SMBJ26CA/SMBJ26CALFCT-ND/286001

    More amperage is probably better at this higher voltage since many power supplies are commonly around 20A.
    TVS - 24v protection. 26v was selected. 75A (Mc Bustafuse) - 100 for 0.16900 each.
    DO-15 LARGER PACKAGE - https://www.digikey.com/product-detail/en/stmicroelectronics/P6KE30CA/497-12048-1-ND/2783790


    USA and CANADA, MAINS VOLTAGE - 120 volts ± 6% = 127.2 v

    TVS - 120v protection. 130v was selected. 39A (Mc Bustafuse, Probably not enough to be a ( "Mc TrippaBreaker", Better have a fuse) - 100 for 0.16900 each.
    R-6 (through hole) https://www.digikey.com/product-detail/en/micro-commercial-co/5KP130A-TP/5KP130A-TPCT-ND/2776380



    More EDITS coming (for larger parts and ones capable of busting your fuse (bustafuse) (you use fuses right?)
  • Cluso99Cluso99 Posts: 18,066
    Back around the 80’s there was a cartoon that was applicable to your discussion. Since I don’t have the cartoon, I’ll describe it.
    A fuse and a transistor on a pcb. The transistor is all burnt and obviously dead. The fuse says “he died to protect me”.
    This is the reality. Fuses are too slow to protect transistors, mosfets, ics, etc.
    For lightning protection, i found a small few ohm smt resistor in series in both phone lines followed by a transil/transorb across the lines caused the resistors to blow in most lightning strikes protecting the remainder of the modem. The few where the pcb was destroyed mostly came from NT (Darwin) where 13KV strikes were more common according to research.
    Unfortunately you can make a product foolproof but you cannot make it idiot proof.
  • Clock LoopClock Loop Posts: 2,069
    edited 2020-08-31 22:38
    Cluso99 wrote: »
    Back around the 80’s there was a cartoon that was applicable to your discussion. Since I don’t have the cartoon, I’ll describe it.
    A fuse and a transistor on a pcb. The transistor is all burnt and obviously dead. The fuse says “he died to protect me”.
    This is the reality. Fuses are too slow to protect transistors, mosfets, ics, etc.
    For lightning protection, i found a small few ohm smt resistor in series in both phone lines followed by a transil/transorb across the lines caused the resistors to blow in most lightning strikes protecting the remainder of the modem. The few where the pcb was destroyed mostly came from NT (Darwin) where 13KV strikes were more common according to research.
    Unfortunately you can make a product foolproof but you cannot make it idiot proof.

    Goes to say with everything, technically. Design a stairway with wide short steps, a handle railing, and you still have someone break their leg.

    But we still gotta try, it may not protect everything, but the cascading failure will be greatly reduced, making repair a feasible possibility.
    And TVS's do work VERY well in protection of semiconductors, since thats what they are.
    I have experienced this personally with IGBT's in motor drives that were connected to a brushless ac motor set to full power.
    The shaft was connected to a BAR that would hit another bar after 1 rotation, DEADLOCKING the motor at full power, without a TVS to ground on each phase leg,
    the IGBT would pop like an M-80, with the proper tvs, (and an array of them on each phase, a few of the TVS would pop and then the fuse did.
    The IGBTs survived fine, even under post long term testing, so did the igbt optoisolator drivers, and all the associated DSP + FPGA logic.

    In non-tvs drives with just fuses, the damage was unpredicable, sometimes the IGBT goes, and others the IGBT and optoisolator drivers, and others even the logic dies.
    TVS's do really work, and are going to pop if your fuse is above the tvs peak pulse. (then you need an array of TVS's)

    Cluso99 wrote: »
    A fuse and a transistor walk into a bar, the transistor says to the fuse, "I bet you I can drop dead from drinking too much, faster than you.", the fuse says, "I agree."

    But this is EXACTLY the kind of suggestions I am talking about, designing a circuit with a PROTECTION TRANSISTOR, LOL... ;)


  • I use optocouplers to protect my Propellers. I fried a number of Props over the years without giving much thought to the damage that inductive loads, like motors, were doing to my micros until this year.
  • lardom wrote: »
    I use optocouplers to protect my Propellers. I fried a number of Props over the years without giving much thought to the damage that inductive loads, like motors, were doing to my micros until this year.



    Do you use an optocoupler array?
    Or just some individual ones like H11AA1 or similarm to protect only the lines going to the motor drives(transistors, mosfets, igbts, etc)?

    Got any part numbers you find as your favorite?
  • Clock LoopClock Loop Posts: 2,069
    edited 2020-08-31 22:39
    Cluso99 wrote: »
    For lightning protection, i found a small few ohm smt resistor in series in both phone lines followed by a transil/transorb across the lines caused the resistors to blow in most lightning strikes protecting the remainder of the modem. The few where the pcb was destroyed mostly came from NT (Darwin) where 13KV strikes were more common according to research.
    Unfortunately you can make a product foolproof but you cannot make it idiot proof.


    Ok, so a few ohm resistor in series on vulnerable input/outputs. Less effective for sustained over voltage/current situations, unless you also include a fuse.
    (the smt resistors start to cook, this was the case in my DCS240) But it didn't have any kind of transorb/TVS, and those are kinda POINTLESS without a fuse to open the circuit.
    (see image below)
    No doubt a 1amp smt fuse would have kept them from cooking as much as mine did, and while the resistors are starting to cook...
    The fuses would have likely protected the 6 pnp/npn transistors that are also done, plus all the other circuitry that isn't obviously damaged. (which i can't find)


    I was an idiot, and the fuse would have protected me.
    The transistors were only rated at 1amp, they could have handled it for the few seconds it takes to pop a fuse,
    combine that with the series 22ohm burnt resistors, and I would most likely still have a 250$ DCC command station.

    But like you say, that 10 AMP fuse is useless and is only there to prevent LAWSUITS in the case where it burnt down someones $20,000 layout, (or even house)


    BUT you also mentioned a transil/transorb (THIS IS A TVS, I GUESS) across the lines, to ground or just I/O? Both?
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2020-08-31 23:24
    Anyone who knows me knows I HATE FUSES! But anyone who knows fuses know that they aren't there to protect the circuit, they are there to prevent a fire. The reason I hate fuses is because they come in all shapes and sizes, especially the ones that you haven't got. Then you have to get them, and be prepared to blow a few finding out that the circuit is shot and you shouldn't have bothered!

    I use all kinds of protection in my commercial gear, not just TVS diodes, but also HSPs (high-speed current protection) that can respond in 50ns. What's more, they don't "blow" but continue to operate normally after the surge. But I also use gas tubes for RS485 lines and series inductance (current lags voltage) and spark gaps, and a separate and a real earth ground etc.

    Even from decades ago I would arrange for a suitable low value low-wattage resistor in series with the mains voltage and sitting high on the pcb so that it wouldn't burn the pcb either. They also take up a lot less room than a bulky glass fuse and clips. Other times I also use polyfuses in low-voltage circuits. However anything that hooks up to car batteries and the like require some extra thought especially where pcb tracks are involved.

    Now there are CBs (circuit-breakers) on mains equipment that are supposed to protect the equipment and a few years ago I was working on 1500kVA 3-phase regulation and even the smartest CBs were hopeless. I was testing the 200A SCR "control" circuit (1000A rating) and there was a timing glitch and thick black smoke issued forth from the pallet sized regulator as the 1" copper control cables started to melt. There was an immediate emergency evacuation of the work shed plus all the other work sheds on the site. What happened? Why didn't the CBs trip? Well the 3-phase CB couldn't trip because the contacts on one phase had fused together! There was another expensive CB on the regulator itself that never tripped either. There is a long long story to this project but towards the end of my relationship with it, I started working on a smart CB that would not ever get to the point that contacts could fuse together and that would also report on the currents etc.


    btw, the buildings were all fitted with all the required smoke and fire alarms, but they never sounded. Turns out they had mounted them at the highest point, perhaps by following conventional wisdom that smoke rises, rather than actual testing. Turns out that the super-heavy duty insulation can be a bit like a tyre/tire fire and it is heavy. There was a cross breeze in the shed due to roller doors open at either end of the 50m or so length of the building and the smoke never reached the detectors!

    P.S. While I might use optos or solid-state isolaters, most of this is avoidable if you have proper routing for common grounds, keeping the heavy currents away from the signal currents, and thereby also the "spikes".
  • Clock Loop wrote: »
    Do you use an optocoupler array?
    Or just some individual ones like H11AA1 or similarm to protect only the lines going to the motor drives(transistors, mosfets, igbts, etc)?
    Got any part numbers you find as your favorite?
    For the time being I'm going to stay with 4-pin optocouplers 'and' sockets. I started with eleven SFH618A-5's from Mouser. They worked great. When I ran out I bought a batch of fifty PC817's from Ebay, which I haven't tested yet.
  • Clock LoopClock Loop Posts: 2,069
    edited 2020-08-31 23:25
    lardom wrote: »
    For the time being I'm going to stay with 4-pin optocouplers 'and' sockets. I started with eleven SFH618A-5's from Mouser.

    OPTOISOLATORS!!

    Looks like they are still available... CUT TAPE @ 10 at 0.93 each. Pricey.
    https://www.digikey.com/product-detail/en/vishay-semiconductor-opto-division/SFH618A-5X007T/SFH618A-5X007TCT-ND/7612324
    https://www.vishay.com/docs/83673/sfh618a.pdf

    lardom wrote: »
    When I ran out I bought a batch of fifty PC817's from Ebay, which I haven't tested yet.

    Those are a cheaper OPTOISOLATORS!!

    CUT TAPE @ 100 for 0.19430 each.
    https://www.digikey.com/product-detail/en/taiwan-semiconductor-corporation/TPC817S1C-RAG/TPC817S1CRAGCT-ND/7359682
    https://www.taiwansemi.com/products/datasheet/TPC817 SERIES_C1612.pdf
  • Cluso99Cluso99 Posts: 18,066
    I used to carry 30A fuses by the handful in my old 6V VW beatle.
    You see, my 6m valve rig used to blow the 30A fuse if I stayed on transmit for more that a minute or two :(
  • Clock LoopClock Loop Posts: 2,069
    edited 2020-09-01 00:39
    Cluso99 wrote: »
    used to blow the 30A fuse

    Why didn't you just wrap the fuse with kitchen foil? Fuses are useless. ;)

    My step-dad did this to get a string of Christmas lights to work again. It worked great.
    I now replace all fuses with the old fuse wrapped in foil.

    IM CONVINCED! Fuses ARE useless.

    fuse-replacement-guide-155107.jpg

  • evanhevanh Posts: 15,126
    Ah, fuses and circuit breakers are there for human safety reasons, not to keep equipment functioning correctly. Fuses still get used well for their purpose.
  • evanhevanh Posts: 15,126
    edited 2020-09-01 05:02
    Clock Loop,
    You might want to rethink keeping that last post ... and consider what can happen to the wiring if the lights go shorted when no fuse exists.
  • @evanh - I think it is a tongue-in-cheek comment about fuses and how they are perceived to be circuit-protection when in fact they are more like fire prevention. We agree that fuses have a role to play...... somewhere else.
  • Cluso99Cluso99 Posts: 18,066
    I was happy to blow the 30A fuses. Much better than killing my battery. It was limiting the power I could draw from my battery.
  • evanhevanh Posts: 15,126
    @evanh - I think it is a tongue-in-cheek comment about fuses and how they are perceived to be circuit-protection when in fact they are more like fire prevention. We agree that fuses have a role to play...... somewhere else.
    I will point out that "Circuit Protection" is an actual safety term. That's why I even looked at this topic.

  • Clock LoopClock Loop Posts: 2,069
    edited 2020-09-01 10:24
    I think it is a tongue-in-cheek

    It was, but more about how fuses are NOT only for fire prevention. (but many think that replacing a fuse is more burdensome than replacing the entire device?)
    BURN THE EARTHS RESOURCES NOW, ASK QUESTIONS LATER.
    DO NOT REPAIR. DO NOT RECYCLE. DO NOT DESIGN FOR LONGEVITY.

    Are we all MR.Muntz now?
    https://en.wikipedia.org/wiki/Muntzing
    Earlmuntz.jpg

    The role model of the 21st Century?

    burn-it-burn-it-down.jpg

    Because, in my opinion, fuses play a role in NOT ONLY fire prevention (the more obvious and commonly used purpose), but that they CAN play a role in circuit protection.
    Sure, many never design a circuit in such a way, but that doesn't mean it cannot be designed as such.

    Math seems to be an ever decreasing role in circuit design, calculating a circuits current draw, can easily help one to determine an upper limit to any given circuit, making fuse selection easy.

    Fuses are useful in not just fire prevention, but user idiocy. I made this point in my initial post about my multi-meter, they do have valid uses in situations OTHER THAN fire prevention.
    But I didn't make this thread about fuses alone..

    The P2 circuit describes the USB as protected.
    "USB protection: current-limiter and short-circuit detection"

    Ok, so it doesn't use a fuse, but it DOES make an attempt at "current-limiter and short-circuit detection".

    "When a short circuit, over-current, or over-temperature fault condition is detected by the USB protection circuit,
    power will be disconnected and the corresponding red fault LED, marked by a warning triangle label, will light until the fault is remedied."

    I will look over the schematic to see how parallax goes about doing this, but that is what I am talking about when I say "Circuit Protection", so no NOT JUST fuses.

    Onward and upward.

    Another form of "fuse" thats not a fuse...
    PPTC devices: https://m.littelfuse.com/~/media/electronics/technical_papers/resettable_ptcs/littelfuse_fundamentals_of_resettable_functionality_in_pptc_devices_technical_paper.pdf.pdf

  • I use PPTCs all the time, but always in low voltage DC supplies and also on RS485/422 signals as well, although I have been using the newer high-speed protector (HSU) for that purpose lately.

    I'm not against a fuse per se, it's the stupid glass tube thing that blows and you have go out or order the right ones, which takes time. That's why I will use PPTCs and even series resistors that have been calculated and tested to fail when the need arises. But those glass tubes and clips... yuck!

  • evanhevanh Posts: 15,126
    Circuit Protection is the safety term for fire/explosion prevention of an electrical circuit. So, they're one and the same thing.
  • evanh wrote: »
    Circuit Protection is the safety term for fire/explosion prevention of an electrical circuit. So, they're one and the same thing.

    Now what?
  • evanhevanh Posts: 15,126
    edited 2020-09-01 12:32
    Meaning it doesn't matter if the cause was a user mistake. The fuse isn't there to save the multimeter - without the fuse, the multimeter has a chance of being an explosive device.

  • Perspective my friend.

    Use the multimeter to measure a 480v 3 phase 100 amp power line, and even if you have that fuse, you will find the multimeter IS an explosive device.

    The fuse IS there to save the multimeter, the BRAIN stops it from being an explosive device.

    Still, you have made this point 3 times now, and I think the point has been made.

    Got any OTHER tips on ways to protect circuits?
  • evanhevanh Posts: 15,126
    Clock Loop wrote: »
    Use the multimeter to measure a 480v 3 phase 100 amp power line, and even if you have that fuse, you will find the multimeter IS an explosive device.
    The original fuse should be rated for more than that. The better meters are good for 1000 Vac. If you chose to not use the correctly rated fuse then sure it can fail badly.
  • Clock Loop wrote: »

    Got any OTHER tips on ways to protect circuits?

    I'll take that as a no. Thanks anyway!

  • evanhevanh Posts: 15,126
    Doesn't feel very tongue-in-cheek.
  • evanh wrote: »
    Doesn't feel very tongue-in-cheek.

    Sorta, just that the tongue was extended out between lips :)

    Must have "blown a fuse"

    Perhaps he is after some input on actual "circuit protection" rather than the semantics and terminology of it.

  • @Clock Loop, quote:
    The P2 circuit describes the USB as protected.
    "USB protection: current-limiter and short-circuit detection"
    Can you post some more info about that?

    @Peter Jakacki, what is a PPTC?
  • lardom wrote: »
    @Clock Loop, quote:
    The P2 circuit describes the USB as protected.
    "USB protection: current-limiter and short-circuit detection"
    Can you post some more info about that?

    I don't see the REV C schematic up yet, I figured it would be on the product page but it is not.
    REV A and REV B are here: https://www.parallax.com/downloads/propeller-2-es-eval-board-schematic
    I guess without the schematic, someone who owns the pcb will need to reverse it, to help you, sorry!
  • lardom wrote: »
    @Peter Jakacki, what is a PPTC?

    Personal Protection with Tongue in Cheek?

    or maybe you are referring to resettable fuses or Polymeric Positive Temperature Coefficient device.
    They have a small amount of resistance and heat up with overcurrent and then the material goes high resistance while still drawing a small amount of leakage current which keeps it warm and in the high resistance state. Once you disconnect the power or interrupt the current, the device cools down again very quickly, back to its low resistance state.

  • Clock LoopClock Loop Posts: 2,069
    edited 2020-09-01 15:36
    although I have been using the newer high-speed protector (HSU) for that purpose lately.

    Man, those things are expensive. But they look feature loaded.

    SOIC - https://www.digikey.com/product-detail/en/analog-devices-inc/LT4356IS-2-TRPBF/LT4356IS-2-TRPBFCT-ND/8549233
    https://www.analog.com/media/en/technical-documentation/data-sheets/LT4356-1-4356-2.pdf
Sign In or Register to comment.