Author Topic: Under-rated fuse question  (Read 3173 times)

0 Members and 1 Guest are viewing this topic.

Offline JesterTopic starter

  • Frequent Contributor
  • **
  • Posts: 898
  • Country: ca
Under-rated fuse question
« on: March 28, 2019, 02:30:20 pm »
I'm working on a project that requires a load resistor to be applied to the 120V Mains for 15 seconds once an hour. The circuit is in a fairly small plastic enclosure.

The temperature rise for 15 seconds once an hour is not a problem, however assuming the single fault condition that the circuit gets latched on continuously, the circuit will over heat catastrophically. The resistor is 500 \$\Omega\$, so 240ma and about 30W. If I insert a 200mA  160ma fuse in series, the fuse burn time @ 240ma is approximately 100s 200s. Assuming we can tolerate temperature rise of 30W for 200s, can I rely on the fuse to be long time reliable with 24 heating cycles a day to 15% 7.5% of the trip time?


« Last Edit: March 29, 2019, 01:03:36 am by Jester »
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 9327
  • Country: fi
Re: Under-rated fuse question
« Reply #1 on: March 28, 2019, 02:51:34 pm »
The fuse will most likely be fine in your conditions (i.e., won't nuisance blow), however, 200mA nominal fuses that guarantee blowing up on 240mA in any time are going to be rare, or unobtainium. In other words: it won't blow up in the fault condition (unless you are very lucky with the actual fuse), and your device catches fire.

You'd need to use something like a 125mA fuse to be sure enough, and then it would be a problem with your 15 second cycle.

Thermal fuse coupled with the resistor is the way to go. Adjust the coupling so that it doesn't blow in the normal case.
 
The following users thanked this post: SeanB, Jester

Offline Tomorokoshi

  • Super Contributor
  • ***
  • Posts: 1212
  • Country: us
Re: Under-rated fuse question
« Reply #2 on: March 28, 2019, 04:40:36 pm »
What type of device is turning on power to the load resistor? Can you program the duty cycle of how the device is controlled?
 
The following users thanked this post: Jester

Offline ejeffrey

  • Super Contributor
  • ***
  • Posts: 4032
  • Country: us
Re: Under-rated fuse question
« Reply #3 on: March 28, 2019, 04:52:31 pm »
I wouldn't want to rely on a fuse stressed that much long term. Can you use a thermal fuse or even a bimetallic thermal switch?
 
The following users thanked this post: Jester

Offline Neilm

  • Super Contributor
  • ***
  • Posts: 1560
  • Country: gb
Re: Under-rated fuse question
« Reply #4 on: March 28, 2019, 07:52:10 pm »
I have seen resistors that have a "pinger" on them. If the resistor gets too hot the pinger opens and disconnects the resistor. The downside for this is it is a single shot device and you would have to replace the whole resistor. Also they are not that available. A thermal fuse that is pressed up against the resistor would be a reasonable option
Two things are infinite: the universe and human stupidity; and I'm not sure about the the universe. - Albert Einstein
Tesla referral code https://ts.la/neil53539
 
The following users thanked this post: Jester

Offline JesterTopic starter

  • Frequent Contributor
  • **
  • Posts: 898
  • Country: ca
Re: Under-rated fuse question
« Reply #5 on: March 29, 2019, 12:44:15 am »
The fuse will most likely be fine in your conditions (i.e., won't nuisance blow), however, 200mA nominal fuses that guarantee blowing up on 240mA in any time are going to be rare, or unobtainium. In other words: it won't blow up in the fault condition (unless you are very lucky with the actual fuse), and your device catches fire.

You'd need to use something like a 125mA fuse to be sure enough, and then it would be a problem with your 15 second cycle.

Thermal fuse coupled with the resistor is the way to go. Adjust the coupling so that it doesn't blow in the normal case.

I looked too quickly, was actually looking along the 340ma line (one column over on the graph), however the 160ma fuse yields a probably acceptable, trip time of about about 200s

The fuse is a OTS fuse $0.43, according to the curve it will blow in  ~200s at 240ma

Link:  https://www.mouser.com/ProductDetail/Schurter/3403015924?qs=%2Fha2pyFaduj1c3MuAHqYD7OT4tNbLcvhQ0QvCGrzUp74v5TF39%252BDEQ%3D%3D

« Last Edit: March 29, 2019, 01:00:39 am by Jester »
 

Offline JesterTopic starter

  • Frequent Contributor
  • **
  • Posts: 898
  • Country: ca
Re: Under-rated fuse question
« Reply #6 on: March 29, 2019, 12:54:33 am »
What type of device is turning on power to the load resistor? Can you program the duty cycle of how the device is controlled?

There is no issue with controlling the on/off time. The resistor will only be on for 15s per hour.  I'm considering the fault case that if something fails such that the load resistor stays on due to a circuit failure, I need to make sure it fails safely. I'm pretty sure we can tolerate the heat build up until the fuse opens. I just want to make sure the fuse will not nuisance blow, when regularly (24 times per day), moving partially up the trip curve.

I guess it comes down to will the element be hotter or cooler at 240ma after 15s vs.  160ma for 10,000s

I also sent the question to Schurter.
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13156
  • Country: ch
Re: Under-rated fuse question
« Reply #7 on: March 29, 2019, 02:29:17 am »
I have seen resistors that have a "pinger" on them. If the resistor gets too hot the pinger opens and disconnects the resistor. The downside for this is it is a single shot device and you would have to replace the whole resistor. Also they are not that available. A thermal fuse that is pressed up against the resistor would be a reasonable option
i.e. the fuses used in older Keithley bench meters?
 
The following users thanked this post: Jester

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Under-rated fuse question
« Reply #8 on: March 29, 2019, 03:17:57 am »
Put a Klixon(R) in series with the load.  (Or if I misremembered or misspelled that brand, it's a thermal switch.)

There are also thermal fuses, which would help here -- but as a fuse, it's a one-and-done case, possibly more annoying than needed.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 
The following users thanked this post: SeanB, Jester

Offline Tomorokoshi

  • Super Contributor
  • ***
  • Posts: 1212
  • Country: us
Re: Under-rated fuse question
« Reply #9 on: March 29, 2019, 04:21:59 am »
Is there anything special about 30W for 15 seconds? For instance, can you do 15W for 30 seconds?
 
The following users thanked this post: Jester

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 9327
  • Country: fi
Re: Under-rated fuse question
« Reply #10 on: March 29, 2019, 07:33:24 am »
I looked too quickly, was actually looking along the 340ma line (one column over on the graph), however the 160ma fuse yields a probably acceptable, trip time of about about 200s

The curves are for "typical unit". They don't include unit-to-unit variation, which tends to be quite large. A proper datasheet should state the minimum current that guarantees a blown fuse in any time at all, and you'll see it's a larger value than what you see on this curve. This information is lacking from this cheap fuse. It's often 1.5x rated current, i.e., 240mA for your 160mA fuse, so if your load is actually 239mA, it's never guaranteed to work. Hence my recommendation for a 125mA fuse, but that could probably nuisance blow.
« Last Edit: March 29, 2019, 08:57:03 am by Siwastaja »
 
The following users thanked this post: Someone, Jester

Offline SteveyG

  • Supporter
  • ****
  • Posts: 997
  • Country: gb
  • Soldering Equipment Guru
Re: Under-rated fuse question
« Reply #11 on: March 29, 2019, 08:11:55 am »
Something like this would justify an electronic 'fuse'. Is this a possibility?
YouTube Channel: https://www.youtube.com/user/sdgelectronics/
Use code: “SDG5” to get 5% off JBC Equipment at Kaisertech
 
The following users thanked this post: Jester

Offline Ian.M

  • Super Contributor
  • ***
  • Posts: 13216
Re: Under-rated fuse question
« Reply #12 on: March 29, 2019, 09:36:08 am »
I have seen resistors that have a "pinger" on them. If the resistor gets too hot the pinger opens and disconnects the resistor. The downside for this is it is a single shot device and you would have to replace the whole resistor. Also they are not that available. A thermal fuse that is pressed up against the resistor would be a reasonable option

I remember those. They used to be fairly common in really old TVs.   A square or rectangular section ceramic bodied axial resistor, mounted vertically, either had a springy steel strip in a T slot up the side of the body with the top bent over in a loop and soldered to the top resistor lead which was a short stub, or there was a formed steel sheet cage round the resistor body with a similar springy strip at the top end.  The steel was typically tin or copper plated for solderability.  The solder joint to the resistor lead was made with low melting point alloy, and if you had to repair it after it had disconnected, it was essential to use a freshly cleaned soldering iron bit with flux but no extra solder so you didn't significantly change the alloy composition.
« Last Edit: March 30, 2019, 01:47:50 am by Ian.M »
 
The following users thanked this post: Jester

Offline Ice-Tea

  • Super Contributor
  • ***
  • Posts: 3229
  • Country: be
    • Freelance Hardware Engineer
Re: Under-rated fuse question
« Reply #13 on: March 29, 2019, 09:44:46 am »
What's keeping you from putting a PTC in series?
 
The following users thanked this post: Jester

Offline JesterTopic starter

  • Frequent Contributor
  • **
  • Posts: 898
  • Country: ca
Re: Under-rated fuse question
« Reply #14 on: March 29, 2019, 01:36:16 pm »
Is there anything special about 30W for 15 seconds? For instance, can you do 15W for 30 seconds?

Special because it's in the specification, so no it has to be 30W for 15s
 

Offline ejeffrey

  • Super Contributor
  • ***
  • Posts: 4032
  • Country: us
Re: Under-rated fuse question
« Reply #15 on: March 29, 2019, 02:41:28 pm »
PTC trip characteristics are much worse than metal fuses. Unless you mean to mount it to the load resistor and turn it into a thermal cutoff.  A thermal fuse or cutoff switch seems like the right solution for this application based on the information we have.
 
The following users thanked this post: Jester

Offline Ice-Tea

  • Super Contributor
  • ***
  • Posts: 3229
  • Country: be
    • Freelance Hardware Engineer
Re: Under-rated fuse question
« Reply #16 on: March 29, 2019, 03:22:57 pm »
PTC trip characteristics are much worse than metal fuses. Unless you mean to mount it to the load resistor and turn it into a thermal cutoff.  A thermal fuse or cutoff switch seems like the right solution for this application based on the information we have.

I know PTCs suck to dimension right. But it might still be valid option in case resetability is an issue or such events are expected on a regular basis.

Put a Klixon(R) in series with the load.  (Or if I misremembered or misspelled that brand, it's a thermal switch.)

Tim

Airpax?
 
The following users thanked this post: Jester

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16385
  • Country: za
Re: Under-rated fuse question
« Reply #17 on: March 29, 2019, 04:33:55 pm »
Klixon self resetting or manual reset unit,or thermal non resettable cut out attached to the resistor. The one time thermal fuse is easy to attach with a small slide on sleeve of silicone coated fibreglass sleeving, and will both provide insulation and good thermal coupling. you typically choose a thermal fuse of around 110C, so that it will not fail in normal use, with the resistor typically reaching 70C surface temperature in the 15 seconds of on, and with the heat soak after. Longer time on it will reach the 110C point and the thermal cutout will melt, most likely after around 5 minutes.

A fuse is not good, it will be replaced with another of different rating, or will be replaced with one not of the same brand, with different trip characteristics, or will be wrapped in foil. You can get small thermal cutouts that will provide protection that are the same size as the one time unit, but they need the non resettable unit as back up as well. Thermal fuse is both rugged, reliable ( they will melt at rated temperature, with up to 5C below it as margin, and will never fail to operate) and cheap, plus needs minimal extra protection, other than a regular fuse to protect the rest of the circuit and wiring from faults, probably here a 1A slow blow fuse.
 
The following users thanked this post: Jester

Offline JesterTopic starter

  • Frequent Contributor
  • **
  • Posts: 898
  • Country: ca
Re: Under-rated fuse question
« Reply #18 on: March 30, 2019, 12:23:32 am »
Thanks for the comments,  I decided to go with a thermal cutout, more expensive than a fuse, however it should be more reliable. 
 

Offline Neomys Sapiens

  • Super Contributor
  • ***
  • Posts: 3268
  • Country: de
Re: Under-rated fuse question
« Reply #19 on: April 03, 2019, 09:07:08 pm »
I would not recommend to make it a self-resetting one, as some are suggesting. If overheating happens (by your description), there was a violation of the control signal, which should be investigated/corrected!
The other possibility is to rebuild the device in such way, that it can tolerate being switched on continuously without overheating.
 
The following users thanked this post: Siwastaja, Jester

Offline OM222O

  • Frequent Contributor
  • **
  • Posts: 768
  • Country: gb
Re: Under-rated fuse question
« Reply #20 on: April 03, 2019, 09:21:12 pm »
have you thought about making an electronic fuse? you can use a triac (solid state relay) which is controlled by a microcontroller. should cost you less than 5$.
 

Offline johnkenyon

  • Regular Contributor
  • *
  • Posts: 123
  • Country: gb
Re: Under-rated fuse question
« Reply #21 on: April 04, 2019, 09:38:55 am »
I would not recommend to make it a self-resetting one, as some are suggesting. If overheating happens (by your description), there was a violation of the control signal, which should be investigated/corrected!
The other possibility is to rebuild the device in such way, that it can tolerate being switched on continuously without overheating.

I second this - the intention is for the device to fail safe and remain failed until fixed.

Having something auto reset will inevitably result in something else getting cooked and failing in a more destructive manner...
 
The following users thanked this post: Jester

Offline JesterTopic starter

  • Frequent Contributor
  • **
  • Posts: 898
  • Country: ca
Re: Under-rated fuse question
« Reply #22 on: April 04, 2019, 03:25:13 pm »
I would not recommend to make it a self-resetting one, as some are suggesting. If overheating happens (by your description), there was a violation of the control signal, which should be investigated/corrected!
The other possibility is to rebuild the device in such way, that it can tolerate being switched on continuously without overheating.

I second this - the intention is for the device to fail safe and remain failed until fixed.

Having something auto reset will inevitably result in something else getting cooked and failing in a more destructive manner...

I agree, I used this one:  https://www.digikey.com/product-detail/en/cantherm/SDF-DF100S/317-1129-ND/1014758
 

Offline JesterTopic starter

  • Frequent Contributor
  • **
  • Posts: 898
  • Country: ca
Re: Under-rated fuse question
« Reply #23 on: April 04, 2019, 03:28:39 pm »
have you thought about making an electronic fuse? you can use a triac (solid state relay) which is controlled by a microcontroller. should cost you less than 5$.

Not enough space, and total BOM cost is about$10, so $5 is significant.
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 9327
  • Country: fi
Re: Under-rated fuse question
« Reply #24 on: April 04, 2019, 04:13:42 pm »
Also, protecting against MCU software malfunction by another MCU, possibly written by the same person, is less robust than using the proper temperature cutout device, which Does The Right Job very reliably, when chosen properly.

Use the time saved by not implementing a separate e-fuse device to review the actual controller a few times more; then slap in the proper thermal fuse (single-use, non-resettable, to get failed products back for your review for root cause analysis) as a safeguard. This is also the cheapest proper solution.

Even if precision, low unit variation fuses existed, they wouldn't protect when the input voltage (and hence resistor current) is lower than expected.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf