EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: mobbarley on May 15, 2011, 02:56:45 am
-
I am trying to come up with something simple and reliable for a fuse blown indicator. For a simple LED indicator I could easily do this, but I have at least 10x fuses to monitor and I would like to provide the LED as well as a single external output for any blown fuses.
-
Something such as this - but this will cause problems when connecting up multiple outputs and I don't want an optoisolator for each.
I have approx 10 outputs that will run at 24-32vdc.
-
I looked into this many years back, i cant remember all the details but the best idea i came up with was to use a little isolated dc/dc to get a separate supply, isolated from the main supply, to drive the optos or mcu and provide the monitoring.
-
Well I think I have found my answer by placing the fuse on the low side of the load - this allows me to use a diode and common ground. However the dissipation at 32v is still going to be higher than I would like for a simple indicator so I might have to look into some kind of driver. ???
-
You start from the wrong design principle - to light the LED when the fuse fails. The right principle would be to turn off the LED when the fuse fails. The later also covers the case when not the fuse but the LED fails.
The circuit also become almost trivia when you use "LED off" to indicates failure.
-
You start from the wrong design principle - to light the LED when the fuse fails. The right principle would be to turn off the LED when the fuse fails.
You're right.
The obvious solution is to connect the LED in parallel with the load.
The only valid reason for connecting the LED in parallel with a fuse would be to simplify the wiring.
-
If you connect the led onto the load you will be unable to tell the difference between a blown fuse and the device being intentionally off.
It's quite common to put a led+res across a fuse so it lights when the fuse blows, you see it in some car fuses.
eg
(http://www.northernautoparts.com/Images/Uploaded/Wirthco/Illuminating_led_fuses/ATM_LED_illuminating_Fuses_Light_Dark.jpg)
-
If you connect the led onto the load you will be unable to tell the difference between a blown fuse and the device being intentionally off.
Oh yes, I didn't think of that, LOL
All right, just put another LED before the fuse, as a power indicator, as well as one in parallel with the load.
-
I've decided with LED across the fuse and i'm okay with that.
-
So once again I wasted my time with people who don't want to know how it is done.
-
imho, this is a good concept. i only concern whether if its counterfeit the purpose of the fuse, ie when something goes wrong (shorted load and hi volt), blown fuse will isolate the load/circuit from power source.
edit: and think of green energy, >99% ok (working) only <1% blown. ie with conventional design, the LED will be 99% ON which a waste from "green energy point of view"
-
If you want to be able to detect a blown fuse without any power on VCC you need a separate isolated supply, ie.. if you want a system that can check all fuses before applying power.
Otherwise the best approach is to connect the opto led between the low side of the fuse and ground.
It inverts the logic but then the output of the opto transistor inverts it back, so you end up with an output that goes high if the fuse fails.
You can use that to drive a visible led and provide an output to signal a fuse is bad.
Probably want some sort of spike protection on the opto led tho, if the motor is large it may put out quite a bit of back emf and could damage the opto led.
-
err. i just look at reply/design #3. placing fuse on the earth side is not really a good idea. during blown fuse, the circuit still exposed to live wire. and that is very wrong for me, esp when somebody dont expect the faulty fuse and try to debug the circuit.
-
err. i just look at reply/design #3. placing fuse on the earth side is not really a good idea. during blown fuse, the circuit still exposed to live wire.
That depends on whether positive or negative side is earthed or the entire circuit could be floating, in which case it makes no difference.
-
So once again I wasted my time with people who don't want to know how it is done.
This is because your opinion that my entire idea is wrong is worthless. Yes most power supplies have a 'fuse ok' led - this is not a power supply and I require a very simple 'fault condition' led.
Thank you for wasting your time.
-
err. i just look at reply/design #3. placing fuse on the earth side is not really a good idea. during blown fuse, the circuit still exposed to live wire. and that is very wrong for me, esp when somebody dont expect the faulty fuse and try to debug the circuit.
Sure - except this isn't mains and it isnt earth referenced, it also isnt in a car where the chassis is ground. Its low voltage (24DC) and fairly low current <1A and there are no other ways to ground except through the wire connected to it - the ~5mA for a led flowing through a faulted load is also acceptable and won't present a hazard.
-
Sure - except this isn't mains and it isnt earth referenced...
just dont make it as a habit. as other said, its not a good "practice". its ok if its only for hobby. but i dont think it will be passed by white haired (seasoned) engineers folk at the standard certification/approving agency.
-
It's possible to do safely with mains, a 4.7nF Y2 rated capacitor wired in series with a neon lamp will probably be enough to satisfy the safety regulations.
-
Thanks for confirming that you only came here to waste our time and that you are incapable of learning.
-
Thanks for confirming that you only came here to waste our time and that you are incapable of learning.
Personally I see the merit in the concept of having the led lit on failure and a couple of examples were already pointed out in the thread of some benefits.
You didn't give him a circuit to do what he wanted, you told him to do something different which may not be appropriate to his application but he obviously wants to do as he described.
I dont see that as any reason to get upset.
In fact, I am doing the same thing (nearly) with 48 indicator warning lights. I don't want them all on so I have to figure out if one is off and there is a problem. Obviously the best way is to have them all off and only turn on in the case of a malfunction. Surely the circuit is dictated by the application, not the other way around?
-
Dear mobbarley:
--While I do not feel competent to comment on all of the electronic issues presented in this thread, I do know something about designing custom devices. I also know something about sticking to your guns as long a possible. No doubt if you were designing a door bell, some contrarian would surface, saying "Why don't you just get a door knocker" A person likes to get advice on how to do it, advice on how not to do it, no so much.
--With regard to the comments made by Boreatwork:
1) He stated:
"You start from the wrong design principle"
--Horse feathers! A man once asked for directions from a down easter Yankee. The Yankee told him "If I was agoin' theyah I wouldn't staht from heah" Same sort of helpful advice, yes?
2) He stated:
"The right principle would be to turn off the LED when the fuse fails. "
--"When he says "The right principle" I think he just means "easier and simpler", "right principle" indeed, Good Lord. Notice he restates the problem so that now we are only dealing with 1 led and not 10. If you have 10 lights lit up an one goes out you might not notice. But if you have no lights lit, and one lights up, you are more likely to notice. Admittedly his idea is simpler, and is a good fall back position if nothing else will work, but I do not think simplicity is the only goal here. Too bad those car fuse guys started from the wrong design principle, no?
3) He stated:
"The later [sic] also covers the case when not the fuse but the LED fails."
--Clumsy syntax. I think he meant to say "latter" which rhymes with ratter, rather that "later" which rhymes with "baiter" With an average lifetime of 20 to 50 thousand hours, I do not think LED failure is a significant worry here, but just for the sake of argument, let us pretend that is is. An LED is bloody well less likely to fail if it is not on all the bloody time.
4) He stated:
"The circuit also become almost trivia [sic] when you use "LED off" to indicates [sic] failure. Perhaps he started from the wrong grammar and spelling principle. One assumes he meant to say "trivial" instead of "trivia", but it is by no means clear.
5)He states:
"I love the sound of a [sic] "you are not helpful" whining in the morning"
--Again with the peculiar grammar. Let us not be unkind, perhaps English is not his first language. In this statement there may be a key to his personality, more's the pity.
6) In his second post he states:
"So once again I wasted my time with people who don't want to know how it is done."
--Where he says "people" I think he meant to say "a person". Too bad those car fuse guys "don't want to know how it is done." They could have had all of the fuses lit up all the time. Kind of like leaving all the lights in your house on so you will know if one burns out.
7) He states in his third post:
"Thanks for confirming that you only came here to waste our time and that you are incapable of learning."
--An entire sentence with no grammar or spelling errors, it a new record. This remark seems to be of a personal nature, and betrays some animus on the part of the speaker. Now you are wasting "our" time, but I do not recall anyone else accusing you of wasting our time. Surly among Gentlemen (and Ladies) advice is freely given and may be freely declined. No need to "tear a passion to tatters, to very rags" as the Bard of Avon said.
--And so Brother mobbarley, I hope you can see by the posting that I agree with almost all of the comments, made by almost everyone in this thread.
--I do have one small design idea for you to consider. You could use some of those peeping annunciators instead of LEDs, for an aural rather than a visual indication. You would have them all peeping away constantly, then when a fuse blows, you would know because one of them would stop peeping. I hope I am using the right design principle here and making the circuit almost trivia.
Best Regards
Clear Ether
-
You could put in a test function to test that all the led's are working when you push the test button
-
There is such a thing as an alarm fuse.
http://www.daitotusin.co.jp/english/contents/c_j/alarm_index.htm (http://www.daitotusin.co.jp/english/contents/c_j/alarm_index.htm)
-
There are also circuit breakers which have the option for auxiliary contacts which close/open when the breaker is tripped.
If you want an alarm you could connect a piezo buzzer in parallel with the fuse. ;D