Blown Fuse Indicator

Thread Starter

mcgyvr

Joined Oct 15, 2009
5,394
Need some help as my brain seems to have turned off since my wife and I had our first child (lack of sleep I'm sure)
I need a blown/good fuse indicator circuit..

Supply voltage is -24VDC
Circuit needs to have the following functionality..

-Load Present + Fuse intact = green LED on
-Load Present + Fuse blown = red LED on
-No load present = no LEDs on and voltage at output of fuse less than 5VDC
 

Thread Starter

mcgyvr

Joined Oct 15, 2009
5,394

kubeek

Joined Sep 20, 2005
5,794
You would need to add a load of your own, so that there is at least some current flowing. Could be very small load, say 100uA to generate a votlage difference at the open fuse.
 

Thread Starter

mcgyvr

Joined Oct 15, 2009
5,394
Can you clarify/rewrite
-No load present = no LEDs on and voltage at output of fuse holder less than 5VDC

I stated that as I suspect some sort of "load sensing" circuitry is required.. (like I believe kubeek was alluding to,etc...)
I want to ensure that if one were to throw a multimeter on the output (unloaded/no fuse in place) that they don't see full line voltage (sensing) and think a fuse is in place or something else is wrong..
 
Last edited:

Thread Starter

mcgyvr

Joined Oct 15, 2009
5,394
Any additional help would be great...
I'm banging my head against the table now so anything would be an improvement over that :)
 

Thread Starter

mcgyvr

Joined Oct 15, 2009
5,394
Bump..
I'm still quite unsure how to tackle this..
Would looking attempting to use a comparator to judge about the presence of a load/bad fuse be the direction to go here?
 

Thread Starter

mcgyvr

Joined Oct 15, 2009
5,394
The simple one would be the fuse blown - load present, a LED across the fuse.
Max.
Yes.. That I've got..
I can easily do the first one also..
But I need to check all 3 boxes..

I just can't figure out how I construct a circuit to know a load is present or not without causing someone to see line voltage at the output when no load is present..
I've never used a comparator before and I'm not sure how it influences a circuit (I'm playing around in TINA right now)..
Nor if thats even the right direction..
 

ebeowulf17

Joined Aug 12, 2014
3,307
Yes.. That I've got..
I can easily do the first one also..
But I need to check all 3 boxes..

I just can't figure out how I construct a circuit to know a load is present or not without causing someone to see line voltage at the output when no load is present..
I've never used a comparator before and I'm not sure how it influences a circuit (I'm playing around in TINA right now)..
Nor if thats even the right direction..
I think I've got a loose concept - block diagram logic but not actual circuit components. No time for a decent sketch right now, but I'll try to post something this evening.

Two questions:
1) If the fuse blows and there is a load connected, is it ok to pass a small test current through the load, as opposed to truly and completely breaking the circuit like a fuse normally does?

2) Is it safe to assume the load would always draw some small, measurable amount of current if it were fed up to 5V through a current source? What is the nature of the load? Nothing weird, like an SMPS or something that might shut itself down if under powered?
 

ebeowulf17

Joined Aug 12, 2014
3,307
Ok, like I said before, I don't have any of the specific components or sub-circuits worked out, but I think I may have a concept that could meet your requirements. If this looks at all like what you had in mind, I can try to work on building sub circuits and simulating it, but it may take me a while. Not my strong suit, but it seems like a fun challenge for me, and I'm intrigued by it.
upload_2018-10-9_22-14-11.png
 

Attachments

Thread Starter

mcgyvr

Joined Oct 15, 2009
5,394
I think I've got a loose concept - block diagram logic but not actual circuit components. No time for a decent sketch right now, but I'll try to post something this evening.

Two questions:
1) If the fuse blows and there is a load connected, is it ok to pass a small test current through the load, as opposed to truly and completely breaking the circuit like a fuse normally does?

2) Is it safe to assume the load would always draw some small, measurable amount of current if it were fed up to 5V through a current source? What is the nature of the load? Nothing weird, like an SMPS or something that might shut itself down if under powered?
First.. thanks.. I appreciate the effort you have put in..
While the concept you have provided may achieve the goals I'm not sure its a valid direction for this..
It has certainly given me some help though in seeing one solution and helping spur more ideas..

To answer your questions..
1)The current circuit I have which has been acceptable up until now just has a transistor across the fuse driving an LED (that circuit is actually built into the fuse holder itself) so yes there is some current being drawn by that to turn on the transistor and enable the red led.
2)The load is unknown or should I say could be anything drawing 30A or less..
This is the current schematic being used that is no longer acceptable.. (because sufficient voltage is measurable at the output of the fuse with no fuse present)
Cir1.PNG
 

ebeowulf17

Joined Aug 12, 2014
3,307
First.. thanks.. I appreciate the effort you have put in..
While the concept you have provided may achieve the goals I'm not sure its a valid direction for this..
It has certainly given me some help though in seeing one solution and helping spur more ideas..

To answer your questions..
1)The current circuit I have which has been acceptable up until now just has a transistor across the fuse driving an LED (that circuit is actually built into the fuse holder itself) so yes there is some current being drawn by that to turn on the transistor and enable the red led.
2)The load is unknown or should I say could be anything drawing 30A or less..
This is the current schematic being used that is no longer acceptable.. (because sufficient voltage is measurable at the output of the fuse with no fuse present)
View attachment 161268
Well, if you didn't need the light indicating when there's no load connected, I think I have a simple solution for indicating a blown fuse without letting too much voltage show on the other side of a fuse. Switch from BJT to MOSFET, and add a Zener and a few resistors to protect the MOSFET gate. With an 18V Zener keeping gate voltage from exceeding 20 (a common limit if it's not logic-level,) the voltage seen on the other side of a blown fuse never exceeds 5-6V, even if there's no load connected.
Fuse-Load-Detection_04.png

*** EDIT: I see @Kjeldgaard posted while I was typing, and that solution looks like it may be simpler and/or have less leakage than mine. Anyway, plenty of solutions that are not too difficult if you don't need missing-load detection!
 

Attachments

Thread Starter

mcgyvr

Joined Oct 15, 2009
5,394
Well, if you didn't need the light indicating when there's no load connected, I think I have a simple solution for indicating a blown fuse without letting too much voltage show on the other side of a fuse. Switch from BJT to MOSFET, and add a Zener and a few resistors to protect the MOSFET gate. With an 18V Zener keeping gate voltage from exceeding 20 (a common limit if it's not logic-level,) the voltage seen on the other side of a blown fuse never exceeds 5-6V, even if there's no load connected.

*** EDIT: I see @Kjeldgaard posted while I was typing, and that solution looks like it may be simpler and/or have less leakage than mine. Anyway, plenty of solutions that are not too difficult if you don't need missing-load detection!
And again..
I don't have enough knowledge to know if I can trust the simulation or not..
Here I still see 13+VDC at the output
ebeo2.PNG
 

Danko

Joined Nov 22, 2017
1,829
Is there something incorrect with this simulation?
I'm still seeing 21.08V on the output with no load/no fuse..
Of course you should see voltage in simulation because leakage current of diode D1 is about 0.00000001A.
Short line after fuse by resistor 100k.
 

Attachments

Last edited:

ebeowulf17

Joined Aug 12, 2014
3,307
And again..
I don't have enough knowledge to know if I can trust the simulation or not..
Here I still see 13+VDC at the output
View attachment 161274
That's bizarre. I don't have enough knowledge to know when to trust the simulation either. Sometimes I think I've done everything right and discover that I failed to model some real-world thing like a parasitic leakage path and that my circuit only works in an impossibly perfect world. So, I've definitely fooled myself with simulations before.

Having said that, even without the simulation, I feel like I understand this circuit pretty well, and I don't see how it can be that far off. My best guess is that the Zener characteristics vary from part to part (we're simming with different Zeners, and I don't have a model for yours to try it out.) If I I increase R1 to really extreme levels (10-100Meg) the voltage at the fuse goes up. Maybe with a different Zener it requires more current to behave normally. You could try other 18V Zeners if you've got them, or try decreasing R1 to allow more current and see if the fuse voltage drops.

You could also skip the Zener entirely if you chose a MOSFET that can tolerate >24Vds.

Beyond that, I'm at a loss. I feel fairly optimistic that this should work just like it does in my LTspice simulation, but I'm not totally sure.
 

Thread Starter

mcgyvr

Joined Oct 15, 2009
5,394
All.. I got some excellent help here today and I appreciate it..
I think I have a suitable path to follow to achieve my needs..
Thanks again..
 
Top