Simple (?) redundant power question

[q9]

Hi there everybody

Probably a dead simple question, but anyhow, here it is.

What I would like to do is to add a circuit to my motorbike headlight, so that if the power is cut, say from the fuse burning out, the new circuit will cut in and keep the light lit. Had a bit of a fright last night, you see :shock:

I know how to hook up a relay to do this, but I don't know how to get the relay to kick in when the power from the other circuit drops out. My initial thoughts are that I need a transistor that is set up to block power going to the relay switch, but I am not sure if that will work, nor am I sure if that is the best option. Ideas would be welcome!

Thanks.

 

[akg]

well.. this is not an answer to ur question..
but ..if the fuse blows out(some problem occured)..it is better to look into the reason and to fix it..than using an additional ckt..

 

[JimB]

A bit difficult.
Presumably you have:
One battery,
One fuse,
One switch,
One headlight.
The fuse could have blown for twp possible reasons:
Something (the headlamp) took too much current.
A random failure (It does happen from time to time).
So if the proposed second fuse circuit provided power to the headlight which was taking too much current for some reason, that fuse would also blow.
If you are that concerned, a much better scheme would be to add a second headlight fed via a separate circuit.

JimB

 

[stevez]

Some critical circuits in some vehicles are equipped with circuit breakers that automatically reset and do so quickly (seconds, not minutes). The only improvement that this offers is if the overcurrent condition was momentary.

As others have said, your backup circuit would only be useful if the fault were in the wiring from the source to the point where the backup ties in - otherwise it's going to see the same condition.

 

[q9]

I understand what you are saying, however, the second circuit will be routed differently through the bikes frame. So if the first circuit shorted out because the wire rubbed through the insulation against the frame I would expect the fuse to blow. The second circuit should then kick in. This circuit would probably also find the short, meaning I have a problem that needs fixing.

However, if the fuse just blows from age/vibration (it happens), or in the case as the other night, the actual fusing housing plastic (nylon) melted around the fuse contacts, which broke the circuit, rather than shorting it.

I have replaced the fuse housing with a bakelite housing and used Earth B.P. connectors to join the wires. So this is more an excercise in "how could I do it" rather than anything else.

Thanks for your replies though.

 

[Klaus]

I know how to hook up a relay to do this, but I don't know how to get the relay to kick in when the power from the other circuit drops out. My initial thoughts are that I need a transistor that is set up to block power going to the relay switch, but I am not sure if that will work, nor am I sure if that is the best option. Ideas would be welcome!

Thanks.

the simplest way is to use a relay in reverse. Your normal light circuit also powers a relay with normally closed contacts, eg. they are held OPEN when the normal light circuit is on.
When that drops out the relay also drops out, closing the contacts to power the light from a different circuit.

You have to figure out a way for the closed contacts not to feed power to unwanted places when the light switch is off. Unless you duplicate the light switch you still have no redundancy for that failing, nor do you have redundancy for a blown head light globe unless you power Hi and Lo beam from a different source (fuse).

Come to think of it, rather than complicating the existing wiring, how about upgrading it to make sure shorts are a very rare occurence.
Is it a British or Italien bike? -ducking for cover now-

Scary, isn't it, when the head light suddenly goes out :shock:
Klaus

 

[q9]

Hehehe...good one Klaus!

It is a 23 year old Yamaha, actually :wink:

I actually thought it would be a fairly simple thing to do - respond to a cut in power on one circuit. I think I need to get my electronics gear down here so I can have a play with things...it's the grown ups Lego!

 

[Nigel Goodwin]

Hehehe...good one Klaus!

It is a 23 year old Yamaha, actually :wink:

I actually thought it would be a fairly simple thing to do - respond to a cut in power on one circuit. I think I need to get my electronics gear down here so I can have a play with things...it's the grown ups Lego!

Unfortunately it's NOT simple, because you are effectively bypassing the fuse - so there's no point in the fuse being there! - it's no different than replacing the fuse with a nail!.

It's VERY, VERY rare for a fuse to fail without a reason, so you need to find the reason, NOT short the fuse out with a relay!

 

[q9]

Not quite Nigel. The fuse didn't actually fail in this instance. It was an inappropriate selection of fuse holder that actually melted nylon over the contacts, breaking the circuit.

But what I am trying to achieve is a seperate, differently routed circuit that can kick in when, and only when the primary circuit fails. I suppose if I want to be tricky, I could also disconnect the primary circuit close to the headlight bulb, so that in the case of a short in the wire, it can't short the second circuit also (would a diode do the job here? and would I get a diode to handle that kind of current?).

But I have no idea how exactly to do it...

 

[Nigel Goodwin]

Not quite Nigel. The fuse didn't actually fail in this instance. It was an inappropriate selection of fuse holder that actually melted nylon over the contacts, breaking the circuit.

But what I am trying to achieve is a seperate, differently routed circuit that can kick in when, and only when the primary circuit fails. I suppose if I want to be tricky, I could also disconnect the primary circuit close to the headlight bulb, so that in the case of a short in the wire, it can't short the second circuit also (would a diode do the job here? and would I get a diode to handle that kind of current?).

But I have no idea how exactly to do it...

By introducing extra elements in the lighting circuit, you're going to substantially reduce it's reliability - which presumably isn't what you want?. It's no good introducing diodes in the circuit, the last thing you want to do is make your lights dim.

Headlight circuits are simple and reliable, in years of biking I never had one fail!.

It sounds like the only reason for this failure is that it wasn't designed correctly? - simply design it correctly and you should have no problems?.

 

[eblc1388]

The fuse didn't actually fail in this instance. It was an inappropriate selection of fuse holder that actually melted nylon over the contacts, breaking the circuit.

Doggy fuse holder. You've said it yourself, with 100% clarity. Just use a proper fuse holder and the incident will not happen again.

Most would agree the component with the highest probability to fail is the lamp filament itself. It might even fail to a short circuit and take out the circuit fuse too. How can you cater for that scenario?

Backing up everything is usually possible, but many are just not feasible.

 

[q9]

You are all correct about my situation. It has been fixed with the new fuse holder. Nothing more to do. Headlight worked to and from work yesterday (45km each way). The only problem is now I have two different fuse sizes. I think I might spend $14 and buy a new fuse box and maybe even buy some new wire and rewire the bike.

But...

Just as an excercise, I'd like to know how to do what I asked to do...I doubt very much that I will in fact implement it on the bike, it was just a simple thought for an excercise...something to keep me amused on a rainy weekend...

 

[eblc1388]

Just as an excercise, I'd like to know how to do what I asked to do...I doubt very much that I will in fact implement it on the bike, it was just a simple thought for an excercise...something to keep me amused on a rainy weekend...

If you insists.

The analysis is very easy and straight forward. You consider a blackbox with +12V as your input and "light" as your output. Then you examine every element inside the blackbox to see what effect it has on the output if it fails. After that, you either do a design to make sure that failure mode cannot happen, or reduces its effect to acceptable consequence or put in backup/duplicate circuit or whatever.

As an example in your case, the elements are:

a. the cable from +12V to fuse holder
-open circuited
-shorted to other cable
-shorted to bodywork

b. fuse holder itself
-open circuited
-bad contact
-short circuited

c. fuse element itself
-open
-bad contact inside element

d. the cable from fuse holder to lamp switch
-open circuited
-shorted to other cable
-shorted to bodywork

e. the lamp switch itself
-open circuited
-intermittent contacts
-contacts stuck together

f. the cable from switch to lamp
-open circuited
-shorted to other cable
-shorted to bodywork

g. the lamp bulb holder
-open circuited
-shorted
-intermittent, bad contact

h. the lamp bulb itself
-filament open
-filament open and short across internal connection, high current

I'll let you work out the consequence and the possible remedy.

In fact, it is not difficult to see that the only solution to all the above is a completely duplicated setup(including the lamp) with separate cable run.

 

[q9]

This has become a lot more about the light and the bike than I ever expected...

If I had said I want a seperate event to occur, say, a buzzer to sound when the circuit failed, I think maybe I would have gotten somewhere :?

What I am really asking, is how do I get a relay to turn something on in response to no voltage? It is easy to make a relay turn on by applying power to it - that is simple. All I really REALLY want to know, is how to make some kind of switch that turns on when power somewhere else turns OFF.

Forget the light. Forget the bike. I am simply trying to learn a bit more about this stuff...

 

[q9]

woopsies! doubled up...

 

[Nigel Goodwin]

What I am really asking, is how do I get a relay to turn something on in response to no voltage? It is easy to make a relay turn on by applying power to it - that is simple. All I really REALLY want to know, is how to make some kind of switch that turns on when power somewhere else turns OFF.

Forget the light. Forget the bike. I am simply trying to learn a bit more about this stuff...

A simple relay will do it, with change over contacts, but that would put your lights permanently ON as well.