Hi guys. I built this circuit in attempt to indicate voltage spikes. the fuse is 750mA, the BJT is PNP type, and the zener is 16vdc.
the theory is that when when the fuse blows the transistor would turn on and keep the LED lit indicating there was a voltage spike and with the fuse intact it would keep the transistor off.
i built the circuit and tested it with the fuse in and with the fuse out and the LED stayed on both times.
any opinions to why? (circuit is attached)
HI,
Assuming:
The horizontal 10K is much too high. With the fuse in, the vertical resistor will allow base current to flow from the base causing the LED to go on.
I would move the horizontal 10 K from the zener to the emitter and make it 1k.
BTW
you should always use reference designations (R1, R2 etc) on your circuits you post. Also specify the input voltage.
You should look at I2T rating (as in I * I * T) of the fuse. They can take terribly long to fuse (blow open) in some cases.
JohnRob:
HI,Assuming:
- The input is on the upper left of your schematic and it is < 16V
- The Emitter of the PNP is connected to the input / fuse
The horizontal 10K is much too high. With the fuse in, the vertical resistor will allow base current to flow from the base causing the LED to go on.
I would move the horizontal 10 K from the zener to the emitter and make it 1k.
BTW
you should always use reference designations (R1, R2 etc) on your circuits you post. Also specify the input voltage.
You should look at I2T rating (as in I * I * T) of the fuse. They can take terribly long to fuse (blow open) in some cases.
normal input voltage is between 10vdc and 14vdc.
can I trouble you to post a schematic of what you described please?
Sorry, I'm not in a place I can draw and post. However it is very simple.
Looking a your drawing:
Move the left most lead of the horizontal 10k to the other side of the fuse. And make it a 1k.
If you can modify your sketch I can verify it.
BTW: Do you have a target spike you are looking to sense? The circuit as conceived will require a substantial "spike" to trip. It will have to be both high and long. In addition, not knowing what the zener diode is, it might fail before the fuse.
John
Hi,
OPs diagram.
Thanks.. Tom... 
I don't see how this works. the fuse or the zener will blow if the voltage is too high.
Allan
allanhurst:
I don't see how this works. the fuse or the zener will blow if the voltage is too high.Allan
The zener acts as a crowbar protection device, it should be able to conduct at least the current that the fuse is rated at.
When a spike occurs that is long enough, current flows through the zener and causes the fuse to blow.
Tom....
It would have to be a pretty big zener. The standard crowbar circuit for this purpose uses a SCR as the high current dump.
MOV's and 'Tranzorbs' (do GE still make them? ) are also good.
Allan
JohnRob:
Sorry, I'm not in a place I can draw and post. However it is very simple.Looking a your drawing:
Move the left most lead of the horizontal 10k to the other side of the fuse. And make it a 1k.
If you can modify your sketch I can verify it.
BTW: Do you have a target spike you are looking to sense? The circuit as conceived will require a substantial "spike" to trip. It will have to be both high and long. In addition, not knowing what the zener diode is, it might fail before the fuse.
John
wont moving the resistor as mentioned keep the transistor "off" all the time?
TomGeorge:
Hi,
OPs diagram.
Can you edit your diagram and show;
- Input point.
- Output point.
- Transistor type and number.
- Label, base, collector and emitter.
- Mark what the fuse is rated at.
Thanks.. Tom...
transistor 2N3906
Fuse 750mA
zener P6KE16CA
the input is normally about 14vdc. and the out put would be that there ground symbol.
DudeGuy:
Hi guys. I built this circuit in attempt to indicate voltage spikes. the fuse is 750mA, the BJT is PNP type, and the zener is 16vdc.the theory is that when when the fuse blows the transistor would turn on and keep the LED lit indicating there was a voltage spike and with the fuse intact it would keep the transistor off.
i built the circuit and tested it with the fuse in and with the fuse out and the LED stayed on both times.
any opinions to why? (circuit is attached)
Did you look at the specs for the fuse? How long will it carry >750 ma before it opens up? I sincerely doubt that would be called a voltage spike. Fuses are really, really slow to respond to voltage spikes. Fuses are for current spikes and I doubt the zener will convert a "spike" to a current for enough time to pop the fuse.
The exception are fuses designed to carry a specific current and then open at perhaps 1% over that amount. Experience says they are quite expensive.
I think you need to look to a "crow bar" circuit using an SCR to pop the fuse.
Paul
I had a chance to sketch the ckt out after my post and my suggestion will not work. You are correct, the LED would not switch as you desire.
I've attached what I believe to be a functional sketch. The key is R2 now draws base current when the fuse is open.
BTW I agree with Paul in that this is not how I would approach a spike problem.
However my answer is valid to your question as I don't know you reasoning for actual issue/goal.
John
@ Paul
this circuit doesn't really protect anything at the moment, I only want to see if there is any voltage spikes. I'm aware that the fuse is for current. I just wanted to see how it would work.
I originally wanted to use any reverse bias from the zener to activate some sort of holding circuit, but I couldn't figure it out by using just transistors. I could use a relay but I don't want to. so I came up with the fuse part.
Does any one know of a zener that will "blow open" 100% of the time? to my understanding zeners can blow open or blow shorted. yes?
a shorted zener would not be helpful at all.
anyway, back to the orignal topic. Im just wanting to see if a spike was present at any time and not just the time that it happens. Thats why i want some type of holding circuit.
I guess what you are calling a "spike" is what I would call a "surge". To me a spike is nano to milli seconds long.
Paul
Paul_KD7HB:
I guess what you are calling a "spike" is what I would call a "surge". To me a spike is nano to milli seconds long.Paul
I would agree with that statement...
I guess on a scope they do look different
DudeGuy:
the input is normally about 14vdc. and the out put would be that there ground symbol.
I don't think so! ! !
One of the aims of this circuit is to disconnected the load from the supply if a spike/surge is detected.
Tom...
Hi,
The 10k to ground may be a problem, as it will prevent the transistor from being biased OFF when the fuse is in tact.
Tom...
PS, Just had my morning java.. :o
i'll give that a try. my thoughts were that the 10k to fuse would supply positive to the PNP and when the fuse blew then the 10k to ground would supply the negative for the PNP. I used the two resistors off the base of the PNP to supply a difference of potential pending the state of the fuse.
It might help if you drew the schematic with and without the fuse.
In TomGeorges schematic there is not path for the base current.
Look at it this way:
The fuse is gone.
The input voltage is now below the 16V of the zener so the zener is not conducting.
There is no path for the base current.
JohnRob:
It might help if you drew the schematic with and without the fuse.In TomGeorges schematic there is not path for the base current.
Look at it this way:
The fuse is gone.
The input voltage is now below the 16V of the zener so the zener is not conducting.There is no path for the base current.
Good point...
If Vout is connected to a load, then Vout pin will go to near gnd, base current will flow through emitter/base.
BUT Vout will have to have a load.
Putting a 1K from Vout to Gnd would ensure base current even without a load.
Tom...
I don't see the current path you are describing. If the base is essentially open circuit, no current is going to flow.
Go back to the sketch (crude) in post #10, Remove the fuse and Zener, you will see a current path from emitter -> base -> R2 -> Ground
Putting the fuse and zener back in. The fuse essentially shorts out the base emitter junction, not allowing any base current to flow out of the transistor.
John
