Hi
Basically what I want to do is protect a voltage sensitve device (costing quite a bit of money) from getting an overvoltage, I want to power the device from a converter that will be powered from a 12 volt battery the output from the converter will be set to 3 volts, this 3 volts must NOT rise otherwise it will destroy the voltage sensitive device, I can protect the device from over amperage by putting a quick blow 0.1 amp fuse in the supply line, I have seen simple circuits such as this but they do not give any information as to the component values to protect 3 volts from an overvoltage. Also the protection must work fast, I don't care if protection circuit destroys itself in the process.

 

You need to look up the triggering voltage for your chosen SCR. It is the sum of the triggering voltage plus the Zener voltage that causes triggering. It will not be very precise at such a low voltage...

Since the load obviously draws less than 100mA, why not make the upstream supply both current-limited to 100mA and voltage regulated to 3.00V, and then just put a secondary shunt regulator tweaked to 3.01V across the load. The power dissipation in the shunt regulator (if it is ever needed) is small enough so that you dont need a crowbar. You can retain the fast blow fuse between the upstream supply and the shunt reg.

 

You need to look up the triggering voltage for your chosen SCR. It is the sum of the triggering voltage plus the Zener voltage that causes triggering. It will not be very precise at such a low voltage...

Since the load obviously draws less than 100mA, why not make the upstream supply both current-limited to 100mA and voltage regulated to 3.00V, and then just put a secondary shunt regulator tweaked to 3.01V across the load. The power dissipation in the shunt regulator (if it is ever needed) is small enough so that you dont need a crowbar. You can retain the fast blow fuse between the upstream supply and the shunt reg.

Thanks for the reply, first of all I am not very electronically gifted, the supply to the converter is from a 12 volt battery, that cannot be altered, the output from the converter can be altered but needs to be 3 volts to supply the voltage sensitive device, the circuit diagram was what I saw on a website and thought that it might be what I wanted to go between the converter output and the voltage sensitive device to protect it should the converter go belly up and produce more than 3 volts.

 

How much current does the load device draw?

 

How much current does the load device draw?

very small (i think ) what it is, is a new touchscreen control panel for an alde wet central heating system, the part that requires the 3 volts is the memory and clock ( it must have 3 volts backup should the main supply of 12volts gets interrupted) the manufactures don't give any specs on it.

 

I doubt the memory and clock require more than a few micro-Amps.
But now I'm confused. If the main 12V supply fails then so will the 3V supply from the converter .
Why use a converter to provide the 3V? A CR2032 battery, or a couple of AA cells, would do the job.

 

1. How much current the 12V battery could supply?
2. How accurate the voltage sensitve device need to active the protection? (ex: 3.01V, 3.05V, 3.1V?)
3. If you need more precisely then maybe you need to using an op amp to do the detection job, because the zener may not accurate enough.

 

I doubt the memory and clock require more than a few micro-Amps.
But now I'm confused. If the main 12V supply fails then so will the 3V supply from the converter .
Why use a converter to provide the 3V? A CR2032 battery, or a couple of AA cells, would do the job.

sorry for being vague, the system is in a caravan, which has a master switch, this can be turned off when leaving or storing the van, the main supply to this control panel is via the master switch so when it is turned off all power to the control panel is lost, to maintain the memory settings and the clock function 3 volts have to be applied to the memory and clock circuit via a small plug, all this is behind a timber panel so I would like to supply the circuit via a converter powered from the + side of the master switch via a jumper lead, a kind of fix and forget solution so batteries would be no good.

 

sorry for being vague, the system is in a caravan, which has a master switch, this can be turned off when leaving or storing the van, the main supply to this control panel is via the master switch so when it is turned off all power to the control panel is lost, to maintain the memory settings and the clock function 3 volts have to be applied to the memory and clock circuit via a small plug, all this is behind a timber panel so I would like to supply the circuit via a converter powered from the + side of the master switch via a jumper lead, a kind of fix and forget solution so batteries would be no good.

Batteries are the manufacturers way of solving the problem so you know what values I am dealing with.

 

Batteries are the manufacturers way of solving the problem so you know what values I am dealing with.

Also this is the converter that I have my eyes on http://www.ebay.co.uk/itm/221555525923?_trksid=p2060778.m1438.l2649&ssPageName=STRK:MEBIDX:IT
don't know if it has it's own protection?

 

I would stick with the battery if it's only required to maintain the memory. The converter will consume far more power than the thing that it is supplying.

 

or possibly this one http://www.ebay.co.uk/itm/DC-DC-Ste...41300798&tpos=top&ttype=price&talgo=undefined

 

I would stick with the battery if it's only required to maintain the memory. The converter will consume far more power than the thing that it is supplying.

maybe so but the batteries that I have on the van are Trojans with a capacity of 240 a/h and are constantly charged from a solar panel (with solar controller) so battery power is not a problem, on the old panel the memory and the clock circuit were 12 volts so no problem, however her indoors could not understand how to use it, the new panel operation is as easy as falling off a log so to speak.

 

Re posts #10, #13. Very nice, but those converters have unknown operating current/efficiency. Whatever battery you have to power them from the "+ side of the master switch" may be drained rather quickly. I suspect it would be much more efficient to use a discrete-component home-brew regulator for the few uA you probably need.

 

So, as it turns out, the voltage is not super-critical and the current requirement is very very small; Therefore, all you need is a simple shunt-regulator consisting of perhaps a 100k (or larger) resistor and a 3.1V zener diode.

 

As Blocco says, a simple shunt regulator would do the job.
Personally, I'd forget about the DC converter and just use a shunt regulator such as A or B below:

The input resistor and capacitor protect against any voltage spikes.
Version A allows adjustment of the output voltage.
In version B the two Schottky diodes could be replaced by a single ordinary silicon diode (e.g. a 1N4148), but the output voltage would then be ~ 3.1V. (I'm confident the memory and clock would tolerate even more than that. Many real-time-clock systems run happily from 2 AAA cells which, when fresh, could provide ~3.3V.)

 

As Blocco says, a simple shunt regulator would do the job.
Personally, I'd forget about the DC converter and just use a shunt regulator such as A or B below:
View attachment 76747
The input resistor and capacitor protect against any voltage spikes.
Version A allows adjustment of the output voltage.
In version B the two Schottky diodes could be replaced by a single ordinary silicon diode (e.g. a 1N4148), but the output voltage would then be ~ 3.1V. (I'm confident the memory and clock would tolerate even more than that. Many real-time-clock systems run happily from 2 AAA cells which, when fresh, could provide ~3.3V.)

Many thanks to all who have contributed to this thread I will have a go at each one and decide which one will be best.

 

One last thing if I was to utilise one of the shunt regulators as Blocco suggests would it be possible to protect the panel should the shunt regulator go short and put the input voltage to it across the output connections or would the shunt regulator destroy itself before that happend, as I said I don't know much about this.

 

would it be possible to protect the panel should the shunt regulator go short

It all depends on which bit of the regulator goes short. You can never guarantee absolute 100% protection, whatever regulator you use.

 

One last thing if I was to utilise one of the shunt regulators as Blocco suggests would it be possible to protect the panel should the shunt regulator go short and put the input voltage to it across the output connections or would the shunt regulator destroy itself before that happend, as I said I don't know much about this.

If the shunt goes short-circuit, you will get zero voltage to the load and the current will be limited by the preceding series resistor so nothing will happen. If you are concerned that the shunt may go open-circuit; just use two e.g. put two 3.1V zeners in parallel.