Author Topic: Electronic fuse.. (Read 14404 times)

Fraser · « **Reply #1 on:** March 23, 2014, 10:36:49 pm »

I know exactly what you are describing. My electronic fuse is circa 1990 and was a kit from Altronics (Australia). It came from a design published in the Australian Magazine Silicon Chip, or possibly Electronics Australia, much like the Bob Pease ESR meter became available from Dick Smith. I have had a look for it to see if any vendors still make the kit but no joy yet. If I find it, I will post here.

notsob · « **Reply #2 on:** March 23, 2014, 10:49:56 pm »

This may be of interest
http://archive.siliconchip.com.au/cms/A_112183/article.html

kripton2035 · « **Reply #3 on:** March 23, 2014, 10:56:48 pm »

also this schematic from nuts and volts september 2009

pickle9000 · « **Reply #4 on:** March 24, 2014, 12:33:47 am »

How about this?

http://www.ebay.ca/itm/Programmab-Over-Current-Protection-DC-200V-75A-Combo-Meter-Voltage-Amp-Power-Ah-/181242410499?pt=LH_DefaultDomain_0&hash=item2a32e3b603

T3sl4co1l · « **Reply #5 on:** March 24, 2014, 12:42:37 am »

I've made things before;

(Actually, drawn only; I didn't make one, but someone else apparently did successfully.)

But I'm not aware of any general purpose, scalable design that's available. There are monolithic devices (protected linear or latching MOSFETs), but they take relatively high current consumption (>>10uA?), and aren't very big.

I have partial plans for such a solution, and I think the build cost would be around $100 in small quantities; would you be interested in that? Anyone? Like, Kickstarter it?

Tim

Fraser · « **Reply #6 on:** March 24, 2014, 01:23:24 am »

@pickle9000,

That's an interesting and very useful little unit that you have identified. I may just have to buy one of those for my 'just in case' stock

Sadly, as a fuse, 5 updates per second may make it too slow to respond, depending upon the circut it is protecting.

Good find though

pickle9000 · « **Reply #7 on:** March 24, 2014, 01:48:04 am »

That's definitely a slo-blow

Pinkus · « **Reply #8 on:** March 24, 2014, 09:10:55 am »

How about building your own box with a set of different PTC fuses
(e.g. http://www.digikey.com/product-search/en/circuit-protection/ptc-resettable-fuses/656272?k=ptc%20fuse)

ovnr · « **Reply #9 on:** March 24, 2014, 10:10:59 am »

I think this was asked a few months ago; I recommended the LT6108 back then, and my recommendation still stands if you want to build one yourself.

Link: http://www.linear.com/product/LT6108

oldway · « **Reply #10 on:** March 24, 2014, 10:11:13 am »

If I had to develop such an electronic fuse, here's how I conceive the product.

We must start from the worst conditions, a short circuit, with a voltage source of low internal resistance and with a very low inductance in the circuit.

The electronic fuse is intended to protect the voltage source and possibly also the circuits it supplies.

It should be noted that in practice it is impossible to protect a semiconductor with a fuse, except when using an ultra fast fuse with i ² dt at least 30% lower than the semiconductor with diodes, thyristors, triacs and GTO's.

The electronic fuse can not change the operation of the circuit it protects.
For this reason, it must have a very low resistance as a fuse have.

I already deducted that the current sensor can not be done by a resistance, I would choose a Hall effect current sensor.

How does a fuse work?

When the current reaches several times the rated current , the fuse blows and the instantaneous power ( peak current x voltage ) is dissipated in the arc that occurs in the fuse.

This introduces an arc voltage drop and also a resistance which limits the current .
The arc is then extinguished either naturally ( elongation of the arc ) or by silica ( sand) that melt and isolates the electrodes.
If we want an electronic fuse , we must do the same .

First, it is obvious that we can not use a relay because the operating delay varies from 10 to 30 ms ...
In a circuit with low inductance and low resistance of the voltage source , the current will reach its highest value (U / Rint) well before the 10 ms.

The current to be interrupted will be very high and the relay will be damaged (and perhaps also other components of the circuit !) .

You have to choose a MOSFET to interrupt the current .
He must act very fast because the di / dt can be very high and the short circuit current may exceed the maximum peak current of the Mosfet.

However, such a high-speed electronic fuse may be annoying , for example, inadvertently triggering when charging a capacitor.

That is why I would choose another solution: Mosfet should limit the current to an adjustable value, and after an adjustable delay, (compatible with the power the Mosfet can dissipate), I would block it.
It should also provide an overheating protection to protect the Mosfet if the electronic fuse is reset several times.

T3sl4co1l · « **Reply #11 on:** March 24, 2014, 11:02:32 am »

Like I said, I have plans which address this; it should prove reliable even in harsh loads (inductors and capacitors).

The biggest challenge is making something usable in reset. I think I can get away with charging a 100uF cap to 50V, for example, but not 470 or 1000.

It would be battery operated, and should last for weeks to months of continuous operation.

Tim

oldway · « **Reply #12 on:** March 24, 2014, 12:14:09 pm »

Short circuit current of an automotive 12V battery can reach 600A...
And from a loaded big electrolitic?
Thousand or more amps...
Doe you really think one or several relays can interrupt that without been damaged?
Have you already tried to interrupt a dc voltage of 50V or more with high current?
Arc between contacts stays and current is not interrupted !

To interrupt dc voltage of 50V or more with high current, you need a special relay with magnets to blow the arc.

oldway · « **Reply #13 on:** March 24, 2014, 12:33:05 pm »

5A is the rated current, not the short circuit current.
A slow acting fuse (a relay need more than 10 ms to open), will not protect anything...For example, if there is no current limitation in the power supply, rectifiers diodes will blow and short circuit the secundary of the transformer.

NB: I really wonder what you want.
To replace a simple and cheap fuse holder by an electronic fuse that protects less than a glass fuse, it makes no sense.
If you use an electronic fuse, it's intended to protect better than a commun glass fuse.

kripton2035 · « **Reply #14 on:** March 24, 2014, 01:12:35 pm »

it's not intended to protect. it's intended to cut the current if it is too important.
if you repair swichmode power supplies, there are failures that blows the fuse instantly.
if you have an electronic fuse, you reset it
if you have real glass fuses, you resoldier another one other and other and other ... till you find the defect !

cosmos · « **Reply #15 on:** March 24, 2014, 01:36:38 pm »

I made something like this once.

It was based on this switch: BTS6163
Switch has internal short circuit protection, and give out a crude estimate of the current.
Fairly robust device as I remember it.

We converted the current to a voltage and feed it into a comparator (quad) to detect trip (1x) and latch the trip afterwards (2x).
Reset of latch and at start-up was from dropping input voltage under xx V (1x).

This was for a customer so I can't give out more detailed schematic, but it was fairly straight forward.

poorchava · « **Reply #16 on:** March 24, 2014, 02:18:24 pm »

MAX4373

oldway · « **Reply #17 on:** March 24, 2014, 02:58:05 pm »

Quote from: kripton2035 on March 24, 2014, 01:12:35 pm

it's not intended to protect. it's intended to cut the current if it is too important.
if you repair swichmode power supplies, there are failures that blows the fuse instantly.
if you have an electronic fuse, you reset it
if you have real glass fuses, you resoldier another one other and other and other ... till you find the defect !

You speak about an electronic fuse for ac or for dc ?
Switch mode power supply do'nt have dc fuses simply because glass fuses can't interrupt 320Vdc or 395Vdc.
I have had enough MOSFET which exploded in my face (I worked by the greatest european manufacturer of SMPS) to say that a fuse don't protect any semi-conductor.

Using a electronic fuse is a non sense if it not very very fast...
And I have some doubts about the possibility of interrupting 395Vdc (dc voltage of a SMPS with PFC)

oldway · « **Reply #18 on:** March 24, 2014, 04:22:52 pm »

But you don't have any specifications of what you had...

kripton2035 · « **Reply #19 on:** March 24, 2014, 04:29:50 pm »

the glass fuse in an smps is at the entry so 230V or 120v "only"

oldway · « **Reply #20 on:** March 24, 2014, 05:05:02 pm »

Quote from: kripton2035 on March 24, 2014, 04:29:50 pm

the glass fuse in an smps is at the entry so 230V or 120v "only"

It seems that ErikTheNorwegian is loking for a dc electronic fuse, not an ac one.

Quote

f you repair swichmode power supplies, there are failures that blows the fuse instantly.
if you have an electronic fuse, you reset it

This is non sense, you will have to find the defective component and replace it first.
Somebody who has a some experience never power on a SMPS without first testing that the bridge rectifier is good and that there is no short circuit in the dc bus.
For not to blow fuses, nor damage components, connect a 100W incandescent lamp in serie when you first power on a SMPS. ( NB: not for SMPS with PFC).
If the lamp is fully bright, there is short-circuit....
If there is something really wrong with the drives of powerfet's, fuses will not protect nothing...be sure the powerfet's will explode before the fuse can interrupt the current...

oldway · « **Reply #21 on:** March 24, 2014, 05:51:12 pm »

Quote from: ErikTheNorwegian on March 24, 2014, 05:28:46 pm

Quote from: oldway on March 24, 2014, 04:22:52 pm
But you don't have any specifications of what you had...

No, hence, this tread start.. where others had a simular fuse.. whats your problem??

My problem is that you don't know what you are looking for and you are not helping us.

An electronic fuse

Ac or dc ?
How much max voltage ?
How much peak current ?
How fast?
What load? INductive, capacitive or resistive?
For what use?
WIth what power supply?
Has this power supply some current limitation?
What's the internal resistance of this power supply?
What about capacitor charging?
What about cost?
What do you think about the schematics available on Internet ? (looking for "electronic fuse" on Google)

And so on...

pickle9000 · « **Reply #22 on:** March 24, 2014, 06:01:11 pm »

@eric

Just to be clear.

I'll use a computer power supply for the example. You plug your fuse protection device into the mains. Then plug your computer power supply into the fuse protection device. Next you replace the fuse in your computer power supply with one that is of a larger current than the setting on your fuse protection device.

The result is that you have a device that saves you the cost of using multiple fuses, smoking traces when testing the computer power supply. Or to use for burn in purposes.

You are testing items that already have (or probably) a fault and this will save time and money (fuse cost).

You could also call this an adjustable circuit breaker.

Did I get it right?

oldway · « **Reply #23 on:** March 24, 2014, 06:35:34 pm »

Quote

Yepp, but no switch mode supplays. Just regular old school electronic. HIFI amps, HIFI CD players, etc..

For this purpose, I use a variac, increasing the voltage from 0V measuring the current.
If the current is to high, I reduce the voltage without blowing any fuse.

Of course, the output of my variac is also protected by a circuit breaker.

oldway · « **Reply #24 on:** March 24, 2014, 07:01:37 pm »

With a variac and an ammeter, you will never blow a fuse...

It's far better for electrolytics condensators to apply increasing voltage from 0 to 230V than to apply the full main voltage with (or without) electronic fuse.


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: Electronic fuse.. (Read 14404 times)

Share me