Hi,

I’ve put together a simple fuse box circuit which I’d like to place between my PC power supply and some diy electronic peripherals (hence the need for protection )

The idea is to have 2 independent channels on the PCB, each capable of handling up to 10A @ 12V.

The circuit is meant to protect mainly against shorts and possibly other nasties like over-voltage/voltage spikes.

In practical terms, I’ll be using four square 4-pin Molex power headers (like the ones used for CPU power on some motherboards) – two headers for inputs and two for outputs.

The parts selection & design questions I’m currently faced with:

(1) What would be the best choice for a protection diode (D1 / D2):

UF4004 ( http://www.fairchildsemi.com/ds/UF/UF4004.pdf)
1N4007 (http://www.fairchildsemi.com/ds/1N/1N4007.pdf )
1N6276AG (TVS) ( http://www.onsemi.com/pub_link/Collateral/1N6267A.PDF )
ICTE12-E3/51 (TVS) (http://www.vishay.com/docs/88356/icte.pdf )
( I have all these particular diodes on hand so if any of them is suitable that would be great. However if you think there’s a much better alternative please let me know).

(2) For the two series fuses, I intend to use a couple of Fast Blowing glass fuses. Is this a good idea?

(3) For the capacitors, I thought of using a couple of 100nF MLCC ceramics (SR155C104KAR) (http://www.avx.com/docs/Catalogs/skycap-sr.pdf). Is this a good choice?

(4) I’d be very happy to incorporate reversed polarity protection in the circuit, but couldn't find a method that’s effective yet doesn’t involve any voltage drop. Any thoughts?

Any advice regarding the above question or any other suggestion on how to improve the circuit would be great

thanks,

Joe

 

Exactly what are the 'protection diodes' supposed to protect?

 

Never seen a fuse box with anything but fuses in them..(well maybe some indicator/alarm circuitry)
And fuses don't provide any overvoltage/spike protection. Only short/overcurrent protection.

And without knowing the nature of the devices being fed from this "fusebox" we cannot answer the slow/fast blow times.. The time curve of a fuse should be chosen to avoid nuisance tripping and to protect the device in an overcurrent situation. As such its device dependent. But in general a "fast blow" would be the best choice. But fast is relative without a number attached to it.

 

Op has probably seen protection diodes like these in other devices.

They are useful mainly after the fuses on the input of load for polarity protection.

Possibly offer some protection from spikes if connecting under load?
In any case there should be no down side. Except maybe they should be after the fuses?
And movs might be better?

 

thanks for the responses, people.

Exactly what are the 'protection diodes' supposed to protect?

they're meant to protect the devices on the output side of the fusebox against over-voltage/votlage spikes, I thiought - or is this completely off base?

And fuses don't provide any overvoltage/spike protection. Only short/overcurrent protection.

I thoguht that was what the diodes were for...I've seen more than a few examples on the net that use TVS diodes like these and claim they provide this kind of protection - or then again, maybe I got it wrong?

And without knowing the nature of the devices being fed from this "fusebox" we cannot answer the slow/fast blow times.. The time curve of a fuse should be chosen to avoid nuisance tripping and to protect the device in an overcurrent situation. As such its device dependent.

the intended devices on the output side of the fusebox are:

1) channel 1 - fan controller (controls the speed of 7 PC fans (12V, 0.2A each max) - for this channel it might be better to use a 2-3A fuse but I wanted the design to allow for other applications in the future, hence the 10A upper limit.

2) channel 2 - pump controller (controls the speed of 2 water cooling pumps (12V, 2A each max). again, maybe I ought to use a lower value fuse here too (6-7A)?

does this help? any thoughts on the suggested fuse values?

But in general a "fast blow" would be the best choice. But fast is relative without a number attached to it.

that makes sense the ones I currently have are just cheap generic ones I got from ebay and they don't have any specifications on the packaging.

do you have any recomendation regarding a particular brand/type that's preferable in this context?

Op has probably seen protection diodes like these in other devices.

yup

Possibly offer some protection from spikes if connecting under load?

yup - I thought the fan controller & pump controller on the output side (see above) constitute "loads" from the prespective of the fuse box - wrong assumption?

In any case there should be no down side. Except maybe they should be after the fuses?

aren't they after the fuses already? (when looked at from the output side)

And movs might be better?

sorry but didn't really understand that part.

 

Your PC power supply shouldn't be producing voltage spikes, so aren't TVS diodes redundant?

 

aren't they after the fuses already? (when looked at from the output side)


Power_in and Power_out..............can be taken either way if you think about it.

I must have it wrong.

I was think in "power out" meant out from the supply.



sorry but didn't really understand that part.

MOVs as in metal oxide varistors.

 

Your PC power supply shouldn't be producing voltage spikes, so aren't TVS diodes redundant?

don't know, and can't check as I don't have a scope (that's why I posted here in the first place).

MOVs as in metal oxide varistors.

thanks for the clarification. but which MOVs, and where? and how?

 

No point IMO in anything but fuses and I'd put switches for turning the outputs on/off.. The output from your PC supply is sufficient/clean enough for a computer.. its going to be fine for anything you attach to it..

 

No point IMO in anything but fuses and I'd put switches for turning the outputs on/off.. The output from your PC supply is sufficient/clean enough for a computer.. its going to be fine for anything you attach to it..

thanks, man, will do - appreciate the advice