over voltage protection

Can any kind person advise me of a simple way to provide over-voltage protection on a 12V DC PSU? The current drawn by the equipment is approx

1A. I've looked at MOVs but they don't seem to protect below about 30V, which I think might be higher than than the PSU output when the PSU regulation has failed. Ideally I suppose I'd like something that would provide protection at about 20V.

Bill

Reply to
Bill Wright
Loading thread data ...

The usual method is a polyfuse in series followed by a transzorb to ground. Use a through-hole polyfuse--their switching behaviour is much more predictable. Tyco makes laminated combinations of the two, called PolyZens, which are nice because the dissipation in the transzorb makes the polyfuse switch much faster and more accurately.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

Are you 'shorting out' excess? Or, 'opening up' to excess?

Stating 1A sounds like you want a 'crowbar' above 12Vdc

If your supply current limits, you might have a problem. If your supply folds back current you can use an appropriate [and cheap] tranzorb. They're fast. Probably digikey or such

You may get away with a raw zener, but don't recommend, because they're slow to turn on. relatively.

Don't ever recommend MOV, since they're rated for 'lifetime of operations. In other words, they disappear AS YOU USE THEM!

Reply to
RobertMacy

Zeners aren't slow, certainly not the avalanche ones (i.e. those whose knees are above about 6 volts). TVS devices are sub-nanosecond.

Avalanche breakdown generates carriers in situ, so you don't have to wait for them to diffuse through undepleted silicon as in normal PN junctions.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

--
If the PSU output goes to 20V, what do you want to have happen? 

John Fields
Reply to
John Fields

This is to protect a LED array. The LEDs will withstand 25V for a short time (I don't know how long but I tried a sample LED and it was OK for the 10 seconds I tried it. The PSU is rated at 3.3A but the load is only

1A, so I thought crowbar protection following a 2A quickblow fuse would be OK.

Bill

Reply to
Bill Wright

That's what I assumed, from my position of ignorance.

Bill

Reply to
Bill Wright

I know you've said this before. And even had an explanation for why I had terrible +/-12 volt square wave from using them in feedback of an Opamp.

But tell me again, I don't understand how I could barely get audio speed square wave [without horrible overshoot spikes because the zeners weren't coming on fast enough]. I didn't get into it much other than hearing from other EMC and microwave people to not use zeners [confirmed by my experience trying to make that sq wave generator] They said the zeners turn on too slowly compared to products DESIGNED to turn on fast.

Reply to
RobertMacy

I think you had a PN rectifier in series with the zener, and it was the rectifier that was being slow.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

Remove M1 from this...

Adjust R2/R1 ratio for 20V, or whatever you want.

Add series R from the junction of R3, R4 to the cathode of the TL431 to keep it from frying.

Increase R4 & R5 to ~5.1K (not critical).

Listen to Larkin whine >:-} ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

The supply does't have a crowbar already?

A good one would. I remember bringing home endless switching supplies from the local surplus store, and they all generally had a crowbar on them, despite being various brands and models.

I'd hope a supply in a computer had some sort of protection.

I haven't actually looked at the switching supplies now used as "ac adapters" to know if they include some protection.

Michael

Reply to
Michael Black

Thanks to those who responded. It was very helpful.

Bill

Reply to
Bill Wright

From memory, when we used those 'slow' zeners for overvoltage protection, we used to keep the zener slightly on with a bias current, so when you needed it, it was already working. had HS blocking diode between zener and circuitry. That sped it up the overvoltage protection clamp quite a bit.

The sq wave generator used two zeners 'facing' each other, using the 'built-in' forward diode. like a 'balanced' tranzorb.

Just how slow can forward conducting diode be? THAT slow? You mean, you can get a huge forward voltage across the zener's junction before it finally comes on? But it's really fast in the breakdown direction? That means back to back diodes in the feedback path, albeit small output voltage, would have done the same?

Maybe should have used a FW high speed rectifier [4 pack] in series with a single zener? Would have been symmetrical for sure.

Any App Notes, or White Papers on the subject you can send me a 'specific' URL that desribe that slow forward and fast breakdown?

Reply to
RobertMacy

Forward conduction in PN diodes is slow, all right--there are app notes on things like SMPS failures caused by huge turn-on overshoots, e.g.

formatting link
The reason is that the thick, highly doped contact region shields out all the E field, so that carriers have to diffuse their way out of that before they get to the depletion zone and feel the electric field. Once the concentration gradient is established, they conduct fine.

Avalanche breakdown is quick, as in avalanche transistor pulsers and avalanche photodiodes. The carriers get generated right there in the depletion zone, where there's no waiting. You can measure it by looking at the modulation of the transmitted light. This paper claims that it works up to >~ 10 GHz in silicon, which is reasonable considering how fast avalanche PDs and transistors are.

formatting link

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

--
http://www.onsemi.com/pub_link/Collateral/HBD854-D.PDF 

Page 27. 
 Click to see the full signature
Reply to
John Fields

Why didn't anyone mention a using a Zener to turn on an SCR?

Just a honkin' big Zener after a fuse should do it, yes?

--
Tim Wescott 
Control system and signal processing consulting 
 Click to see the full signature
Reply to
Tim Wescott

Zeners are generally too soft for crow-bar use. I've had good success with a TL431 tripping a "honkin' big" SCR. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.