Simple series Zener diode to drop voltage - okay? 5

[VEBill]

I think I already know the answer (because it seems to be trivial), but I really want to make sure I'm not missing anything.

I'm installing a system which is rated for a supply voltage range of 9 to 30 volts DC. The available power supply is a nominal "28 volts", but the power supply specification actually allows the supply to wander up as high as above 30.5 volts. So there's an issue since 30.5 > 30.0.

I'm considering using a simple Zener diode in series with the + supply line to knock off about 6 volts. This should drop the voltage down into the sweet spot (~22 volts) and avoid the over-voltage issue.

The current is fairly low (~25mA) so the power dissipation is not a significant road block.

I've seen this simple technique used before, but it was always presented as a 'hack' (to use the modern word).

So, the question: Is there any reaon not to use a simple series Zener diode to drop a few volts? And side-effects or details that I need to watch out for?

Thanks.

 

[melone]

Could you just use a forward biased diode? You should be able to get something that could handle some pretty serious current, plus, you only have to drop 0.5V, right? Those hefty diodes will drop ~1V or more.

However, as long as you exceed the minimum voltage and current for avalance breakdown, you should be fine with a zener.

 

[VEBill]

I'd like to get a significant voltage drop, something like about 6 volts, to get well away from the maximum. If I use forward biased diodes then I'd need at least several. Thus, it would be better to use one Zener if it'll work.

The current is quite low, about 25mA. Thus a 6 volt Zener would be under 0.25 watt, so I could probably use a 1-watt version (to be confirmed by worst case analysis later).

The 'normal' solution seems to be to use a $1k Mil-Spec DC-DC converter (28 in , perhaps 15 out) and so on. Seems to be a bit much...

 

[Skogsgurra]

No issues at all. But, if you want a fire-sure solution, then put two zeners in series. Use two 6 V (do they exist? you will probably find 5.1 V instead) zeners.

Not that it is likely to happen, but if one of them gets shorted, the other one will still drop around 5 volts.

You are right when you say that this is trivial. Actually, I don't think you will have any problem at all connecting your PSU directly to the device. Half-a volt above spec shouldn't kill your low-powered application.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[ScottyUK]

skogs,

6.2V is a standard zener voltage.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Skogsgurra]

OK, checked with my distributor. They have the 6.2. But it is a funny non-standard value. Most other voltages are in the E6 or E12 series. Any reason that 6.2 exists? Special use somewhere?

On the other hand, 5.1 isn't in the E12 series either - but 5.6 is. Funny world, this electronic world.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[VEBill]

IIRC, somewhere around 6.2 volts the Zener and avalanche effects balance out to produce the best temperature coefficient.

 

[VEBill]

Looking at a table of 'standard' Zener diode voltages, it approximately follows the pattern (...4.7, 5.1, 5.6, 6.2, etc.) of the standard resistor values (based on 10% steps).

I mentioned 'about 6 volts' because it seemed to be enough to get away from the over voltage condition, and yet not be so high that the power dissipation on the Zener becomes significant.

 

[itsmoked]

Yep 6.2 is the original vanilla flavor.
And one in series is fine. Do make sure you give it a lot of thermal head room. I have seen so many caramelized boards around zeners.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[OperaHouse]

Cases like this would be a good opportunity for using a three terminal regulator in constant current mode.

 

[VEBill]

I want a (more-or-less) constant voltage drop, not a constant current. I could use a three-terminal regulator (connected normally), but then I'd need to add all the recommended extra bits (capacitors) and it'd become a small project onto itself.

So I'm still leaning towards the series 6.2v Zener diode to knock-off about six volts. The thing that had me a bit worried was that this technique was always presented as a 'hack' (a cheat). So I wanted to double-check in case I'm missing something.

 

[BobbyNewmark]

Will there be significant inrush current into the load's decoupling capacitors? A small series resistor might be a good idea.

Will the load's minimum current be above the Zener's leakage current?

 

[sreid]

One reason for using a three terminal regulator is that they are difficult to kill (like a 7824). Short them and they current limit. If they over heat, they shut down.

 

[VEBill]

There's almost certainly a 3-terminal regulator in the unit. One other reason to drop the voltage a bit externally is to avoid having the system shutting down on hot days.

Also, the MTBF should improve if we keep the power supply further away from the upper limit; probably more than enough to compensate for the reduction in MTBF from having an 'extra' part, the Zener diode.

Typically, a 3-terminal regulator application note calls for capacitors on the input and output pins. Although capacitors are cheap, it suddenly becomes a 'design project' and a paper storm ensues.

 

[danw2]

I used to think it trivial, but I burned up the on-the-motherboard power supply of an Advantech WebLink module (blind box running Windows CE before I discovered just how crippled CE was), a $1,000 loss.

The device was spec'd at 24V, and I measured the supply at 24.7 volts. They wouldn't cover it under warranty. Traces burned off the board.

Dan