Zener Heat

T

Tim M

Guest
I'm designing a power supply that involves putting a bunch of zeners in
series, specifically 6 1N4742As (12V, 1W) and 12 1N4735As (6.2V, 1W).
Yes, I know that's a lot of zeners, but current drain of my circuit is
only 40-50 mA. However, I notice the zeners getting very hot, even
though I'm well below the max power dissipation per diode.

What sort of heat sinking considerations should I be taking. The
1N4742As are glass, the others metal. I have them laid out next to each
other lengthwise (see below), which probably wasn't the best choice for
heat dissipation.

| | |
--- --- ---
| | | | | |
--- --- ---
| | |

Thanks,

Tim
 
I'm designing a power supply that involves putting a bunch of zeners
in series, specifically 6 1N4742As (12V, 1W) and 12 1N4735As (6.2V,
1W). Yes, I know that's a lot of zeners, but current drain of my
circuit is only 40-50 mA. However, I notice the zeners getting very
hot, even though I'm well below the max power dissipation per diode.

Did anyone mention that that is a really stupid design, and why is he
running 40-50 ma through them all??????

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----
 
"Tim M" <tym1@spammenot.cwru.edu> wrote in message
news:cuk9dh$fmq$1@eeyore.INS.cwru.edu...
I'm designing a power supply that involves putting a bunch of zeners in
series, specifically 6 1N4742As (12V, 1W) and 12 1N4735As (6.2V, 1W).
Yes, I know that's a lot of zeners, but current drain of my circuit is
only 40-50 mA. However, I notice the zeners getting very hot, even
though I'm well below the max power dissipation per diode.

What sort of heat sinking considerations should I be taking. The
1N4742As are glass, the others metal. I have them laid out next to each
other lengthwise (see below), which probably wasn't the best choice for
heat dissipation.

| | |
--- --- ---
| | | | | |
--- --- ---
| | |

Thanks,
To get any credable answer you will have to tell us more about why you are
putting so many zeners in series and what you are trying to achieve with
this.

> Tim
 
I am using multiple diodes because I would like a reliable voltage drop
across a variety of currents, with the maximum being 50 mA and the
nominal current being much less, probably around 25 mA. I also need to
pull off multiple different supply voltages for different parts of my
circuit. I am powering multiple ICs (older TTL which draw more current)
some of which require eclcectic operating voltages.

My maximum current estimate is drawn from the summed maximum current
rating from all devices in the circuit.
That's not how zeners are used. Why don't you tell us what the
"eclectic" voltages are, referenced to common, and what is the loading
on each, what is the input voltage referenced to common.
 
Tim M wrote:
I was also thinking of making a (long) heat sink out of aluminum and
having it contact the package of each zener, probably with some heat
conductive paste in between. With the devices so closely spaced, there
isn't a lot of working room at the leads.
You will probably do better if you solder a small copper or brass shim
stock wing on each lead of the diodes right against the body. The
leads are the main conduit of heat out of the die. A pair if 1/4'
wide by 1" high wings can dump a lot of heat, compared to the diode
and ordinary circuit board pads.

--
John Popelish
 
"Tim M" <tym1@spammenot.cwru.edu> wrote in message
news:culna6$i65$1@eeyore.INS.cwru.edu...
Zeners are in series, and to respond to another question, I do realize
the 12V diodes will dissipate about twice as much power as the 6.2V ones.

I'm feeding it with approx. 155 VDC, and the string of zeners adds up to
146.4 (at test conditions there's 149 V across them, which is close
enough for me). I only have 1/4-W resistors handy, so the test setup
has two 330-ohm resistors in parallel to give 165-ohm at 0.5 W.

-Tim
snip
From what you have told us so far this is either a top secret product for
the military or you have but a limited idea of what you are doing.
Be careful - 150Vdc is dangerous - and get someone in to do whatever it is
you are playing with before you harm yourself.
 
On Sat, 12 Feb 2005 16:21:37 +0000, amdx wrote:

"Tim M" <tym1@spammenot.cwru.edu> wrote in message
news:cuk9dh$fmq$1@eeyore.INS.cwru.edu...
I'm designing a power supply that involves putting a bunch of zeners in
series, specifically 6 1N4742As (12V, 1W) and 12 1N4735As (6.2V, 1W).



Hi Tim,
Let's see if we can understand what you are doing.

1) Are the zeners in Series?

This would meaning you would have 98 volts at the top of the string.
I get (6 * 12) + (12 * 6.2) = 146.4V.

At 50 mA (which you seem to have snipped), that's 146.4 * 0.05 = 7.32
watts. Each 6.2V zener dissipates 6.2 * 0.05 = .31W, and each 12V zener
dissipates 12 * .05 = .6 W.

2) What is the voltage you are feeding the string with ?
3) What is the value of your voltage dropping resistor?

If you could answer these three questions, then we can probably help.
Yeah, those two question. ;-)

Good Luck!
Rich
 
You don't have to use a zener string to drop all that voltage.
There is another way to make a low voltage power supply from
line power without using a transformer, if you only need a few
dozen milliamps.
Take a full bridge (four diodes). Feed it from the mains
through a series capacitor. The capacitor's impedance will limit
current into the bridge. You can filter the bridge's output and
regulate it with a zener, since you will have limited current.
The current out of the bridge will be directly proportional to
the capacitance of your blocking capacitor. For 50 mA you would
need something like 1 uF. So with no load your 5 volt zener
conducting 50 mA would dissipate .25 watts, very doable.

You might want to have a resistor in there in addition to the
capacitor, to alleviate possible inrush problems.
 
On Sat, 12 Feb 2005 17:21:26 -0500, Tim M wrote:

Actually, I think it should be 36 mA and not 3.6 mA.

I wish to do thing sans transformers for size and weight issues, plus I
don't want the finished project to require two separate electrical outlets.

The complete power supply circuit works under 36 mA load except for the
heating of the zeners.
Well, since the thing's already a death trap, you might as well use the
capacitor-diode-zener trick. Just run a half-wave doubler by way of a
capacitor to the hot side of the mains. Filter and zener the output of
that. Size the series capacitor to get your load current into spec, and of
course you can filter it to your heart's content. ;-)

Have Fun!
Rich
 
On Sat, 12 Feb 2005 16:04:08 +0000, Fred Bloggs wrote:
valiana@fastmail.fm wrote:
i have used their milia treatment for milia seed on my eye brown area
....

That's makes about as much sense as anything else in this thread....
Never argue with a bot. People might not be able to tell the difference.
--
The Pig Bladder From Uranus, Still Waiting for
Some Hot Babe to Ask What My Favorite Planet Is.
 
On Sat, 12 Feb 2005 05:37:42 -0600, me wrote:

I'm designing a power supply that involves putting a bunch of zeners
in series, specifically 6 1N4742As (12V, 1W) and 12 1N4735As (6.2V,
1W). Yes, I know that's a lot of zeners, but current drain of my
circuit is only 40-50 mA. However, I notice the zeners getting very
hot, even though I'm well below the max power dissipation per diode.

Did anyone mention that that is a really stupid design, and why is he
running 40-50 ma through them all??????
I didn't specifically say, "stupid," but in another post I said that
rather than try series dropping zeners from a nixie supply, since it's a
death trap already, just feed a half-wave doubler with a capacitor and
zener regulate it, with a current path, of course.

But that was just a few minutes ago, so by tomorrow maybe he'll have
noticed. ;-)

Cheers!
Rich
 
I've noticed, and I haven't fallen into the deathtrap yet. ;)

Adding copper fins to the leads of the zeners does dissipate heat nicely
so that the temperature is within spec, but I agree that this design is
a bit flaky for continuous use.

I will give the doubler-capacitor and capacitor-rectifier tricks a try.
Will have to see what I have in the parts bin.

-Tim

Rich Grise wrote:

On Sat, 12 Feb 2005 05:37:42 -0600, me wrote:


I'm designing a power supply that involves putting a bunch of zeners
in series, specifically 6 1N4742As (12V, 1W) and 12 1N4735As (6.2V,
1W). Yes, I know that's a lot of zeners, but current drain of my
circuit is only 40-50 mA. However, I notice the zeners getting very
hot, even though I'm well below the max power dissipation per diode.

Did anyone mention that that is a really stupid design, and why is he
running 40-50 ma through them all??????


I didn't specifically say, "stupid," but in another post I said that
rather than try series dropping zeners from a nixie supply, since it's a
death trap already, just feed a half-wave doubler with a capacitor and
zener regulate it, with a current path, of course.

But that was just a few minutes ago, so by tomorrow maybe he'll have
noticed. ;-)

Cheers!
Rich
 
Adding copper fins to the leads of the zeners does dissipate heat nicely
so that the temperature is within spec,
Wonderful- nothing like lots of exposed metal at high ungrounded voltage.

but I agree that this design is
a bit flaky for continuous use.

I will give the doubler-capacitor and capacitor-rectifier tricks a try.
Will have to see what I have in the parts bin.
That's even more foolish than your original idea.
 
"kell" <kellrobinson@billburg.com> wrote in message
news:1108333498.198539.185420@c13g2000cwb.googlegroups.com...
So the series line capacitor is a no-no?

Only if you need to be told why there is a resistor in series with the
capacitor, why there is one in parallel with the capacitor, why a transient
suppressor is necessary, what is the difference between X and Y rated
capacitors, and what class ll means.

Otherwise there are millions of successful products out there with reactive
ballasts.
 
You seem to get a big turn-on slapping people's shit.
What I "need" ain't what you got.
 
Tim M wrote:
All of these things are from suggestions in this thread and not my
"foolish" idea, sir. I don't see you hassling them.
The foolish idea of using zeners for a 150V voltage drop was your idea.
We know who you are the BTW, shooting your mouth off with all that right
wing born again christian crap.
 
"kell" <kellrobinson@billburg.com> wrote in message
news:1108339200.620761.94210@o13g2000cwo.googlegroups.com...
You seem to get a big turn-on slapping people's shit.
What I "need" ain't what you got.
glad you are so sure
 
"Tim M" <tym1@spammenot.cwru.edu> wrote in message
news:cul7vv$i8q$1@eeyore.INS.cwru.edu...
I am using multiple diodes because I would like a reliable voltage drop
*That* you will not get:

Eventually, one zener will fail stressing the others a bit more so one of
those fail - and directly on the mains too. At least the eventual explosion
of the entire stack and the group-suicide of the IC's should be impressive!

circuit. I am powering multiple ICs (older TTL which draw more current)
some of which require eclcectic operating voltages.
Use a wall-wart and a custom switcher to generate the special voltages - how
hard can it be?

With limited board space as well, I need to crowd the diodes together.
Planar transformers can be *very* small - even with mains isolation.
 

Welcome to EDABoard.com

Sponsor

Back
Top