Diode Lead Length

Anand P. Paralkar

unread,

Feb 6, 2013, 4:33:27 AM2/6/13

to

Hi,

I would like to use the 1N5283 diode in a certain application. Like
some other applications in power electronics, this diode datasheet too
mentions a certain Lead Length to be maintained.

(See http://www.datasheetcatalog.org/datasheet/motorola/1N5298.pdf)

Could someone please explain this? Why does this distance have to be
maintained? (Its got to do with the thermal characteristics alright -
but how?)

I also see thermal resistance versus lead length curve and a note which
says "(MOST HEAT CONDUCTION IS THROUGH THE CATHODE LEAD)". How's this?

I tried searching and found some other datasheets mentioning 3/8" lead
length, but no explanation.

Your inputs please?

Regards,
Anand

Tim Wescott

unread,

Feb 6, 2013, 10:46:25 AM2/6/13

to

On Wed, 06 Feb 2013 15:03:27 +0530, Anand P. Paralkar wrote:

> Hi,
>
> I would like to use the 1N5283 diode in a certain application. Like
> some other applications in power electronics, this diode datasheet too
> mentions a certain Lead Length to be maintained.
>
> (See http://www.datasheetcatalog.org/datasheet/motorola/1N5298.pdf)
>
> Could someone please explain this? Why does this distance have to be
> maintained? (Its got to do with the thermal characteristics alright -
> but how?)

I assume that it's a minimum lead length that's not to be exceeded?
They're probably counting on the circuit board to contribute to the heat
dissipation from the part.

> I also see thermal resistance versus lead length curve and a note which
> says "(MOST HEAT CONDUCTION IS THROUGH THE CATHODE LEAD)". How's this?

Does "most heat GENERATION happens IN the cathode" make more sense?

> I tried searching and found some other datasheets mentioning 3/8" lead
> length, but no explanation.
>
> Your inputs please?

They're rating the parts for an installation into a board where the board
itself is conducting heat away from the part.

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

Jim Thompson

unread,

Feb 6, 2013, 11:05:43 AM2/6/13

to

On Wed, 06 Feb 2013 09:46:25 -0600, Tim Wescott
<t...@seemywebsite.please> wrote:

>On Wed, 06 Feb 2013 15:03:27 +0530, Anand P. Paralkar wrote:
>
>> Hi,
>>
>> I would like to use the 1N5283 diode in a certain application. Like
>> some other applications in power electronics, this diode datasheet too
>> mentions a certain Lead Length to be maintained.
>>
>> (See http://www.datasheetcatalog.org/datasheet/motorola/1N5298.pdf)
>>
>> Could someone please explain this? Why does this distance have to be
>> maintained? (Its got to do with the thermal characteristics alright -
>> but how?)
>
>I assume that it's a minimum lead length that's not to be exceeded?
>They're probably counting on the circuit board to contribute to the heat
>dissipation from the part.
>
>> I also see thermal resistance versus lead length curve and a note which
>> says "(MOST HEAT CONDUCTION IS THROUGH THE CATHODE LEAD)". How's this?
>
>Does "most heat GENERATION happens IN the cathode" make more sense?

Poorly stated... most heat flow is via the cathode lead, for
mechanical reasons...

http://www.analog-innovations.com/SED/Diode_Package.pdf

[snip]

The same reason many I/C's now have a metal slug in the package, under
the die and at least in thermal contact (and in many, it's also
electrical... a quality ground).

...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.

George Herold

unread,

Feb 6, 2013, 1:28:48 PM2/6/13

to

On Feb 6, 4:33 am, "Anand P. Paralkar"

<anand.paral...@gnospammale.com> wrote:
> Hi,
>
> I would like to use the 1N5283 diode in a certain application. Like
> some other applications in power electronics, this diode datasheet too
> mentions a certain Lead Length to be maintained.
>

> (Seehttp://www.datasheetcatalog.org/datasheet/motorola/1N5298.pdf)

>
> Could someone please explain this? Why does this distance have to be
> maintained? (Its got to do with the thermal characteristics alright -
> but how?)
>
> I also see thermal resistance versus lead length curve and a note which
> says "(MOST HEAT CONDUCTION IS THROUGH THE CATHODE LEAD)". How's this?
>
> I tried searching and found some other datasheets mentioning 3/8" lead
> length, but no explanation.
>
> Your inputs please?
>
> Regards,
> Anand

I've never used one of those. But the app note sorta spells it out.
The current is temperature dependent,
The temperature depends on power dissipation and thermal conductivity,
For best control keep power low (low forward voltage)
and thermal conductivity low... short leads.

Jim nailed the cathode part.

George H.

Tim Wescott

unread,

Feb 6, 2013, 4:13:13 PM2/6/13

to

Thanks Jim. That's something that I had all the factoids rolling around
in my head, but never put two and two together. I'll remember now.

Does this apply to surface mount as well? Uniformly? Or might there be
oddball cases where it's the anode that has the good contact?

--
My liberal friends think I'm a conservative kook.
My conservative friends think I'm a liberal kook.
Why am I not happy that they have found common ground?

Tim Wescott, Communications, Control, Circuits & Software
http://www.wescottdesign.com

Jim Thompson

unread,

Feb 6, 2013, 4:44:32 PM2/6/13

to

On Wed, 06 Feb 2013 15:13:13 -0600, Tim Wescott <t...@seemywebsite.com>

I don't know for sure, but, for diodes and zeners, it would depend on
the "starting material". If it's P diffused into N, it's the cathode
that's on the header; and vice versa.

Most I/C's are P-type substrate (*), which is at the most negative
potential (GND or -VEE or -VSS, etc)

(*) In the past, we played games and made two chip amplifiers, like
the MC1524, where one chip was P-substrate, and one was N-substrate,
to optimize vertical NPN's and PNP's... but I've not seen that sort of
gimmick in years.

Phil Allison

unread,

Feb 6, 2013, 4:55:49 PM2/6/13

to

"Anand P. Paralkar"

>
> I would like to use the 1N5283 diode in a certain application. Like some
> other applications in power electronics, this diode datasheet too mentions
> a certain Lead Length to be maintained.

** No such requirement is in the data.

> (See http://www.datasheetcatalog.org/datasheet/motorola/1N5298.pdf)
>
> Could someone please explain this? Why does this distance have to be
> maintained?

** There is no distance to be maintained.

FFS learn to read - it only says " keep leads short ".

.... Phil

John Larkin

unread,

Feb 9, 2013, 9:44:56 PM2/9/13

to

On Wed, 06 Feb 2013 15:03:27 +0530, "Anand P. Paralkar"

As people have noted, most of the heat path is usually through the leads. Fig 2
makes the unreasonable assumption that both leads hit infinite heat sinks at
some distance from the body. On a PC board, that's not the case, although
soldering to some big copper pours is a rough approximation.

Current limiter diodes are expensive and not very accurate. They seem to be
obsolete, or nearly so.

What's your application?

--

John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

John Larkin

unread,

Feb 9, 2013, 9:46:35 PM2/9/13

to

Axial power diodes are often soldered dual nailhead designs, thermally
symmetric.

Cydrome Leader

unread,

Feb 27, 2013, 2:16:56 PM2/27/13

to

Anand P. Paralkar <anand.p...@gnospammale.com> wrote:

> Hi,
>
> I would like to use the 1N5283 diode in a certain application. Like
> some other applications in power electronics, this diode datasheet too
> mentions a certain Lead Length to be maintained.
>
> (See http://www.datasheetcatalog.org/datasheet/motorola/1N5298.pdf)
>
> Could someone please explain this? Why does this distance have to be
> maintained? (Its got to do with the thermal characteristics alright -
> but how?)

they assume the full lead acts as a heatsink, and if you chop it off, you
lose heatsinking, and power rating.

Of course they don't really account for a PCB with a trace also acting as
a heatsink, and that's upto you to determine if it's less or more
effective than a lead sticking out into the air.

I've seen things such as solder turrets used as heatsinks for power
diodes.