Thermal conductivity calculation ?? for comparing thermal interface materials

[John Larkin]

John wrote:



The hole isn't central. The header can indeed 'lift' slightly.

I like to use a *large* washer ( commonly generically called a bicycle washer ) when mounting TO-247 to distribute the pressure.

Spring washers (bellevilles) are ideal... the clamp force is well
defined, and there's compliance for the case that the grease or pad or
whatever settle some.

John

 

[Pooh Bear]

Spring washers (bellevilles) are ideal... the clamp force is well
defined, and there's compliance for the case that the grease or pad or
whatever settle some.

I recall 'discovering' Belleville washers about 13 yrs ago.

Trouble is - I've never found a stockist who understands their application in electronics. I've never seen any data that helps make a
selection.

Graham

 

[John]

Spring washers (bellevilles) are ideal... the clamp force is well
Neither have I. The app notes always just list "Belleville washer"
with no data on diameter or spring pressure. I guess there just isn't
enough of a difference between the different ones that are available?

I whipped out my trusty McMaster-Carr catalog and they have Belleville
washers with serrations on top and bottom (for socket-head cap screws
only), top only, and non-serrated. Interestingly, the top/bottom
serrated ones are the least expensive? I thought the plain ones would
have been.

Since I'm using pads (which will compress), I'm ordering a bunch of
the #6 top-serrated ones along with some washers that are fatter (a
bit) than the washers.

John

-- remove SPAMMENOT for e-mail responses --

 

[John]

Since I'm using pads (which will compress), I'm ordering a bunch of
Let's try that again....
Since I'm using pads (which will compress), I'm ordering a bunch of
the #6 top-serrated Belleville washers along with some large flat
washers.

I *believe* that's the recommeded setup...have to check the mounting
app note again.

John

-- remove SPAMMENOT for e-mail responses --

 

[John Larkin]

I recall 'discovering' Belleville washers about 13 yrs ago.

Trouble is - I've never found a stockist who understands their application in electronics. I've never seen any data that helps make a
selection.

Well, you have to hire an engineer to figure out complicated stuff
like that.

John

 

[Spehro Pefhany]

Neither have I. The app notes always just list "Belleville washer"
with no data on diameter or spring pressure. I guess there just isn't
enough of a difference between the different ones that are available?

Machinery's Handbook doesn't have design data on them- they say it
would take too many formulas and dozens of curves for the constants.

Nice the way you can put them in 'series' or 'parallel' to make up
different forces/deflections.

 

[Pooh Bear]

Machinery's Handbook doesn't have design data on them- they say it
would take too many formulas and dozens of curves for the constants.

Nice the way you can put them in 'series' or 'parallel' to make up
different forces/deflections.

I did get some data on them faxed to me - before I had internet access - probably before you'd have found that data on the web too. I can't
really remember now what conclusion I came to.

In the end I used single coil spring washers.

Graham

 

[Rich Grise]

Neither have I. The app notes always just list "Belleville washer"
with no data on diameter or spring pressure. I guess there just isn't
enough of a difference between the different ones that are available?

I whipped out my trusty McMaster-Carr catalog and they have Belleville
washers with serrations on top and bottom (for socket-head cap screws
only), top only, and non-serrated. Interestingly, the top/bottom
serrated ones are the least expensive? I thought the plain ones would
have been.

Since I'm using pads (which will compress), I'm ordering a bunch of
the #6 top-serrated ones along with some washers that are fatter (a
bit) than the washers.

I just looked them up, and was kind of surprised to find that belleville
washers are much simpler of a design than I thought - it's just an
ordinary, round, flat washer that's not flat, but kind of, um, banked -
like the sides of a very wide, not very tall cone.

I was thinking of those things that look like half of the little cymbals
on a tambourine, that they use under nailheads to post bills on telephone
poles, and stuff like that - the closest McMaster-Carr has is "Finishing
Washer", but that's not it either.

But I didn't see anything (at McM-C) like those little cymbal-shaped
thingies.

Anybody know what I'm talking about?

Thanks,
Rich

 

[John]

I was thinking of those things that look like half of the little cymbals
I had a much more complex curve (profile) in my mind for the washer
too. Perhaps they used to be made that way until someone realized
that they could get most of its performance with a much simpler washer
shape (i.e., less expensive)?

John

-- remove SPAMMENOT for e-mail responses --

 

[Terry Given]

I had a much more complex curve (profile) in my mind for the washer
too. Perhaps they used to be made that way until someone realized
that they could get most of its performance with a much simpler washer
shape (i.e., less expensive)?

John

Belleville washers have been around a LONG time. I recently did a bit
of research into springs (read as: found an excuse to buy more
engineering books) and found this gem:

Mechanical Springs, A.M. Wahl, Penton 1944. see Ch. XIV - Initially
coned disk (belleville) springs....

Cheers
Terry

 

[Spehro Pefhany]

Belleville washers have been around a LONG time. I recently did a bit
of research into springs (read as: found an excuse to buy more
engineering books) and found this gem:

Mechanical Springs, A.M. Wahl, Penton 1944. see Ch. XIV - Initially
coned disk (belleville) springs....

Cheers
Terry

A very long time. Any hint as to why they are called "Belleville"
springs?

My copy of "Mechanical Design of Machines" (copyright 1939-1965) gives
equations for the force/deflection and maximum stresses at the edges,
but notes they have been unable to confirm with measurements, and are
inferring the stress levels (perhaps 180,000 psi) based on endurance
life tests.


Best regards,
Spehro Pefhany

 

[John]

A very long time. Any hint as to why they are called "Belleville"
Found this while looking for the proper way to use them. From the
Adolf Schnorr GmbH + Co. KG "Handbook for Disc Springs":

"Although the disc spring has found a wider
application during the last few decades, it is
still an old established machine component.
The original inventor is not known, but more
than 130 years ago (on 26.12.1861 to be
precise) Julien Francois Belleville of Dunkirk
was granted French Patent Number 52399
for a spring design which already contained
the principle of the disc spring. The importance
this invention achieved is un known,
but the fact that even today France and the
Anglo Saxon countries still speak of “Belleville Springs” infers a
broad dissemination of
this or similar springs. Today this tends to
denote a disc spring of inferior quality, which
still reflects the not always satisfactory design
and function of springs at that time. This
is no wonder considering that in the last
century neither the theoretical conditions for
calculations nor the necessary materials for
manufacture were available.
Not until 1917 did Fr. Dubois develop the
theory on which the calculation of the disc
spring is based in his dissertation “The
Strength of the Conical Shell” [1] at the ETH in
Zurich. However, it still took several decades
until this was adopted in practice. For a long
time disc springs continued to be calculated
– if at all – in accordance with the theory of
the flat perforated plate. Then in 1936 two
Americans, Almen and László, published a
simplified method of calculation which allowed
a quick and practically correct method
for calculating disc springs"

-- remove SPAMMENOT for e-mail responses --

 

[Joseph2k]

My head is ready to explode trying to figure this out.
I have two electrically insulating materials being considered for use
with TO-247AC devices being mounted to a heat sink. They have the
following thermal conductivity and thickness specs:

- Berquist Sil-Pad K10 pad: 1.3 W/m-K conductivity, .006" thick.
- Aavid 4180 Alum. Oxide insulator: 15.06 W/m-K conductivity, 0.080"
thick.

In order to compare their thermal resistances, can I simply divide the
thermal conductivity by the thickness? The units and equations I can
find on the Web make this, what I thought to be very simple, decision
process very difficult.

- K10 thermal resistance = 1.3 / .006 = 216.67
- 4180 thermal resistance = 15.06 / 0.080 = 188.25

Therefore, even though the Aavid aluminum oxide insulator is 13.33
times thicker, it's thermally conductive enough to be the better
choice (thermally, that is)?

Essentially, I'm looking for the best (but still safe) TO-247AC,
electrically insulating, thermal interface material I can find. I'm
currently using Aavid's UltraStick compound (and love it), but it's
not electrically insulating and I need that for this application
(can't prevent touching of the heat sinks).

Thanks!
-- remove SPAMMENOT for e-mail responses --

I come to add to your choices, try:
http://www.chomerics.com/products/thermal.htm

 

[Joseph2k]

Neither have I. The app notes always just list "Belleville washer"
with no data on diameter or spring pressure. I guess there just isn't
enough of a difference between the different ones that are available?

I whipped out my trusty McMaster-Carr catalog and they have Belleville
washers with serrations on top and bottom (for socket-head cap screws
only), top only, and non-serrated. Interestingly, the top/bottom
serrated ones are the least expensive? I thought the plain ones would
have been.

Since I'm using pads (which will compress), I'm ordering a bunch of
the #6 top-serrated ones along with some washers that are fatter (a
bit) than the washers.

John

-- remove SPAMMENOT for e-mail responses --

Crikey mate, just google for belleville washers and learn what you want to
know.

 

[Joseph2k]

"John"



** I think you are being tricked by figures that are non comparables.

The K10 material is soft and conforms to the shape of the device and
heatsink surface leaving no air gaps - so the thermal conductivity figure
is directly usable.

An Alumina insulator is hard needing thermal grease on both sides to
fill voids and work efficiently.

IME with high powered TO3 devices - thin mica with a smear of thermal
grease both sides is still the best.




........ Phil

Please tell me Phil, where do you get pre-punched thin mica now days?

 

[John]

Crikey mate, just google for belleville washers and learn what you want to
Did that after my post. Still couldn't find a LOT of the info I
needed though. Many, many of the manufacturers don't agree on some of
the most basic things, like how far to "set" the washers (50% of
height, 75%, fully set. etc.). Very frustrating.

Sometimes it's just better to ask for a recommendation from those who
use the darn things.

John
-- remove SPAMMENOT for e-mail responses --

 

[John]

I come to add to your choices, try:
Thanks Joseph, they have a couple with very impressive specs.

John
-- remove SPAMMENOT for e-mail responses --

 

[Joseph2k]

Did that after my post. Still couldn't find a LOT of the info I
needed though. Many, many of the manufacturers don't agree on some of
the most basic things, like how far to "set" the washers (50% of
height, 75%, fully set. etc.). Very frustrating.

Sometimes it's just better to ask for a recommendation from those who
use the darn things.

John
-- remove SPAMMENOT for e-mail responses --

Ok, i was a bit rough. If you want thermal conductivity - fully set. If
you want vibration resistance 75%. if you want to allow the part some room
to move without busting something 50%. Easy, once you ask yourself how it
might behave under various tightenings.