electrically insulating, thermally conducting material

classic Classic list List threaded Threaded
13 messages Options
Reply | Threaded
Open this post in threaded view
|

electrically insulating, thermally conducting material

jeremy.rutman-2
you might try anodized aluminum, mica, or kapton as electrically insulating
thermal conductors. whats needed is the exact opposite of whats good in
thermoelectric generators by the way. you could find the 'figure of demerit',
the inverse of sigma_electrical over sigma_thermal as used by TEG people.


Dan Frederiksen <[hidden email]> wrote
> I initially wanted water cooling for compactness but alas the damn little
>components are electrically active on the thermal face.. so some electrically

>insulating and thermally dreadful film is needed in between and I really
>don't like that.
> oil cooling has come up but not sure if that is any good at taking heat >away.

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev
Reply | Threaded
Open this post in threaded view
|

Re: electrically insulating, thermally conducting material

Dan Frederiksen-2
Shalom :)

but I need it to be a specific product that's not too expensive and not
to poorly thermally conducting. heatpads seem to fail on all counts :)

I'm currently leaning towards a large two piece internal heatsink with
forced air to keep it 'simple'. it is an unfortunate complication that
the transistors are live on their backside but I imagine there is a good
thermal reason for that design.

it can work, it's just finding the right heatsinks and arranging it
elegantly so it's easy to build for others. would be nice with
transistors that have zero internal loss :)

Dan

[hidden email] wrote:

> you might try anodized aluminum, mica, or kapton as electrically insulating
> thermal conductors. whats needed is the exact opposite of whats good in
> thermoelectric generators by the way. you could find the 'figure of demerit',
> the inverse of sigma_electrical over sigma_thermal as used by TEG people.
>
>
> Dan Frederiksen <[hidden email]> wrote
>  
>> I initially wanted water cooling for compactness but alas the damn little
>> components are electrically active on the thermal face.. so some electrically
>>    
>
>  
>> insulating and thermally dreadful film is needed in between and I really
>> don't like that.
>> oil cooling has come up but not sure if that is any good at taking heat >away.
>>    
>
> _______________________________________________
> For subscription options, see
> http://lists.sjsu.edu/mailman/listinfo/ev
>
>  

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev
Reply | Threaded
Open this post in threaded view
|

Re: electrically insulating, thermally conducting material

Morgan LaMoore
Dan,

Go to http://www.digikey.com and search for "heat sink pad". They have
over 300 insulating pads and give you information on the thermal
conductivity. Be sure to buy some nylon screws from them, too, to
attach the transistors to the heatsinks without shorts!

Thermal pads and nylon screws are pretty standard for attaching a
connected tab to a heatsink that you want isolated. You should be able
to find a good variety at many electronics suppliers.

-Morgan LaMoore

On Nov 27, 2007 10:13 PM, Dan Frederiksen <[hidden email]> wrote:

> Shalom :)
>
> but I need it to be a specific product that's not too expensive and not
> to poorly thermally conducting. heatpads seem to fail on all counts :)
>
> I'm currently leaning towards a large two piece internal heatsink with
> forced air to keep it 'simple'. it is an unfortunate complication that
> the transistors are live on their backside but I imagine there is a good
> thermal reason for that design.
>
> it can work, it's just finding the right heatsinks and arranging it
> elegantly so it's easy to build for others. would be nice with
> transistors that have zero internal loss :)
>
> Dan
>
>
> [hidden email] wrote:
> > you might try anodized aluminum, mica, or kapton as electrically insulating
> > thermal conductors. whats needed is the exact opposite of whats good in
> > thermoelectric generators by the way. you could find the 'figure of demerit',
> > the inverse of sigma_electrical over sigma_thermal as used by TEG people.
> >
> >
> > Dan Frederiksen <[hidden email]> wrote
> >
> >> I initially wanted water cooling for compactness but alas the damn little
> >> components are electrically active on the thermal face.. so some electrically
> >>
> >
> >
> >> insulating and thermally dreadful film is needed in between and I really
> >> don't like that.
> >> oil cooling has come up but not sure if that is any good at taking heat >away.
> >>
> >
> > _______________________________________________
> > For subscription options, see
> > http://lists.sjsu.edu/mailman/listinfo/ev
> >
> >
>
> _______________________________________________
> For subscription options, see
> http://lists.sjsu.edu/mailman/listinfo/ev
>

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev
Reply | Threaded
Open this post in threaded view
|

Re: electrically insulating, thermally conducting material

Dan Frederiksen-2
I know digikey has heatpad but the ones I saw had poor thermal
conductivity and bizarrely high price. more than 50$ for a letter sized
sheet. a single sheet.
that kind of expense for a small item is not good for an open source design.

and not that it's elemental to the issue but I don't think you need
plastic screws for that though. it's not metal in the screw hole of the
transistor.

Dan



Morgan LaMoore wrote:

> Dan,
>
> Go to http://www.digikey.com and search for "heat sink pad". They have
> over 300 insulating pads and give you information on the thermal
> conductivity. Be sure to buy some nylon screws from them, too, to
> attach the transistors to the heatsinks without shorts!
>
> Thermal pads and nylon screws are pretty standard for attaching a
> connected tab to a heatsink that you want isolated. You should be able
> to find a good variety at many electronics suppliers.
>
> -Morgan LaMoore
>
> On Nov 27, 2007 10:13 PM, Dan Frederiksen <[hidden email]> wrote:
>  
>> Shalom :)
>>
>> but I need it to be a specific product that's not too expensive and not
>> to poorly thermally conducting. heatpads seem to fail on all counts :)
>>
>> I'm currently leaning towards a large two piece internal heatsink with
>> forced air to keep it 'simple'. it is an unfortunate complication that
>> the transistors are live on their backside but I imagine there is a good
>> thermal reason for that design.
>>
>> it can work, it's just finding the right heatsinks and arranging it
>> elegantly so it's easy to build for others. would be nice with
>> transistors that have zero internal loss :)
>>
>> Dan
>>
>>
>> [hidden email] wrote:
>>    
>>> you might try anodized aluminum, mica, or kapton as electrically insulating
>>> thermal conductors. whats needed is the exact opposite of whats good in
>>> thermoelectric generators by the way. you could find the 'figure of demerit',
>>> the inverse of sigma_electrical over sigma_thermal as used by TEG people.
>>>
>>>
>>> Dan Frederiksen <[hidden email]> wrote
>>>
>>>      
>>>> I initially wanted water cooling for compactness but alas the damn little
>>>> components are electrically active on the thermal face.. so some electrically
>>>>
>>>>        
>>>      
>>>> insulating and thermally dreadful film is needed in between and I really
>>>> don't like that.
>>>> oil cooling has come up but not sure if that is any good at taking heat >away.
>>>>
>>>>        
>>> _______________________________________________
>>> For subscription options, see
>>> http://lists.sjsu.edu/mailman/listinfo/ev
>>>
>>>
>>>      
>> _______________________________________________
>> For subscription options, see
>> http://lists.sjsu.edu/mailman/listinfo/ev
>>
>>    
>
> _______________________________________________
> For subscription options, see
> http://lists.sjsu.edu/mailman/listinfo/ev
>
>  

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev
Reply | Threaded
Open this post in threaded view
|

Re: electrically insulating, thermally conducting material

Morgan LaMoore
In reply to this post by Morgan LaMoore
Dan Frederiksen,

Oops, I just noticed that you said that thermal pads aren't good
enough. I completely disagree that thermal pads are too expensive;
many cost less than $0.20 each in low quantities, and under $0.05 in
mass production quantities.

As far as thermal conductivity is concerned, consider this example:

Your power stage uses 25 transistors in parallel.
You are switching 50 kW.
2 kW of waste heat is generated in the transistors.
You use these thermal pads:
http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=BER167-ND
0.2 K/W, $0.10 each

Thus, the waste heat in each transistor is 2 kW/25=80 W
The temperature across each thermal pad is 80W*0.2K/W = 16 degrees

These don't seem like unacceptable numbers to me.

Good luck!

-Morgan LaMoore

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev
Reply | Threaded
Open this post in threaded view
|

Re: electrically insulating, thermally conducting material

Dan Frederiksen-2
no that's not too bad although it's still 16 degrees which could be
spent on more power. there are some problematic issues though. the
breakdown voltage is 300V. that might be enough but seems uncomfortably
close for a max system voltage that could be 350v. (but maybe)
further it's a TO220 pad. I am using TO247 transistors for which pads
seem to cost alot more.  that could be changed for a revision of the
design but all in all these heatpads don't scream nice design to me

another stray thought, maybe the waste heat from the controller can be
used as cabin heater.

Dan
 

Morgan LaMoore wrote:

> Dan Frederiksen,
>
> Oops, I just noticed that you said that thermal pads aren't good
> enough. I completely disagree that thermal pads are too expensive;
> many cost less than $0.20 each in low quantities, and under $0.05 in
> mass production quantities.
>
> As far as thermal conductivity is concerned, consider this example:
>
> Your power stage uses 25 transistors in parallel.
> You are switching 50 kW.
> 2 kW of waste heat is generated in the transistors.
> You use these thermal pads:
> http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=BER167-ND
> 0.2 K/W, $0.10 each
>
> Thus, the waste heat in each transistor is 2 kW/25=80 W
> The temperature across each thermal pad is 80W*0.2K/W = 16 degrees
>
> These don't seem like unacceptable numbers to me.
>
> Good luck!
>
> -Morgan LaMoore
>
> _______________________________________________
> For subscription options, see
> http://lists.sjsu.edu/mailman/listinfo/ev
>
>  

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev
Reply | Threaded
Open this post in threaded view
|

Re: electrically insulating, thermally conducting material

Dan Frederiksen-2
In reply to this post by Morgan LaMoore
I noticed something else in their spec sheet:
http://www.bergquistcompany.com/objects/data_sheets/PDS_HF115AC_0305E.pdf
a thermal performance more like 1K/W. quite a significant difference
from the stated on digikey.
do you understand what's going on there?

if that's true it will need a temperature difference of 80 degrees
across that thin pad to transport 80watt
 
Dan


Morgan LaMoore wrote:

> You use these thermal pads:
> http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=BER167-ND
> 0.2 K/W, $0.10 each
>
> Thus, the waste heat in each transistor is 2 kW/25=80 W
> The temperature across each thermal pad is 80W*0.2K/W = 16 degrees
>
> These don't seem like unacceptable numbers to me.
>
> Good luck!
>
> -Morgan LaMoore
>
> _______________________________________________
> For subscription options, see
> http://lists.sjsu.edu/mailman/listinfo/ev
>
>  

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev
Reply | Threaded
Open this post in threaded view
|

Re: electrically insulating, thermally conducting material

Peter Gabrielsson
Search digikey for isoplus.

However, isolation always come at a cost, devil is in the details....

On Nov 27, 2007 9:44 PM, Dan Frederiksen <[hidden email]> wrote:

> I noticed something else in their spec sheet:
> http://www.bergquistcompany.com/objects/data_sheets/PDS_HF115AC_0305E.pdf
> a thermal performance more like 1K/W. quite a significant difference
> from the stated on digikey.
> do you understand what's going on there?
>
> if that's true it will need a temperature difference of 80 degrees
> across that thin pad to transport 80watt
>
> Dan
>
>
>
> Morgan LaMoore wrote:
> > You use these thermal pads:
> > http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=BER167-ND
> > 0.2 K/W, $0.10 each
> >
> > Thus, the waste heat in each transistor is 2 kW/25=80 W
> > The temperature across each thermal pad is 80W*0.2K/W = 16 degrees
> >
> > These don't seem like unacceptable numbers to me.
> >
> > Good luck!
> >
> > -Morgan LaMoore
> >
> > _______________________________________________
> > For subscription options, see
> > http://lists.sjsu.edu/mailman/listinfo/ev
> >
> >
>
> _______________________________________________
> For subscription options, see
> http://lists.sjsu.edu/mailman/listinfo/ev
>



--
www.electric-lemon.com

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev
Reply | Threaded
Open this post in threaded view
|

Re: electrically insulating, thermally conducting material

(-Phil-)
In reply to this post by Dan Frederiksen-2
I wonder if you could do some of the stunts the extreme PC overclockers do,
and immerse your whole controller output stage in liquid helium.  With the
temp coefficient of MOSFETs working in your favor, you might get a lot
closer to that zero internal loss..... =)

-Phil
----- Original Message -----
From: "Dan Frederiksen" <[hidden email]>
To: "Electric Vehicle Discussion List" <[hidden email]>
Sent: Tuesday, November 27, 2007 8:13 PM
Subject: Re: [EVDL] electrically insulating, thermally conducting material


> Shalom :)
>
> but I need it to be a specific product that's not too expensive and not
> to poorly thermally conducting. heatpads seem to fail on all counts :)
>
> I'm currently leaning towards a large two piece internal heatsink with
> forced air to keep it 'simple'. it is an unfortunate complication that
> the transistors are live on their backside but I imagine there is a good
> thermal reason for that design.
>
> it can work, it's just finding the right heatsinks and arranging it
> elegantly so it's easy to build for others. would be nice with
> transistors that have zero internal loss :)
>
> Dan
>
> [hidden email] wrote:
>> you might try anodized aluminum, mica, or kapton as electrically
>> insulating
>> thermal conductors. whats needed is the exact opposite of whats good in
>> thermoelectric generators by the way. you could find the 'figure of
>> demerit',
>> the inverse of sigma_electrical over sigma_thermal as used by TEG people.
>>
>>
>> Dan Frederiksen <[hidden email]> wrote
>>
>>> I initially wanted water cooling for compactness but alas the damn
>>> little
>>> components are electrically active on the thermal face.. so some
>>> electrically
>>>
>>
>>
>>> insulating and thermally dreadful film is needed in between and I really
>>> don't like that.
>>> oil cooling has come up but not sure if that is any good at taking heat
>>>  >away.
>>>
>>
>> _______________________________________________
>> For subscription options, see
>> http://lists.sjsu.edu/mailman/listinfo/ev
>>
>>
>
> _______________________________________________
> For subscription options, see
> http://lists.sjsu.edu/mailman/listinfo/ev
>

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev
Reply | Threaded
Open this post in threaded view
|

Re: electrically insulating, thermally conducting material

(-Phil-)
In reply to this post by Morgan LaMoore
I would not use nylon screws.  The best trick for TO-220 and other "tabbed"
case styles is to put a hold-down clamp across the body secured with screws
on either side.  It could be a long bar with screws in-between each
transistor.  This way, the body of the transistor serves as the dielectric
for the bar, and you can still maintain a good mechanical pressure against
your thermal dielectric pad.

-Phil
----- Original Message -----
From: "Morgan LaMoore" <[hidden email]>
To: "Electric Vehicle Discussion List" <[hidden email]>
Sent: Tuesday, November 27, 2007 8:29 PM
Subject: Re: [EVDL] electrically insulating, thermally conducting material


> Dan,
>
> Go to http://www.digikey.com and search for "heat sink pad". They have
> over 300 insulating pads and give you information on the thermal
> conductivity. Be sure to buy some nylon screws from them, too, to
> attach the transistors to the heatsinks without shorts!
>
> Thermal pads and nylon screws are pretty standard for attaching a
> connected tab to a heatsink that you want isolated. You should be able
> to find a good variety at many electronics suppliers.
>
> -Morgan LaMoore
>
> On Nov 27, 2007 10:13 PM, Dan Frederiksen <[hidden email]>
> wrote:
>> Shalom :)
>>
>> but I need it to be a specific product that's not too expensive and not
>> to poorly thermally conducting. heatpads seem to fail on all counts :)
>>
>> I'm currently leaning towards a large two piece internal heatsink with
>> forced air to keep it 'simple'. it is an unfortunate complication that
>> the transistors are live on their backside but I imagine there is a good
>> thermal reason for that design.
>>
>> it can work, it's just finding the right heatsinks and arranging it
>> elegantly so it's easy to build for others. would be nice with
>> transistors that have zero internal loss :)
>>
>> Dan
>>
>>
>> [hidden email] wrote:
>> > you might try anodized aluminum, mica, or kapton as electrically
>> > insulating
>> > thermal conductors. whats needed is the exact opposite of whats good in
>> > thermoelectric generators by the way. you could find the 'figure of
>> > demerit',
>> > the inverse of sigma_electrical over sigma_thermal as used by TEG
>> > people.
>> >
>> >
>> > Dan Frederiksen <[hidden email]> wrote
>> >
>> >> I initially wanted water cooling for compactness but alas the damn
>> >> little
>> >> components are electrically active on the thermal face.. so some
>> >> electrically
>> >>
>> >
>> >
>> >> insulating and thermally dreadful film is needed in between and I
>> >> really
>> >> don't like that.
>> >> oil cooling has come up but not sure if that is any good at taking
>> >> heat >away.
>> >>
>> >
>> > _______________________________________________
>> > For subscription options, see
>> > http://lists.sjsu.edu/mailman/listinfo/ev
>> >
>> >
>>
>> _______________________________________________
>> For subscription options, see
>> http://lists.sjsu.edu/mailman/listinfo/ev
>>
>
> _______________________________________________
> For subscription options, see
> http://lists.sjsu.edu/mailman/listinfo/ev
>

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev
Reply | Threaded
Open this post in threaded view
|

Re: electrically insulating, thermally conducting material

Dan Frederiksen-2
In reply to this post by (-Phil-)
hehe I'll keep it in mind

(-Phil-) wrote:

> I wonder if you could do some of the stunts the extreme PC overclockers do,
> and immerse your whole controller output stage in liquid helium.  With the
> temp coefficient of MOSFETs working in your favor, you might get a lot
> closer to that zero internal loss..... =)
>
> -Phil
> ----- Original Message -----
> From: "Dan Frederiksen" <[hidden email]>
> To: "Electric Vehicle Discussion List" <[hidden email]>
> Sent: Tuesday, November 27, 2007 8:13 PM
> Subject: Re: [EVDL] electrically insulating, thermally conducting material
>
>
>  
>> Shalom :)
>>
>> but I need it to be a specific product that's not too expensive and not
>> to poorly thermally conducting. heatpads seem to fail on all counts :)
>>
>> I'm currently leaning towards a large two piece internal heatsink with
>> forced air to keep it 'simple'. it is an unfortunate complication that
>> the transistors are live on their backside but I imagine there is a good
>> thermal reason for that design.
>>
>> it can work, it's just finding the right heatsinks and arranging it
>> elegantly so it's easy to build for others. would be nice with
>> transistors that have zero internal loss :)
>>
>> Dan
>>
>> [hidden email] wrote:
>>    
>>> you might try anodized aluminum, mica, or kapton as electrically
>>> insulating
>>> thermal conductors. whats needed is the exact opposite of whats good in
>>> thermoelectric generators by the way. you could find the 'figure of
>>> demerit',
>>> the inverse of sigma_electrical over sigma_thermal as used by TEG people.
>>>
>>>
>>> Dan Frederiksen <[hidden email]> wrote
>>>
>>>      
>>>> I initially wanted water cooling for compactness but alas the damn
>>>> little
>>>> components are electrically active on the thermal face.. so some
>>>> electrically
>>>>
>>>>        
>>>      
>>>> insulating and thermally dreadful film is needed in between and I really
>>>> don't like that.
>>>> oil cooling has come up but not sure if that is any good at taking heat
>>>>  >away.
>>>>
>>>>        
>>> _______________________________________________
>>> For subscription options, see
>>> http://lists.sjsu.edu/mailman/listinfo/ev
>>>
>>>
>>>      
>> _______________________________________________
>> For subscription options, see
>> http://lists.sjsu.edu/mailman/listinfo/ev
>>
>>    
>
> _______________________________________________
> For subscription options, see
> http://lists.sjsu.edu/mailman/listinfo/ev
>
>  

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev
Reply | Threaded
Open this post in threaded view
|

Re: electrically insulating, thermally conducting material

Dan Frederiksen-2
In reply to this post by (-Phil-)
I'm thinking about that too. I have some TO247s which when held down in
the screw hole flip up from the surface!! BAD

(-Phil-) wrote:

> I would not use nylon screws.  The best trick for TO-220 and other "tabbed"
> case styles is to put a hold-down clamp across the body secured with screws
> on either side.  It could be a long bar with screws in-between each
> transistor.  This way, the body of the transistor serves as the dielectric
> for the bar, and you can still maintain a good mechanical pressure against
> your thermal dielectric pad.
>
> -Phil
> ----- Original Message -----
> From: "Morgan LaMoore" <[hidden email]>
> To: "Electric Vehicle Discussion List" <[hidden email]>
> Sent: Tuesday, November 27, 2007 8:29 PM
> Subject: Re: [EVDL] electrically insulating, thermally conducting material
>
>
>  
>> Dan,
>>
>> Go to http://www.digikey.com and search for "heat sink pad". They have
>> over 300 insulating pads and give you information on the thermal
>> conductivity. Be sure to buy some nylon screws from them, too, to
>> attach the transistors to the heatsinks without shorts!
>>
>> Thermal pads and nylon screws are pretty standard for attaching a
>> connected tab to a heatsink that you want isolated. You should be able
>> to find a good variety at many electronics suppliers.
>>
>> -Morgan LaMoore
>>
>> On Nov 27, 2007 10:13 PM, Dan Frederiksen <[hidden email]>
>> wrote:
>>    
>>> Shalom :)
>>>
>>> but I need it to be a specific product that's not too expensive and not
>>> to poorly thermally conducting. heatpads seem to fail on all counts :)
>>>
>>> I'm currently leaning towards a large two piece internal heatsink with
>>> forced air to keep it 'simple'. it is an unfortunate complication that
>>> the transistors are live on their backside but I imagine there is a good
>>> thermal reason for that design.
>>>
>>> it can work, it's just finding the right heatsinks and arranging it
>>> elegantly so it's easy to build for others. would be nice with
>>> transistors that have zero internal loss :)
>>>
>>> Dan
>>>
>>>
>>> [hidden email] wrote:
>>>      
>>>> you might try anodized aluminum, mica, or kapton as electrically
>>>> insulating
>>>> thermal conductors. whats needed is the exact opposite of whats good in
>>>> thermoelectric generators by the way. you could find the 'figure of
>>>> demerit',
>>>> the inverse of sigma_electrical over sigma_thermal as used by TEG
>>>> people.
>>>>
>>>>
>>>> Dan Frederiksen <[hidden email]> wrote
>>>>
>>>>        
>>>>> I initially wanted water cooling for compactness but alas the damn
>>>>> little
>>>>> components are electrically active on the thermal face.. so some
>>>>> electrically
>>>>>
>>>>>          
>>>>        
>>>>> insulating and thermally dreadful film is needed in between and I
>>>>> really
>>>>> don't like that.
>>>>> oil cooling has come up but not sure if that is any good at taking
>>>>> heat >away.
>>>>>
>>>>>          
>>>> _______________________________________________
>>>> For subscription options, see
>>>> http://lists.sjsu.edu/mailman/listinfo/ev
>>>>
>>>>
>>>>        
>>> _______________________________________________
>>> For subscription options, see
>>> http://lists.sjsu.edu/mailman/listinfo/ev
>>>
>>>      
>> _______________________________________________
>> For subscription options, see
>> http://lists.sjsu.edu/mailman/listinfo/ev
>>
>>    
>
> _______________________________________________
> For subscription options, see
> http://lists.sjsu.edu/mailman/listinfo/ev
>
>  

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev
Reply | Threaded
Open this post in threaded view
|

Re: electrically insulating, thermally conducting material

Cor van de Water
In reply to this post by jeremy.rutman-2
Note that thermal conducting pads come in different sorts
of K/W, you can choose which amount of thermal conductance
or isolation that you need for your application.
So in the PDF you may be looking at a different variant
than the one on the Digikey site, keep that in mind.
In addition, the K/W is usually dependent on the
surface AND the thickness of the pad, in addition to
the variation of it per different material, so there are
3 variables you need to put into the calculations.


Cor van de Water
Systems Architect
Proxim Wireless Corporation http://www.proxim.com
Email: [hidden email]    Private: http://www.cvandewater.com
Skype: cor_van_de_water     IM: [hidden email]
Tel: +1 408 542 5225    VoIP: +31 20 3987567 FWD# 25925
Fax: +1 408 731 3675    eFAX: +31-87-784-1130
Second Life: www.secondlife.com/?u=3b42cb3f4ae249319edb487991c30acb

-----Original Message-----
From: [hidden email] [mailto:[hidden email]] On Behalf Of Peter Gabrielsson
Sent: Tuesday, November 27, 2007 10:18 PM
To: Electric Vehicle Discussion List
Subject: Re: [EVDL] electrically insulating, thermally conducting material

Search digikey for isoplus.

However, isolation always come at a cost, devil is in the details....

On Nov 27, 2007 9:44 PM, Dan Frederiksen <[hidden email]> wrote:

> I noticed something else in their spec sheet:
> http://www.bergquistcompany.com/objects/data_sheets/PDS_HF115AC_0305E.
> pdf a thermal performance more like 1K/W. quite a significant
> difference from the stated on digikey.
> do you understand what's going on there?
>
> if that's true it will need a temperature difference of 80 degrees
> across that thin pad to transport 80watt
>
> Dan
>
>
>
> Morgan LaMoore wrote:
> > You use these thermal pads:
> > http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=BER167
> > -ND
> > 0.2 K/W, $0.10 each
> >
> > Thus, the waste heat in each transistor is 2 kW/25=80 W The
> > temperature across each thermal pad is 80W*0.2K/W = 16 degrees
> >
> > These don't seem like unacceptable numbers to me.
> >
> > Good luck!
> >
> > -Morgan LaMoore
> >
> > _______________________________________________
> > For subscription options, see
> > http://lists.sjsu.edu/mailman/listinfo/ev
> >
> >
>
> _______________________________________________
> For subscription options, see
> http://lists.sjsu.edu/mailman/listinfo/ev
>



--
www.electric-lemon.com

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev

_______________________________________________
For subscription options, see
http://lists.sjsu.edu/mailman/listinfo/ev