[hpsdr] LTC 2208 A/D Thermal Characteristics

[Rob Frohne]

Hi Scott,

That extremely small thermal resistance cannot be between the junction 
and the place where Charles and others are measuring the temperature.  
It must be between the junction and the part of the case under the chip 
where it is supposed to be soldered to a good heat sink.  I am not 
saying there is a problem, only that your analysis is not conclusive, 
given the measurements that we can fairly easily perform.  Here are the 
words of wisdom from the LTC2208 data sheet regarding heat considerations:

Most of the heat generated by the LTC2208 is transferred
from the die through the bottom-side exposed pad. For
good electrical and thermal performance, the exposed
pad must be soldered to a large grounded pad on the PC
board. It is critical that the exposed pad and all ground
pins are connected to a ground plane of sufficient area
with as many vias as possible.

The interesting thing is that their example board didn't contain any 
vias to the ground plane that I could see anyway, so it seems that they 
didn't follow their own advice, or maybe I'm blind.  :-)

In any case, increasing the heat sinking will not cause the device to 
operate worse, only perhaps a little better, so I hesitate to discourage 
folk like Francis from doing as they have with more heat sinking at 
least at this point.

73,

Rob, KL7NA

Scott Cowling wrote:
> ***** High Performance Software Defined Radio Discussion List *****
>
> Hi Mercurians,
>
> The thermal resistance of the LTC 64-pin UP package (the one that the 
> LTC2208CUP comes in) is specified at 0.24°C/W from the junction to the 
> case.
>
> The maximum allowed operating junction temperature is 125°C.
>
> The A/D dissipates roughly 1.25W at 3.3V and 1.45W at 3.6V. We are 
> running at 3.3V, but lets assume a worst-case power dissipation of 
> 1.5W, which is higher that it will ever be.
>
> 1.5W * 0.24°C/W = 0.36°C temperature rise from the case to the junction.
>
> That means the junction temperature, T(j), will be only about 1/3 of a 
> degree higher than whatever temperature you measure on the case.
>
> Since the maximum junction temperature is 125°C, the case can get as 
> hot as 124°C WITHOUT EXCEEDING THE MAXIMUM ALLOWED JUNCTION TEMPERATURE.
>
> Since the case temperatures mentioned here are in the 50°C to 70°C 
> range, the chip die is operating at well below the maximum allowable 
> temperature.
>
> While I will never argue against "cooler is better", I will argue 
> against making a problem where there is none.
>
> Could we make it cooler? Sure we could. Do you want to add noise by 
> hooking up a BIG ground plane to the pad on the underside of the chip? 
> Maybe. Designs are about tradeoffs. The designers traded off board 
> area, noise pickup and thermal resistance and came up with a design 
> that works.
>
> It is cool *enough*
> It is quiet *enough*
> It fits on the board
> It works
>
> That doesn't mean that it is necessarily the optimum choice. By all 
> means, modify it, change it, make it better; that is what HPSDR is all 
> about. There are many solutions, some of which are certainly better 
> that the one that we chose. But please don't call it "broken" just 
> because it feels hot to your finger. :-)
>
> Here are some reference points:
> At 50°C, it takes 2 minutes to give you a third degree burn.
> At 60°C, it takes 5 seconds to inflict a third-degree burn.
> At 70°C, it takes only 1 second to give you a third degree burn.
> source: http://www.cqc.state.ny.us/newsletter/estime.htm
>
> It's hot! Be careful!
>
> 73,
> Scotty WA2DFI
>
>
>
>
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at hpsdr.org
> Subscription help: http://lists.hpsdr.org/listinfo.cgi/hpsdr-hpsdr.org
> HPSDR web page: http://hpsdr.org
> Archives: http://lists.hpsdr.org/pipermail/hpsdr-hpsdr.org/

-- 
Rob Frohne, Ph.D., P.E.
E.F. Cross School of Engineering
Walla Walla University
100 SW 4th Street
College Place, WA 99324
(509) 527-2075			 http://people.walllawalla.edu/~rob.frohne

-------------- next part --------------
A non-text attachment was scrubbed...
Name: frohro.vcf
Type: text/x-vcard
Size: 298 bytes
Desc: not available
URL: <http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/attachments/20090114/df80e42d/attachment-0003.vcf>