[hpsdr] LTC 2208 A/D Thermal Characteristics

[Ray Anderson]

Scott-

Not to belabor the point, but I found the Linear Tech document < http://www.linear.com/designtools/packaging/Linear_Technology_Thermal_Resistance_Table.pdf > where you apparently found the 0.24 °C/W number for the UP package. It is theta(jc) as I suspected. On the same document they quote theta(ja) as 28°C/W (which is even higher than quoted on the actual LTC2208 datasheet). These numbers are predicated on having the mounting heat sink path as recommended by the vendor.

So yes I think 20°C/W (or even 28°C/W) is a more realistic number " just because it appears on the data sheet"...

Bottom line is that as long as the silicon is below 100°C there is a comfortable safety margin. 

Regards,

-Ray  WB6TPU


> -----Original Message-----
> From: hpsdr-bounces at lists.hpsdr.org [mailto:hpsdr-bounces at lists.hpsdr.org]
> On Behalf Of Scott Cowling
> Sent: Wednesday, January 14, 2009 8:17 AM
> To: hpsdr at lists.hpsdr.org
> Subject: Re: [hpsdr] LTC 2208 A/D Thermal Characteristics
> 
> ***** High Performance Software Defined Radio Discussion List *****
> 
> I have reread my post several times, and I do not
> see where I ever said "at the top of the case".
> 
> The 0.24°C/W is the number off of the LTC package information.
> 
> If you feel that it is an unrealistic number,
> feel free to substitute any other number that you like.
> 
> If you think it is the number to the underside,
> feel free to measure it there. Add a few °C/W for
> the chip-to-board connection so you don't have to
> unsolder your chip to measure it.
> 
> If you think that 20°C/W (specified thermal
> resistance between junction and ambient) is a
> realistic number just because it appears on the data sheet, use that one.
> 
> Every one of these choices will result in a
> conclusion that the part is operating within
> spec. There will not be a problem until you reach
> nearly 100°C (measured at your favorite spot)
> using the worst-case numbers above.
> 
> Phil, if you choose to oversimplify grounding and
> thermal tradeoffs and call it "nonsense", feel
> free to do so (I oversimplified it too). My
> experience, on the other hand, tells me otherwise.
> 
> None of this changes the fact that there is
> plenty of thermal headroom on Mercury.
> 
> My intent was not to discourage experimentation
> (I specifically stated the opposite), but to
> dispel the myth that there is somehow a design or
> manufacturing defect in Mercury boards with ADC
> temperatures at 50°C, 60°C or even 70°C.
> 
> You are going to have to come up with
> temperatures a lot higher than these before you have a *real* problem. :-)
> 
> 73,
> Scotty WA2DFI
> 

This email and any attachments are intended for the sole use of the named recipient(s) and contain(s) confidential information that may be proprietary, privileged or copyrighted under applicable law. If you are not the intended recipient, do not read, copy, or forward this email message or any attachments. Delete this email message and any attachments immediately.



 1231954387.0