Aluminum-layer PCBs

Copper-clad Al is a useful interface for some power smd circuits. Its major limitation is interconnection. These have to be minimized, compounded and functional in the higher-temperature reflow.

Socketing is highly impractical, considering the cost of higher current female parts, so repair/removal is a major headache - usually all attachments being better off scrapped on failure.

Interface to heat conducting surface also requires serious mechanical thin'in.

You may find these used incidentally inside higher-powered commodity SSR and other power semiconductor modules, capped by a filled polymer housing that guides placement of faston or screw-bearing terminals.

Al is also found in the internal layers if some smaller power printed circuit assemblies (laptops etc) in order to distribute spot temperatures on printed wiring. Vias require predrilling, to fill with uncommitted kevlar in the autoclave. Not cheap.

On Wed, 09 Mar 2016 10:47:25 -0800, John Larkin Gave us:

For CE most of those companies "self certify".

Which must mean, buy a bag of stickers.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

A multi-channel light controller with a backup battery power supply. The two bottom plates are boring, each is just a replicated set of

12 TRIAC/optotriac pairs for the 230V channels. The upper contains all the 12V halogen lamp channels. On the right there is a set of 16 OmniFETs (with internal over-current and over-temperature protection circuits). The four big MOSFETs in the center are the main synchronous rectifier (with 40 amps capacity, i.e. 480W) fed by a toroid tranformer. Two other FETs form an SSR which connects a 13.8V Pb-acid battery in the case of power shortage. There is also a high-current Micrel LDO which acts as a charger.

The whole thing proves that the IMS substrate is a good choice -- I've made the PCBs myself using the classic thermal transfer approach. The soldering part was fun, though. :-)

The device have been working correctly for about 2 years of continuous operation and is cool even under the full load, but, frankly, I don't consider it to be a success -- the left part is overly complex. Later I've invented a much simpler design based on ~300kHz small-core magamps, with similar dimensions.

Best regards, Piotr

This is for theater work?

--
 Thanks, 
    - Win

On Wed, 09 Mar 2016 02:30:29 -0500, DecadentLinuxUserNumeroUno wrote as underneath :

Thank you for that useful link... C+