View Single Post
Old 08-28-2011, 10:40 AM   #20
lawn psycho
Senior Member
 
Join Date: Jul 2009
Location: On the move...
Posts: 987
Thanks: 113
Thanked 248 Times in 133 Posts
Default

Quote:
Originally Posted by LIforrelaxin View Post
As for any of the work arounds here, etc. that I keep reading... I recomend none of them... No offense guys... but the link for using a secondary battery in parallel??? that is asking for a problem, and with the electronics in the cars today you are asking for an even bigger problem.... As for the 9V solution... I would like to understand more about this? but quite frankly I think your fooling yourself thinking that a 9V battern is keep your memory alive....
The problem with this type of setup is you have numerous parts with regulation and blocking. 9V is several diode drops below 12V. In most cases we assume voltage dropping to maybe the 9.8V range.

The reason why people think it works is we overbuild our designs. Period. However all our devices have given current density limitations through subtrate contacts to ground. Current actually flows from - to + as some of you probably recall.

Take 3V across a low ohmic contact and guess what happens during that transient? Look at the simple equation for capacitor current and doing this is ugly. A parasitic BJT in a power MOSFET only needs 0.7V to turn on and destroy itselt.

If you over power the contacts you can get electromigration. If in a radio, this can cause issues that may not be destructive but could be subtle. Phase locked loops can get out of whack, reception range effected, you name it. You may not think you did damage but you did.

In design we see the field returns and get to understand failure modes. Nothing is fool-proof as hard as we try. We never do biased reliability testing at 9V for auto applications which should tell you something.

Many technicians will claim I am full of it because they "get away with it". However as a designer of high power applications with automotive and appliance products this is terrible practice. We have to put stuff in our designs just so that when people do bad things to parts we can demonstrate why the failure was user related and not our part. This is also why warranty claims get denied.

In todays systems we have a lot of detection systems to look for shorts and undervoltage protection because it's such an issue. Most of the issues are from the assembly process and when maintenance is done on the car.

Off too bed. zzzzzz
lawn psycho is offline   Reply With Quote