OK, some basic on SMPS design:

   1. All SMPS designs have losses, some more than others. This results is 
   heat being generated and devices get warm. The two devices that may exhibit 
   this are the switch FET and inductor.
   In a good SMPS, the FET will remain cool-ish and so will the inductor. 
   This design uses an old-ish chip that is not very efficient, but should be 
   good enough.
   2. Do you actually have a problem? How do you know? In the light of 
   point 1. above, your FET will get warm, but how hot is yours? Quantify the 
   issue - is the FET at 30C? 40C? 50C? 
   3. If the FET is actually getting very hot, i.e. too hot to touch, it 
   may be due to internal losses (RDSon for that FET is quite high, so it may 
   get quite warm) or it may be a sub-spec FET.... 
   4. Sometimes in SMPS designs, you can get RF oscillation - it may be 
   that for some reason that is happening here...

...however the designer of that clock is a knowledgeable guy who has build 
a load of clocks- first thing should be to ask him.  Have you done that?

When tacking an issue like this, firstly establish whether you actually DO 
have a issue - speak to the guy that designed & built it. Remember that 
SMPSs are fickle beasts and sometimes misbehave and need care & feeding- 
judging by your apparent level of knowledge, this is not something you 
should do yourself - speak to the designer,

Secondly, forget about drilling holes and fans etc. for now - absolutely NO 
WAY should that be necessary - this is a proven design by a good supplier - 
if you feel its not right, SPEAK TO HIM !

Thirdly. a little knowledge is a dangerous thing - forget the random stuff 
about "electro-migration" - it simply is complete rubbish in this context. 
To quote from a recent Intel research paper:

"Electromigration (EM) remains a serious problem for the reliability of 
VLSI integrated circuits, and will become an even more serious issue as 
future IC's employ linewidths below 0.2 micron...."


Note the scale we are talking about here - 0.2 micron is one five 
thousandths of a millimetre, i.e. 7.9 MILLIONTH of an inch. The tracks on 
the PCB on that clock are at least 1000 times bigger. i.e. Forget about it.

So, like all engineering and other problems:

   1. Establish whether you actually have a problem or whether what you are 
   observing is expected behaviour
   2. If indeed you believe that you have a problem, quantify the behaviour 
   (measure temperatures etc.)
   3. Speak to the designer to resolve the issue - give him that data from 
   item 2 - , and...
   4. ... if that doesn't work, return the product for a refund (assuming 
   your haven't done anything silly like drill holes & fit a fan).
   5. That's it

HTH

Nick

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send an email to [email protected].
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/5ee52223-bdda-4494-a8d2-1fea39b0c4d3%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to