[time-nuts] 5070B once more.... (actually 5370A fans)

Magnus Danielson magnus at rubidium.dyndns.org
Fri May 22 19:06:08 UTC 2009


Hi!

> Regarding the comments below on the 5370:  there are always two
> questions with temperature:  meeting spec and reliability.  Instruments
> vary as to which is a bigger issue.  Some have temperature
> proof measurement techniques that will work virtually until
> something burns up, so you can get lulled into a false sense of
> security just because the instrument meets spec.

I think most people would agree that reliability comes in the first 
room, specs in the second. It is fairly meaningless to have super-spec 
when you know that your box is approaching dysfunctional at a high 
speed, and you don't know when what effect will kick in to fiddle with 
your performance before it says "poof" on you.

Infact, cooling (for reliability) and performance to some degree can 
become troublesome, as the noise of the airflow and vortexes can affect 
performance and forced air onto ovens will make very efficient 
connection between ambient temperature and oven shield, so any shift 
there will quickly and effectively cool or heat the oven. Self convected 
boxes has longer time constants than forced air boxes.

Many times when I "investigate" the design of boxes I react on their 
inability to design for reliability. I react on very basic design errors 
such as mechanical mounting of heavy components (solder is not long term 
stable fixing when heat and vibration comes into play) and heat (when 
the heat sink forms a heat-pocket into which all heat gathers and stays 
rather than distribute the heat sources so dissapation through self 
convection becomes easy is just another example).

Repairing failed gear can be rewarding in terms of understanding failure 
modes. Becoming experienced in various forms of failures is as important 
as understanding how to perform the task of the project.

Cheers,
Magnus



More information about the time-nuts mailing list