Peter L. Peres <plp@actcom.co.il> wrote:
> 
> On Thu, 14 Oct 2004, Sebastian Garcia wrote:
> 
> > I know that it works doing so. But again, the oven free response is
> > exponential, and it's impossible to fit in the paste manufacturer's
> > recommended error margin for the thermal profile. Moreover, the
> > plateau-shaped profiles were developed at the earlier days of reflow
> > soldering, when the ovens where provided only with IR heaters. Today,
> > the recommended profile for convection ovens replaces the plateau zone
> > with a slope shape.
> 
> The oven response is essentially linear if the heaters are significantly
> hotter than the oven (this is true when you turn the heaters full on in a
> toaster oven). 


As i said in my first post, i measured the oven temperature step response 
(full power) with thermocouples (in air) and plotted the data... exponential.
I can understand if you *approximate* as linear the first seconds of the 
exponential, but i see no reason for a linear temperature response in a 
"standard" oven. 


> It is exponential if the heater is controlled to a lower
> temperature (as it is in production ovens). So you are both right ... and
> you can 'fix' the slope between linear and exponential by setting the
> heater temperature. Very hot = more linear, cooler = exponential, exactly
> set to target temp = asymptotic (will 'never' reach target temp). I have
> plotted several ovens at several (constant) power settings and my
> conclusion is that you can approximate any (heating) temperature profile
> you wish by setting specific power levels for specific amounts of time, in
> open loop. For more rapid cooling slope opening the oven door for a
> specific opening width (using a ceramic wedge shim of the right size)
> works great. So with the right practice, you can 'be' the controller. I
> used my wristwatch as timer and a B&K thermocouple instrument for
> temperature reading.


That's what i need to avoid, to be the controller!
Opening the door for fast cooling may seem to work but it creates abrupt
temperature gradients that may lead to component stressing.

 
> > I wish to follow the manufacturer's recommendations, to guarantee
> > long-term reliability of the soldered boards.
> 
> That's always good but can you achieve it with homemade equipment ?


I don't know. 
I just need to do my job and try to achieve the best approximation to the 
solder paste manufacturer's recommendations. Then, we'll see...

Thanks for your input.


Regards,

S.-
_______________________________________________
http://www.piclist.com
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist