[volt-nuts] DIY Air bath

[Marvin E. Gozum]

Some DIY ideas for an environmental chamber.  Note, one thing most 
don't do is alter relative humidity.  To do that you can inject steam.

A key item is getting a chamber big enough for your need.  The heat 
control is fairly easy to design, many approaches.  Cooling to the 
40F ish area using a Peltier device is simple solution for a chamber 
but its cost efficiency and regulation could be a problem as the cuft 
of the chamber grows.  I think the heating part is generally easy and 
reliable, its getting the right size to save the labor of building 
one from scratch, that has both thermal glass, and insulation to build on.

In the eevblog.com post earlier, Dave uses a reptile incubator, there 
is a variant sold as a portable cooler/refrigerator under the Coleman 
name but the Peltier is set only to cool.  The ReptiPro 5000 or its 
clones, has reportedly unreliable electronics, either the Peltier or 
its thermostat are prone to fail early, but its worth a mod to get a 
good sized chamber that has the infrastructure to build both a 
heating and cooling chamber in one.

You can find incubators new or eBay that will save you the labor of 
building more tightly regulated heat only chambers.  Infant animal 
incubators, egg incubators and bio-lab incubators are typical search terms.

Toaster or convection ovens work for just heating, and the only value 
of changing or supplementing the thermostat is for tighter regulation 
and adding a fan to insure the heat is evenly distributed in larger chambers.

If you are lucky to find a used human baby incubator, and have the 
room for it, it has both the size, fan, and regulation for good even 
heating with portholes for working inside the chamber.






At 06:48 PM 1/13/2012, Bob Smither wrote:
>Will wrote:
> > Varying room temperature is a remarkable source of error when making
> > measurements in home environment. It would be nice to have a simple
> > air bath for the items under test. And even better if it was big
> > enough for the multimeter too. There are commercial units available
> > but very expensive.
> >
> > A DIY solution seems to be possible: a simple PI(D) controller built
> > around an NTC resistor, op-amp, pass-transistor and heater plus a fan
> > to keep the air moving inside the box. That approach allows
> > temperature higher than room temperature only but in most cases
> > stability is important and not the absolute value. A Peltier element
> > is almost as easy to drive as a heater resistor, but dissipated heat
> > probably makes the thermal design much more challenging.
> >
> > I am not an expert and there many questionmarks related air
> > circulation, thermistor location and probably other things I can't
> > even imagine.
>
>Hi Will,
>
>For years I have used a modified Rival brand consumer grade 
>convection oven as a
>test chamber.  Although I designed it for use up to 200C, it can be 
>used at any
>temperature above ambient.
>
>The oven has a usefully large "cooking" chamber - 16" x 14" x 
>8.5".  The heater
>and fan are in a section above the chamber separated by a metal 
>partition.  The
>heated air exits that section and enters the chamber through several slots in
>the metal partition.
>
>I redid the wiring so that the fan runs all the time and I designed 
>a simple on
>/ off controller to control the heater element.  The controller is in series
>with the oven bimetallic thermostat which serves as an over temperature safety
>cutoff - I just set it well above the temperature I set the controller to.
>
>The controller uses a diode that is placed in the chamber as the sensing
>element.  A comparator compares the diode voltage to a fixed level 
>derived from
>a reference and a 10 turn pot.
>
>The heater is switched by a solid state relay that is controlled by the
>comparator output.  The SSR has the nice feature of zero voltage turn on that
>reduces any electrical interference.
>
>I have tried different placements of the diode sensor.  If it is 
>located in the
>heater / fan section above the chamber the system is very over damped and it
>takes a long time to reach equilibrium (the diode "overreacts" to the power
>applied to the heater element and shuts down).  If it is located in 
>the chamber,
>the oven will overshoot the target temperature as the applied power is not
>sensed quickly enough.  The best location seems to be in the chamber but close
>(~1") to one of the slots where the heated air enters the chamber.
>
>Although the SSR is an on / off device, when close to the set point it behaves
>in almost an analog fashion as there is some 60 cycle pickup on the sense lead
>(that goes to the diode sensor) so in fact it operates at various 
>duty cycles as
>it applies power to the heater.
>
>I just ran an experiment with three thermocouples in the chamber at different
>positions (close to wall, in middle of chamber, ...).  The lowest 
>setting on my
>controller is above room temperature at 34C.  Here are the minimums 
>and maximums
>through several cycles of the controller at the three different locations as
>recorded on a Doric Trendicator 412A:
>
>Position 1: 33.8C - 34.2C
>Position 2: 33.7C - 34.2C
>Position 3: 33.7C - 34.3C
>
>I have used this oven for over 25 years.  Recently the fan motor 
>finally gave up
>but it was easy to replace.
>
>Hope this gives you some ideas.
>
>Bob Smither



Sincerely,



Marv Gozum
Philadelphia, PA