Re: I-Roast 2 - what's with all the fuss?

I think the fuss is though the iRoast has a programmable interface, it still
is a toy of an interface. From true process control position, it doesn't
come close to that definition. It still is 'Kentucky windage' at best.

Let me explain.

From a process control definition, one would have beginning and ending
temperature setpoints and a time to accomplish them. This sets up a ramp,
either up or down, for the controller to follow. In heat treatment, where my
experience comes from, there may be many ramps, both up and down, to achieve
a particular crystalline structure. Then the heat controller, a PID
controller, would adjust the output to the heater to achieve the ramp based
on a thermocouple(s) within the load. Most don't know what PID stands for.
The output of a PID algorithm is based on the sum of three separate
equations; the P or proportional component, which means how far is the
actual temperature from the setpoint. The I or integral calculates how long
and far the temp is from setpoint and will continue to build if the temp
lags the setpoint for any length of time. And the D or derivate, looks at
how fast the temp is rising or cooling and makes a negative adjustment so
the temperature doesn't overshoot the setpoint.

If the iRoast was truly programmable or the engineers at Heathware
understood process control, then the interface would use ambient temp for
the beginning setpoint and then each segment temp would be the end
temperature setpoint and the time programmed in would create the ramp. They
would also put a thermocouple in the beans to really know what was
happening. Then roasts would be truly repeatable and one could truly nuance
the best or desirable qualities out of the beans.

As far as I can tell, the temp controller is strictly 'I' or integral. It
slowly build to setpoint and it takes a long time to overshoot. Then I don't
know what it does. My kill-a-watt meter jumps all over the place when it
overshoots. The fan usage should be more controllable also. Right now it is
on high for 3 minute OR below 390F OR if the temp is way above setpoint and
it panics and turns the fan on.

Given that, I don't know if the iRoast is built to really follow a ramp and
do true programmability. What we get may be a compromise so the unit doesn't
end up in a puddle like some toon from Roger Rabbit. For $175US it does a
lot, but it could do a lot, lot more.

Jay


.

Relevant Pages

  • Re: [lm-sensors] 2.6.15: lm90 0-004c: Register 0x13 read failed (-1)
    ... Next driver resets SMBus and it works for some time ... This is the driver for the SMB Host controller on Acer Labs Inc. M1541 and M1543C South Bridges. ... "Initialize the base address of the i2c controller"); static unsigned short ali15x3_smba = 0; static int ali15x3_setup{u16 a; unsigned char temp; /* Check the following things: ...
    (Linux-Kernel)
  • Current progress pictures of my brew sculpture...
    ... The HLT has 4 couplings welded to it, ... The mash tun is a simple igloo with a copper slotted manifold in ... I have a temp controller that monitors the hlt and mash temps, ...
    (rec.crafts.brewing)
  • Re: PID tuning for air roasters
    ... controller that I am not using. ... "With the proportional band alone, ... equilibrium at some point away from the main setpoint. ... so that it maintains the ramp without overshooting. ...
    (alt.coffee)
  • Re: Brewtus I and II Owners // "PID" and plumb-in NOW
    ... The boiler has two surfaces - one touching water & one touching air. ... Apparently the stock controller does have several degrees of swing when ... swing when I hooked a PID to the II. ... less swing of intra-shot temp stability so temp control is simpler. ...
    (alt.coffee)
  • Re: Windows 2000 keeps locking up
    ... your suggestions in my original post. ... I have a temp. ... I did check the BIOS temp reading, but BIOS actually locks up as soon ... I disabled the UDMA ATA-100 controller in the BIOS ...
    (microsoft.public.win2000.general)
Loading