Control, Thermostats and Language

CSGnet Archive CSG2010
1

[Fred Nickols (2010.05.23.0717 PDT)]

As I indicated I would, I posted the question below to some
internet groups.

Whenever the
subject of control comes up, someone is almost sure to introduce the
thermostat. Here’s a question for you. Pick only one
response. If you share your response with me or the list, I will tally
them and then share the “correct” response.

A thermostat acts
to control:

a. Room temperature

b. Furnace or A/C
operation (on/off)

c. The temperature it
senses

d. The temperature
that has been set

My
purposes were (a) to see how people responded and (b) to use the item to make
the point that what is controlled are perceptions; in the case of the
thermostat, that’s the sensed temperature.

So
far, no one has chosen c – the “correct” answer. Most
choose b and one has chosen a. Two chose d. Two people opted for
all four choices and one responded with an explanation of how a thermostat
works but did not choose a response. I expect more responses today and
tomorrow.

I
think there is a problem with language; specifically, with the term
“control.” For many people it does not carry any technical
meaning such as “bring a controlled variable to some reference value.”
For these same people it does seem to mean something along the lines of
“I can make it do what I want.” For them, that’s
turning the furnace or A/C on or off, or bringing the room temperature to some
set value, or even setting the desired temperature. They view themselves
as controlling all these things.

Moreover,
even with the technical usage, there is a blurred line between what the
thermostat controls and what I am trying to control. The thermostat controls
sensed temperature (i.e., it operates the furnace or A/C so as to bring the
sensed temperature to the setpoint temperature). I, on the other hand, am
trying to control what I sense to be the room temperature (i.e., which I do by
setting the setpoint to a temperature I want and leave the rest of the task to
the furnace or A/C).

I
have a hunch that sorting all this out will be tricky but worthwhile.

Regards,

Fred Nickols

Managing
Partner

Distance Consulting LLC

1558 Coshocton
Ave - Suite 303

Mount Vernon,
OH 43050-5416

fred@nickols.us

“Assistance at a
Distance”

2

[From Bill Powers (2010.05.23.1015 MDT)]

Fred Nickols (2010.05.23.0717 PDT) --

Excellent little project. I suggest developing it further. The
attached paper might provide some more ideas.

Best,

Bill P.

FolkBeliefs.PDF (1.02 MB)

3

[From Fred Nickols
(2010.05.24.1052 PDT)]

I will be responding later today
to the lists where I posed the thermostat question. I intend providing
the following explanations regarding the answers. Does anyone spot any
flaws in my explanations?

Does the thermostat act to control room temperature?

Technically, the answer is no. Room temperature is
affected as a result of the furnace adding or not adding warm air to the room
via the ducts from the furnace. And the thermostat does turn the furnace
on and off. But, in the strict sense of “control” as meaning
to bring something (in this case, room temperature,) to a specified value and
keeping it there, the answer is no, the thermostat does not control room
temperature.

Does the thermostat act to control furnace operation?

Again technically, the answer is no. It is true that
the thermostat switches the furnace on and off but this is simply a means of
starting or stopping the flow of warm air so as to control something else.

Does the thermostat act to control the sensed (i.e.,
measured) temperature?

Yes. The actions of the thermostat (turning the
furnace on and off) start and stop the flow of warm air into the room.
The thermostat has a sensing device that measures the temperature in the
immediate vicinity of the thermostat (which is why thermostat placement is so
important). The “immediate vicinity” of the thermostat
is not the same as the larger room. In fact, one can place a lamp or a
candle in the vicinity of the thermostat and the sensed or measured temperature
will be decidedly different from the temperature measured elsewhere in the
room. What the thermostat acts to control is the temperature it senses or
measures. It exercises this control by turning the furnace on or off.

Does the thermostat act to control the desired or set point
temperature?

No. The set point temperature is controlled or
adjusted by a human being who slides a lever to the desired value or, in the
case of a digital thermostat, enters the desired temperature by pressing a
button.

Regards,

Fred Nickols

Managing
Partner

Distance Consulting LLC

1558 Coshocton
Ave - Suite 303

Mount Vernon,
OH 43050-5416

www.nickols.us

fred@nickols.us

“Assistance at a
Distance”

···

From: Fred Nickols
[mailto:fred@nickols.us]
Sent: Sunday, May 23, 2010 7:18 AM
To: ‘Control Systems Group Network (CSGnet)’
Subject: Control, Thermostats and Language

[Fred Nickols (2010.05.23.0717 PDT)]

As I indicated I would, I posted the question below to some
internet groups.

Whenever the
subject of control comes up, someone is almost sure to introduce the
thermostat. Here’s a question for you. Pick only one
response. If you share your response with me or the list, I will tally
them and then share the “correct” response.

A thermostat acts
to control:

a. Room temperature

b. Furnace or A/C
operation (on/off)

c. The temperature it
senses

d. The temperature
that has been set

My
purposes were (a) to see how people responded and (b) to use the item to make
the point that what is controlled are perceptions; in the case of the
thermostat, that’s the sensed temperature.

So
far, no one has chosen c – the “correct” answer. Most
choose b and one has chosen a. Two chose d. Two people opted for
all four choices and one responded with an explanation of how a thermostat
works but did not choose a response. I expect more responses today and
tomorrow.

I
think there is a problem with language; specifically, with the term
“control.” For many people it does not carry any technical
meaning such as “bring a controlled variable to some reference
value.” For these same people it does seem to mean something along
the lines of “I can make it do what I want.” For them,
that’s turning the furnace or A/C on or off, or bringing the room
temperature to some set value, or even setting the desired temperature.
They view themselves as controlling all these things.

Moreover,
even with the technical usage, there is a blurred line between what the
thermostat controls and what I am trying to control. The thermostat
controls sensed temperature (i.e., it operates the furnace or A/C so as to
bring the sensed temperature to the setpoint temperature). I, on the
other hand, am trying to control what I sense to be the room temperature (i.e.,
which I do by setting the setpoint to a temperature I want and leave the rest
of the task to the furnace or A/C).

I
have a hunch that sorting all this out will be tricky but worthwhile.

Regards,

Fred Nickols

Managing
Partner

Distance Consulting LLC

1558 Coshocton
Ave - Suite 303

Mount Vernon,
OH 43050-5416

fred@nickols.us

“Assistance at a
Distance”

4

[From Rick Marken (2010.05.24.1255)]

Fred Nickols
(2010.05.24.1052 PDT)–

I will be responding later today
to the lists where I posed the thermostat question. I intend providing
the following explanations regarding the answers. Does anyone spot any
flaws in my explanations?

Not one. Looks great!

Best

Rick

···

Does the thermostat act to control room temperature?

Technically, the answer is no. Room temperature is
affected as a result of the furnace adding or not adding warm air to the room
via the ducts from the furnace. And the thermostat does turn the furnace
on and off. But, in the strict sense of “control” as meaning
to bring something (in this case, room temperature,) to a specified value and
keeping it there, the answer is no, the thermostat does not control room
temperature.

Does the thermostat act to control furnace operation?

Again technically, the answer is no. It is true that
the thermostat switches the furnace on and off but this is simply a means of
starting or stopping the flow of warm air so as to control something else.

Does the thermostat act to control the sensed (i.e.,
measured) temperature?

Yes. The actions of the thermostat (turning the
furnace on and off) start and stop the flow of warm air into the room.
The thermostat has a sensing device that measures the temperature in the
immediate vicinity of the thermostat (which is why thermostat placement is so
important). The “immediate vicinity” of the thermostat
is not the same as the larger room. In fact, one can place a lamp or a
candle in the vicinity of the thermostat and the sensed or measured temperature
will be decidedly different from the temperature measured elsewhere in the
room. What the thermostat acts to control is the temperature it senses or
measures. It exercises this control by turning the furnace on or off.

Does the thermostat act to control the desired or set point
temperature?

No. The set point temperature is controlled or
adjusted by a human being who slides a lever to the desired value or, in the
case of a digital thermostat, enters the desired temperature by pressing a
button.

Regards,

Fred Nickols

Managing
Partner

Distance Consulting LLC

1558 Coshocton
Ave - Suite 303

Mount Vernon,
OH 43050-5416

www.nickols.us

fred@nickols.us

“Assistance at a
Distance”

From: Fred Nickols
[mailto:fred@nickols.us]
Sent: Sunday, May 23, 2010 7:18 AM
To: ‘Control Systems Group Network (CSGnet)’
Subject: Control, Thermostats and Language

[Fred Nickols (2010.05.23.0717 PDT)]

As I indicated I would, I posted the question below to some
internet groups.

Whenever the
subject of control comes up, someone is almost sure to introduce the
thermostat. Here’s a question for you. Pick only one
response. If you share your response with me or the list, I will tally
them and then share the “correct” response.

A thermostat acts
to control:

a. Room temperature

b. Furnace or A/C
operation (on/off)

c. The temperature it
senses

d. The temperature
that has been set

My
purposes were (a) to see how people responded and (b) to use the item to make
the point that what is controlled are perceptions; in the case of the
thermostat, that’s the sensed temperature.

So
far, no one has chosen c – the “correct” answer. Most
choose b and one has chosen a. Two chose d. Two people opted for
all four choices and one responded with an explanation of how a thermostat
works but did not choose a response. I expect more responses today and
tomorrow.

I
think there is a problem with language; specifically, with the term
“control.” For many people it does not carry any technical
meaning such as “bring a controlled variable to some reference
value.” For these same people it does seem to mean something along
the lines of “I can make it do what I want.” For them,
that’s turning the furnace or A/C on or off, or bringing the room
temperature to some set value, or even setting the desired temperature.
They view themselves as controlling all these things.

Moreover,
even with the technical usage, there is a blurred line between what the
thermostat controls and what I am trying to control. The thermostat
controls sensed temperature (i.e., it operates the furnace or A/C so as to
bring the sensed temperature to the setpoint temperature). I, on the
other hand, am trying to control what I sense to be the room temperature (i.e.,
which I do by setting the setpoint to a temperature I want and leave the rest
of the task to the furnace or A/C).

I
have a hunch that sorting all this out will be tricky but worthwhile.

Regards,

Fred Nickols

Managing
Partner

Distance Consulting LLC

1558 Coshocton
Ave - Suite 303

Mount Vernon,
OH 43050-5416

www.nickols.us

fred@nickols.us

“Assistance at a
Distance”

–
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

5

[From Fred Nickols (2010.05.24.1547 PDT)]

Rick Marken (2010.05.24.1255)]

Fred Nickols (2010.05.24.1052 PDT)–

I will be responding later today to the lists where I
posed the thermostat question. I intend providing the following
explanations regarding the answers. Does anyone spot any flaws in my
explanations?

Not one. Looks great!

Best

Rick

Thanks, Rick. I think so too. However, I can easily
imagine the howls that are going to go up in relation to the thermostat not
controlling furnace operation. It’s not hard to envision someone
(and probably lots of them) saying, “What? You mean to tell me that
the thermostat turns the furnace off and on and that’s not controlling
furnace operation?”

Right now, my aim is to indicate that “control”
refers to bringing some variable to a specific state and keeping it
there. Even, then, I’m sure someone will say, “Well, how
about the ‘on’ and ‘off’ of the furnace as specific
states?” Temperature is a variable, hence a target for
control. The on-off state of the furnace also seems to be a variable
(with only two states). Hence, it too is a target for control and part of
the heating system of which the thermostat is also a part.

My original intent was to use the question to generate a little
interest and then use the “correct” answer to show how the
thermostat is controlling its “perception” (i.e., the measured or
sensed temperature) and how, essentially, we as closed-loop, negative feedback
systems, do pretty much the same thing (i.e., act in ways that control our
perceptions of some variable we’ve targeted for control). But I’m
not sure how to handle any objections about the furnace operation not being
controlled.

Got any ideas anyone?

Regards,

Fred Nickols

Managing Partner

Distance Consulting LLC

1558 Coshocton Ave - Suite 303

Mount Vernon, OH 43050-5416

www.nickols.us | fred@nickols.us

“Assistance at a Distance”

···

Does
the thermostat act to control room temperature?

Technically,
the answer is no. Room temperature is affected as a result of the furnace
adding or not adding warm air to the room via the ducts from the furnace.
And the thermostat does turn the furnace on and off. But, in the strict
sense of “control” as meaning to bring something (in this case,
room temperature,) to a specified value and keeping it there, the answer is no,
the thermostat does not control room temperature.

Does
the thermostat act to control furnace operation?

Again
technically, the answer is no. It is true that the thermostat switches
the furnace on and off but this is simply a means of starting or stopping the
flow of warm air so as to control something else.

Does
the thermostat act to control the sensed (i.e., measured) temperature?

Yes.
The actions of the thermostat (turning the furnace on and off) start and stop
the flow of warm air into the room. The thermostat has a sensing device
that measures the temperature in the immediate vicinity of the thermostat
(which is why thermostat placement is so important). The
“immediate vicinity” of the thermostat is not the same as the
larger room. In fact, one can place a lamp or a candle in the vicinity of
the thermostat and the sensed or measured temperature will be decidedly
different from the temperature measured elsewhere in the room. What the
thermostat acts to control is the temperature it senses or measures. It
exercises this control by turning the furnace on or off.

Does
the thermostat act to control the desired or set point temperature?

No.
The set point temperature is controlled or adjusted by a human being who slides
a lever to the desired value or, in the case of a digital thermostat, enters
the desired temperature by pressing a button.

Regards,

Fred Nickols

Managing Partner

Distance Consulting LLC

1558 Coshocton Ave - Suite 303

Mount Vernon, OH 43050-5416

www.nickols.us | fred@nickols.us

“Assistance at a Distance”

From: Fred
Nickols [mailto:fred@nickols.us]
Sent: Sunday, May 23, 2010 7:18 AM
To: ‘Control Systems Group Network (CSGnet)’
Subject: Control, Thermostats and Language

[Fred
Nickols (2010.05.23.0717 PDT)]

As
I indicated I would, I posted the question below to some internet groups.

Whenever
the subject of control comes up, someone is almost sure to introduce the
thermostat. Here’s a question for you. Pick only one
response. If you share your response with me or the list, I will tally
them and then share the “correct” response.

A
thermostat acts to control:

a. Room temperature

b. Furnace or A/C operation (on/off)

c. The temperature it senses

d. The temperature that has been set

My purposes were (a) to see how people
responded and (b) to use the item to make the point that what is controlled are
perceptions; in the case of the thermostat, that’s the sensed
temperature.

So far, no one has chosen c – the
“correct” answer. Most choose b and one has chosen a.
Two chose d. Two people opted for all four choices and one responded with
an explanation of how a thermostat works but did not choose a response. I
expect more responses today and tomorrow.

I think there is a problem with language;
specifically, with the term “control.” For many people it
does not carry any technical meaning such as “bring a controlled variable
to some reference value.” For these same people it does seem to
mean something along the lines of “I can make it do what I
want.” For them, that’s turning the furnace or A/C on or off,
or bringing the room temperature to some set value, or even setting the desired
temperature. They view themselves as controlling all these things.

Moreover, even with the technical usage,
there is a blurred line between what the thermostat controls and what I am
trying to control. The thermostat controls sensed temperature (i.e., it
operates the furnace or A/C so as to bring the sensed temperature to the
setpoint temperature). I, on the other hand, am trying to control what I
sense to be the room temperature (i.e., which I do by setting the setpoint to a
temperature I want and leave the rest of the task to the furnace or A/C).

I have a hunch that sorting all this out will
be tricky but worthwhile.

Regards,

Fred Nickols

Managing Partner

Distance Consulting LLC

1558 Coshocton Ave - Suite 303

Mount Vernon, OH 43050-5416

www.nickols.us | fred@nickols.us

“Assistance at a Distance”

–
Richard S. Marken PhD
rsmarken@gmail.com
www.mindreadings.com

6

[From Bill Powers (2010.05.24.1823 MDT)]

Fred Nickols (2010.05.24.1547 PDT) --

FN: Right now, my aim is to indicate that "control" refers to bringing some variable to a specific state and keeping it there. Even, then, I'm sure someone will say, "Well, how about the 'on' and 'off' of the furnace as specific states?" Temperature is a variable, hence a target for control. The on-off state of the furnace also seems to be a variable (with only two states). Hence, it too is a target for control and part of the heating system of which the thermostat is also a part.

This is why I have tried to promote a specific definition of control: A controls B if, for every disturbance of B, A changes its effect on B so as to cancel the effect of the disturbance.

CAR: If a wind starts blowing the car off course, the driver will change his effect on the car's direction to cancel the effect of the wind. The driver is controlling the course of the car.

FURNACE: If something turns the furnace off when the thermostat is telling it to be on, the thermostat will NOT alter its action so as to turn the furnace back on (it has no direct way to sense whether the furnace is on or off). The thermostat is not controlling the state of the furnace.

My original intent was to use the question to generate a little interest and then use the "correct" answer to show how the thermostat is controlling its "perception" (i.e., the measured or sensed temperature) and how, essentially, we as closed-loop, negative feedback systems, do pretty much the same thing (i.e., act in ways that control our perceptions of some variable we've targeted for control). But I'm not sure how to handle any objections about the furnace operation not being controlled.

Just say that disturbances of the furnace's state can't be resisted by the thermostat. If you turn the power to the furnace off, the thermostat has no way to turn it on again.

Also, you can point out that the thermostat has no particular duty cycle of the furnace (ratio of time on to total time) that it prefers. It has to vary the on-off ratio according to heat losses and gains -- open windows or ovens warming up -- in order to maintain a constant temperature. Controlling some variable is not the same thing as affecting, determining, or influencing it. Control has to do with keeping something in a particular state while disturbances are tending to change it.

Best,

Bill P.

7

[From Fred Nickols (2010.05.24.1913 PDT)]

I think I get most of what you're saying below, Bill, but I'm still puzzling
over something.

The thermostat opens or closes a switch that turns the furnace off or on.
If the furnace is off and if the sensed temperature drops below the set
point value then the thermostat will close the switch. Under ordinary
circumstances, that will turn on the furnace and move warm air into the
room.

However, if the furnace has been turned off manually, closing the switch in
the thermostat will not cause the furnace to start up. If I use a hair
dryer or ice cubes to raise or lower the temperature that is sensed or
measured by the thermostat, it will open or close the switch even though the
furnace is not turning on or off. More to the point, the thermostat is not
controlling the sensed/measured temperature.

I'm beginning to believe that all the thermostat is controlling is the
position of the switch that turns the furnace on or off and it can do so
correctly even if the furnace is off.

So what is controlling the sensed/measured temperature? I think that is the
entire heating system, not the thermostat alone.

What do you think?

Fred Nickols

···

-----Original Message-----
From: Control Systems Group Network (CSGnet)
[mailto:CSGNET@LISTSERV.ILLINOIS.EDU] On Behalf Of Bill Powers
Sent: Monday, May 24, 2010 5:51 PM
To: CSGNET@LISTSERV.ILLINOIS.EDU
Subject: Re: Control, Thermostats and Language

[From Bill Powers (2010.05.24.1823 MDT)]

Fred Nickols (2010.05.24.1547 PDT) --

FN: Right now, my aim is to indicate that "control" refers to
bringing some variable to a specific state and keeping it
there. Even, then, I'm sure someone will say, "Well, how about the
'on' and 'off' of the furnace as specific states?" Temperature is a
variable, hence a target for control. The on-off state of the
furnace also seems to be a variable (with only two states). Hence,
it too is a target for control and part of the heating system of
which the thermostat is also a part.

This is why I have tried to promote a specific definition of control:
A controls B if, for every disturbance of B, A changes its effect on
B so as to cancel the effect of the disturbance.

CAR: If a wind starts blowing the car off course, the driver will
change his effect on the car's direction to cancel the effect of the
wind. The driver is controlling the course of the car.

FURNACE: If something turns the furnace off when the thermostat is
telling it to be on, the thermostat will NOT alter its action so as
to turn the furnace back on (it has no direct way to sense whether
the furnace is on or off). The thermostat is not controlling the
state of the furnace.

My original intent was to use the question to generate a little
interest and then use the "correct" answer to show how the
thermostat is controlling its "perception" (i.e., the measured or
sensed temperature) and how, essentially, we as closed-loop,
negative feedback systems, do pretty much the same thing (i.e., act
in ways that control our perceptions of some variable we've targeted
for control). But I'm not sure how to handle any objections about
the furnace operation not being controlled.

Just say that disturbances of the furnace's state can't be resisted
by the thermostat. If you turn the power to the furnace off, the
thermostat has no way to turn it on again.

Also, you can point out that the thermostat has no particular duty
cycle of the furnace (ratio of time on to total time) that it
prefers. It has to vary the on-off ratio according to heat losses and
gains -- open windows or ovens warming up -- in order to maintain a
constant temperature. Controlling some variable is not the same thing
as affecting, determining, or influencing it. Control has to do with
keeping something in a particular state while disturbances are
tending to change it.

Best,

Bill P.

8

[From Bruce Gregory (2010.05.25.0628 EDT)]

[From Fred Nickols (2010.05.24.1913 PDT)]

So what is controlling the sensed/measured temperature? I think that is the
entire heating system, not the thermostat alone.

What do you think?

BG: I’m with you, Fred. But I am afraid that may be a strike against you.

Namaste,

Bruce

9

[From Bill Powers (2010.05.25.0704 MDT)]

Fred Nickols (2010.05.24.1913 PDT)

FN: So what is controlling the
sensed/measured temperature? I think that is the entire heating
system, not the thermostat alone.

BP: Yes, of course you’re right. The furnace is simply the output
function of the control system, the other parts being in the wall unit:
the comparator (the contacts that are closed when the sensed temperature
drops enough) and the sensor itself. People see the wall unit and think
that is the whole thermostat. But the wall unit couldn’t do much if the
controller didn’t include a furnace (located somewhere more appropriate
than on the wall of the living room, usually)

To see why the furnace isn’t controlled, all you have to do is see that
there is no sensor detecting the state of the furnace, and no output
function to use in correcting errors if the furnace is sensed as being in
the wrong state. In fact you could make the furnace into a lower-order
control system if you wanted to, for example by sensing the temperature
inside the furnace and making sure that the heat-exchange fins are at
reference temperature of exactly 600 degrees F when the furnace is on,
and a lower temperature when it’s off. But that whole control system
would still be only the output function of an air-temperature
controller.

Perhaps a more intuitive example is steering a car by turning the
steering wheel. Does the driver control the angle of the steering wheel?
As a lower-level control system, yes (as Dag Forssell can tell you, if
someone tugs on the steering wheel while you’re driving, you will resist
as energetically as needed). But in terms of steering the car, you can’t
have a preferred angle of the steering wheel and maintain it against
disturbances – if you want the car to stay on the road. The steering
wheel, whether part of a lower-order control system or not, is just the
output function of the car-steering system. There is no particular angle
at which you intend to hold it as a way of steering the car. You
vary the steering wheel angle to control the sensed
direction of the car. If you don’t vary the wheel angle, the wind will
blow the car off the road.

The thermostat varies the heat output of its furnace as a way of
controlling the sensed air temperature. If a lower-grade fuel is
used, the thermostat will just put out a little less heat for a little
longer on each duty cycle.

Best,

Bill P.

10

[From Fred Nickols (2010.05.25.0731 PDT)]

Thanks, Bill. In terms of my original questions, then, the
correct answer is “None of the above.”

The point about sensing is the key point.

Regarding your steering wheel example and Dag’s point
about someone tugging at the wheel, I would imagine that the controlled
variable isn’t the angle of the steering wheel but, rather, the unimpeded
or free control of the movement of the steering wheel. If someone were to
tug at the steering wheel while I was driving, I would also counteract that,
but I think I would be doing so to maintain control of the steering wheel, control
of steering, and control of the car. Maintaining a particular angle doesn’t
seem to me to be the controlled variable (although it might be while executing
a particular turn, say into a parking space).

Regards,

Fred Nickols

Managing Partner

Distance Consulting LLC

1558 Coshocton Ave - Suite 303

Mount Vernon, OH 43050-5416

www.nickols.us | fred@nickols.us

“Assistance at a Distance”

[From Bill Powers (2010.05.25.0704 MDT)]

Fred Nickols (2010.05.24.1913 PDT)

FN: So what is controlling the sensed/measured
temperature? I think that is the entire heating system, not the
thermostat alone.

BP: Yes, of course you’re right. The furnace is simply the output function of
the control system, the other parts being in the wall unit: the comparator (the
contacts that are closed when the sensed temperature drops enough) and the
sensor itself. People see the wall unit and think that is the whole thermostat.
But the wall unit couldn’t do much if the controller didn’t include a furnace
(located somewhere more appropriate than on the wall of the living room, usually)

To see why the furnace isn’t controlled, all you have to do is see that there
is no sensor detecting the state of the furnace, and no output function to use
in correcting errors if the furnace is sensed as being in the wrong state. In
fact you could make the furnace into a lower-order control system if you wanted
to, for example by sensing the temperature inside the furnace and making sure
that the heat-exchange fins are at reference temperature of exactly 600 degrees
F when the furnace is on, and a lower temperature when it’s off. But that whole
control system would still be only the output function of an air-temperature
controller.

Perhaps a more intuitive example is steering a car by turning the steering
wheel. Does the driver control the angle of the steering wheel? As a
lower-level control system, yes (as Dag Forssell can tell you, if someone tugs
on the steering wheel while you’re driving, you will resist as energetically as
needed). But in terms of steering the car, you can’t have a preferred angle of
the steering wheel and maintain it against disturbances – if you want the car
to stay on the road. The steering wheel, whether part of a lower-order control
system or not, is just the output function of the car-steering system. There is
no particular angle at which you intend to hold it as a way of steering the
car. You vary the steering wheel angle to control the sensed
direction of the car. If you don’t vary the wheel angle, the wind will blow the
car off the road.

The thermostat varies the heat output of its furnace as a way of controlling
the sensed air temperature. If a lower-grade fuel is used, the thermostat will
just put out a little less heat for a little longer on each duty cycle.

Best,

Bill P.

11

[From Bill Powers (2010.05.25.0855 MDT)]

Fred Nickols (2010.05.25.0731 PDT) --

FN: Regarding your steering wheel example and Dag's point about someone tugging at the wheel, I would imagine that the controlled variable isn't the angle of the steering wheel but, rather, the unimpeded or free control of the movement of the steering wheel.

BP: It's simpler that that, I think. The higher-level system controls the direction in which the car is going (or its position in a traffic lane). It does so by varying the reference signal being send to a lower control system that keeps the wheel at whatever angle is currently set as a reference level. When someone tugs at the wheel, the first error is at the lower level: the wheel moves a little, creating an error relative to the reference signal being received from the car-steering system. That immediately produces opposing forces from the muscles of the arms, before the steering system senses any error in the position of the car. If the disturbance is large enough to cause the wheel to move at all (that wouldn't take much force), the car will start to veer off course, and very soon the higher system will see an error in the direction or the position of the car in its lane. The higher system will change the reference position for the wheel-control system, just enough to keep the car in its lane even with the force still acting. A still-higher system will try to remove the disturbance, I would wager, by yelling WHAT THE HECK ARE YOU DOING??!!!. The person tugging on the wheel has to be careful when releasing the pressure, to allow the driver to reduce his opposing force slowly to zero. Letting go suddenly would leave the driver twisting the wheel to one side!

I actually did this to Dag on a freeway, having great faith in his control systems. He can tell you about it. I didn't tug very hard and was prepared to stop instantly if the experiment failed, and I reduced the amount of tug slowly at the end.

Best,

Bill P.

12

[From Dag Forssell (2010.05.25.10:30 PDT)]

[Bill Powers (2010.05.25.0855
MDT)]

I actually did this to Dag on a
freeway, having great faith in his control systems. He can tell you about
it. I didn’t tug very hard and was prepared to stop instantly if the
experiment failed, and I reduced the amount of tug slowly at the
end.

OK. This was when I brought Bill back to LAX in November 1991, as
mentioned in “Ten Minutes”.

I must have written about this on CSGnet. Sure enough, a search of the
Eudora archive for Powers, Freeway, and Airport (Eudora’s search function
is pretty good) yields this:
[From Dag Forssell (930618.1200)]

Bill Powers (930618.1000 MDT)

…The variables that are under control are under CONTROL.
There
is no doubt about a controlled variable when you find one. You
can’t simply take hold of a real controlled variable and move it
around. If you try you’ll have to struggle to get your way.
I once drove Bill to the airport on a Los Angeles freeway. We were
in the fast lane with cars all around. Bill asked: “Would you
like
to see a good, tight control system?” “Of course,” I
said. Bill
reached over from the passenger seat, grabbed the steering wheel
and pulled on it, rather hard. I tensed immediately. The car never
veered from my intended position in our lane more than a few inches.
Bill let go. Again, no deviation of the car’s position.
I understood the demonstration of a good, tight control system.Phil read CSGnet and I remember discussing this with him as he
polished People as Living Things. For his version of this story, see page

Best, Dag

13

[From Dag Forssell (2010.05.25.10:40 PDT)]

Having done my search, I failed to check out this one which also showed up.
This excerpt is part of a substantial post which bears rereading.

[From Bill Powers (970311.1939 MST)]

<snip>
The basic point is this: the disturbance must be applied in such a way that
it could be resisted successfully without excessive effort on the part of
the controller. This means using just enough disturbance to be sure you're
seeing the action that prevents it from having any effect. It means keeping
the speed of variations well within the bandwidth of control, so the person
can easily keep up with the changes. It means keeping the amplitude of the
disturbance small enough so that the person can easily maintain control.
Think of applying the disturbance through a rubber band, not directly to the
knot. That's why rubber bands were used in this demo: to make control
possible even while disturbances are being applied.

I've told this one before, but it bears repeating here. One day, while Dag
Forssell was driving me on the freeway from his house to the airport, I
asked him if he wanted to see a really tight control system. Naturally, he
said yes. So I took hold of the side of steering wheel and started pulling
down on it (making sure not to pull too suddenly). By increasing the pull at
a modest rate, I actually raised it to a pretty substantial level before Dag
called a halt to the experiment. The car, traveling along at 60 miles per
hour in traffic, naturally stayed right in its lane, its path not deviating
more than a foot. Dag opposed my disturbance very successfully -- I made
sure he did, because as a passenger in the same car I had a personal
interest in NOT disrupting his control of the car. When you think about
applying disturbances in doing the Test, imagine that you could be killed if
the controller actually failed to control as well as usual.

···

Subject: Applying disturbances