RC's and thermal modeling

G

George Herold

Guest
So from a previous thread,

https://groups.google.com/group/sci.electronics.basics/browse_thread/thread/6563c2244c745710?hl=am#

it was suggested by John L. that I try some RC modeling. I did that
today and the results were not what I expected. I sent a step voltage
into a single RC (R=1 C=1) and then into a string of 10 R/C's. (R=0.1
and C=0.1). With the long string the voltage was a bit slower at the
beginning, but it reached a 1/e value quicker than the single RC. My
colleague says I have reproduced the error function (erf). And this
looks about right.

Very interesting,

Thanks John

George H.
 
On Jan 20, 4:36 pm, George Herold <gher...@teachspin.com> wrote:
So from a previous thread,

https://groups.google.com/group/sci.electronics.basics/browse_thread/...

it was suggested by John L. that I try some RC modeling.  I did that
today and the results were not what I expected.  I sent a step voltage
into a single RC (R=1 C=1) and then into a string of 10 R/C's.  (R=0.1
and C=0.1).  With the long string the voltage was a bit slower at the
beginning, but it reached a 1/e value quicker than the single RC.  My
colleague says I have reproduced the error function (erf).  And this
looks about right.

Very interesting,

Thanks John

George H.
Click to expand...
Here's the LTspice file if you are interested... (might be easier to
make it yourself?)

Version 4
SHEET 1 1012 680
WIRE 144 -496 80 -496
WIRE 224 -400 224 -432
WIRE 304 -400 304 -432
WIRE 384 -400 384 -432
WIRE 80 -272 80 -496
WIRE 144 -272 80 -272
WIRE 224 -176 224 -208
WIRE 304 -176 304 -208
WIRE 80 -64 80 -272
WIRE 144 -64 80 -64
WIRE 224 32 224 0
WIRE 80 128 80 -64
WIRE 80 128 -32 128
WIRE 112 128 80 128
WIRE 272 192 192 192
WIRE 352 192 272 192
WIRE 432 192 352 192
WIRE 512 192 432 192
WIRE 592 192 512 192
WIRE 672 192 592 192
WIRE 752 192 672 192
WIRE 832 192 752 192
WIRE 912 192 832 192
WIRE -32 224 -32 208
WIRE 192 224 192 192
WIRE 192 224 -32 224
WIRE -32 256 -32 224
WIRE 0 0 0 0
FLAG -32 256 0
FLAG 224 32 0
FLAG 224 -176 0
FLAG 304 -176 0
FLAG 224 -400 0
FLAG 304 -400 0
FLAG 384 -400 0
SYMBOL res 208 112 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R1
SYMATTR Value 0.1
SYMBOL res 288 112 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R2
SYMATTR Value 0.1
SYMBOL res 368 112 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R3
SYMATTR Value 0.1
SYMBOL res 448 112 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R4
SYMATTR Value 0.1
SYMBOL res 528 112 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R5
SYMATTR Value 0.1
SYMBOL res 608 112 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R6
SYMATTR Value 0.1
SYMBOL res 688 112 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R7
SYMATTR Value 0.1
SYMBOL res 768 112 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R8
SYMATTR Value 0.1
SYMBOL res 848 112 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R9
SYMATTR Value 0.1
SYMBOL res 928 112 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R10
SYMATTR Value 0.1
SYMBOL cap 176 128 R0
SYMATTR InstName C1
SYMATTR Value .1
SYMBOL cap 256 128 R0
SYMATTR InstName C2
SYMATTR Value .1
SYMBOL cap 336 128 R0
SYMATTR InstName C3
SYMATTR Value .1
SYMBOL cap 416 128 R0
SYMATTR InstName C4
SYMATTR Value .1
SYMBOL cap 496 128 R0
SYMATTR InstName C5
SYMATTR Value .1
SYMBOL cap 576 128 R0
SYMATTR InstName C6
SYMATTR Value .1
SYMBOL cap 656 128 R0
SYMATTR InstName C7
SYMATTR Value .1
SYMBOL cap 736 128 R0
SYMATTR InstName C8
SYMATTR Value .1
SYMBOL cap 816 128 R0
SYMATTR InstName C9
SYMATTR Value .1
SYMBOL cap 896 128 R0
SYMATTR InstName C10
SYMATTR Value .1
SYMBOL voltage -32 112 R0
WINDOW 0 -74 28 Left 0
WINDOW 3 -235 -36 Left 0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V1
SYMATTR Value PULSE(0 10 0 1m 1m 100 200 1)
SYMBOL res 240 -80 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R11
SYMATTR Value 1
SYMBOL cap 208 -64 R0
SYMATTR InstName C11
SYMATTR Value 1
SYMBOL res 240 -288 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R12
SYMATTR Value .5
SYMBOL cap 208 -272 R0
SYMATTR InstName C12
SYMATTR Value .5
SYMBOL res 320 -288 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R13
SYMATTR Value .5
SYMBOL cap 288 -272 R0
SYMATTR InstName C13
SYMATTR Value .5
SYMBOL res 240 -512 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R14
SYMATTR Value .333
SYMBOL cap 208 -496 R0
SYMATTR InstName C14
SYMATTR Value .333
SYMBOL res 320 -512 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R15
SYMATTR Value .333
SYMBOL cap 288 -496 R0
SYMATTR InstName C15
SYMATTR Value .333
SYMBOL res 400 -512 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R16
SYMATTR Value .333
SYMBOL cap 368 -496 R0
SYMATTR InstName C16
SYMATTR Value .333
TEXT -272 280 Left 0 !.tran 0 10 0 1m

George H.
 
George Herold a écrit :
So from a previous thread,

https://groups.google.com/group/sci.electronics.basics/browse_thread/thread/6563c2244c745710?hl=am#

it was suggested by John L. that I try some RC modeling. I did that
today and the results were not what I expected. I sent a step voltage
into a single RC (R=1 C=1) and then into a string of 10 R/C's. (R=0.1
and C=0.1). With the long string the voltage was a bit slower at the
beginning, but it reached a 1/e value quicker than the single RC. My
colleague says I have reproduced the error function (erf). And this
looks about right.

Very interesting,

Thanks John

George H.
Click to expand...
Heat flux (power) is more seen as a current which is integrated by the
thermal mass (cap). Then you have two thermal resistances: the one from
your heated structure between each blob of capacitance node as you've
drawn, and the convection one, which is tied from each cap node and a
voltage source representing ambient temperature.


--
Thanks,
Fred.
 
On Thu, 20 Jan 2011 18:57:00 -0800 (PST), George Herold
<gherold@teachspin.com> wrote:

On Jan 20, 7:03 pm, Fred Bartoli <myname_with_a_dot_inbetw...@free.fr
wrote:
George Herold a écrit :





So from a previous thread,

https://groups.google.com/group/sci.electronics.basics/browse_thread/...

it was suggested by John L. that I try some RC modeling.  I did that
today and the results were not what I expected.  I sent a step voltage
into a single RC (R=1 C=1) and then into a string of 10 R/C's.  (R=0.1
and C=0.1).  With the long string the voltage was a bit slower at the
beginning, but it reached a 1/e value quicker than the single RC.  My
colleague says I have reproduced the error function (erf).  And this
looks about right.

Very interesting,

Thanks John

George H.

Heat flux (power) is more seen as a current which is integrated by the
thermal mass (cap). Then you have two thermal resistances: the one from
your heated structure between each blob of capacitance node as you've
drawn, and the convection one, which is tied from each cap node and a
voltage source representing ambient temperature.

--
Thanks,
Fred.- Hide quoted text -

- Show quoted text -

Yeah, That's right Fred. I was just simulating a step change in the
temperature. Simulating the other losses is an interesting question.
It looks to me that other resistive losses, will just be a loss of
voltage and not a change in the time response. No that's not right.
I'll loss more near the hot end than at the far end...

I can simulate that too. Say is there a way to get numbers out of an
LTspice simulation?

George H.
Click to expand...
You could use the "ltline" lossy transmission line model with L==0, or
there's a couple of pure RC line models, urc and urc2. The
metalization lines in ICs are so lossy that L is essentially zero, so
they behave like thermal conductors.

There are no thermal inductors, which is why temperatures don't
overshoot in purely thermal-conductive systems. There *is* a tricky RC
circuit that has above-unity gain at some frequencies... I wonder if
the thermal version would be interesting.

John
 
On Jan 20, 7:03 pm, Fred Bartoli <myname_with_a_dot_inbetw...@free.fr>
wrote:
George Herold a écrit :





So from a previous thread,

https://groups.google.com/group/sci.electronics.basics/browse_thread/...

it was suggested by John L. that I try some RC modeling.  I did that
today and the results were not what I expected.  I sent a step voltage
into a single RC (R=1 C=1) and then into a string of 10 R/C's.  (R=0.1
and C=0.1).  With the long string the voltage was a bit slower at the
beginning, but it reached a 1/e value quicker than the single RC.  My
colleague says I have reproduced the error function (erf).  And this
looks about right.

Very interesting,

Thanks John

George H.

Heat flux (power) is more seen as a current which is integrated by the
thermal mass (cap). Then you have two thermal resistances: the one from
your heated structure between each blob of capacitance node as you've
drawn, and the convection one, which is tied from each cap node and a
voltage source representing ambient temperature.

--
Thanks,
Fred.- Hide quoted text -

- Show quoted text -
Click to expand...
Yeah, That's right Fred. I was just simulating a step change in the
temperature. Simulating the other losses is an interesting question.
It looks to me that other resistive losses, will just be a loss of
voltage and not a change in the time response. No that's not right.
I'll loss more near the hot end than at the far end...

I can simulate that too. Say is there a way to get numbers out of an
LTspice simulation?

George H.
 
John Larkin wrote:
On Thu, 20 Jan 2011 18:57:00 -0800 (PST), George Herold
gherold@teachspin.com> wrote:

On Jan 20, 7:03 pm, Fred Bartoli<myname_with_a_dot_inbetw...@free.fr
wrote:
George Herold a écrit :





So from a previous thread,

https://groups.google.com/group/sci.electronics.basics/browse_thread/...

it was suggested by John L. that I try some RC modeling. I did that
today and the results were not what I expected. I sent a step voltage
into a single RC (R=1 C=1) and then into a string of 10 R/C's. (R=0.1
and C=0.1). With the long string the voltage was a bit slower at the
beginning, but it reached a 1/e value quicker than the single RC. My
colleague says I have reproduced the error function (erf). And this
looks about right.

Very interesting,

Thanks John

George H.

Heat flux (power) is more seen as a current which is integrated by the
thermal mass (cap). Then you have two thermal resistances: the one from
your heated structure between each blob of capacitance node as you've
drawn, and the convection one, which is tied from each cap node and a
voltage source representing ambient temperature.

--
Thanks,
Fred.- Hide quoted text -

- Show quoted text -

Yeah, That's right Fred. I was just simulating a step change in the
temperature. Simulating the other losses is an interesting question.
It looks to me that other resistive losses, will just be a loss of
voltage and not a change in the time response. No that's not right.
I'll loss more near the hot end than at the far end...

I can simulate that too. Say is there a way to get numbers out of an
LTspice simulation?

George H.

You could use the "ltline" lossy transmission line model with L==0, or
there's a couple of pure RC line models, urc and urc2. The
metalization lines in ICs are so lossy that L is essentially zero, so
they behave like thermal conductors.

There are no thermal inductors, which is why temperatures don't
overshoot in purely thermal-conductive systems. There *is* a tricky RC
circuit that has above-unity gain at some frequencies... I wonder if
the thermal version would be interesting.

John
Click to expand...
Yes it would--as in, breaking the second law of thermodynamics by making
heat flow from cold to hot.

Unfortunately there's no way to wire the two thermal masses in series--a
thermal mass is always a _grounded_ capacitor.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058

email: hobbs (atsign) electrooptical (period) net
http://electrooptical.net
 
On Jan 20, 10:32 pm, John Larkin
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
On Thu, 20 Jan 2011 18:57:00 -0800 (PST), George Herold





gher...@teachspin.com> wrote:
On Jan 20, 7:03 pm, Fred Bartoli <myname_with_a_dot_inbetw...@free.fr
wrote:
George Herold a crit :

So from a previous thread,

https://groups.google.com/group/sci.electronics.basics/browse_thread/....

it was suggested by John L. that I try some RC modeling. I did that
today and the results were not what I expected. I sent a step voltage
into a single RC (R=1 C=1) and then into a string of 10 R/C's. (R=0.1
and C=0.1). With the long string the voltage was a bit slower at the
beginning, but it reached a 1/e value quicker than the single RC. My
colleague says I have reproduced the error function (erf). And this
looks about right.

Very interesting,

Thanks John

George H.

Heat flux (power) is more seen as a current which is integrated by the
thermal mass (cap). Then you have two thermal resistances: the one from
your heated structure between each blob of capacitance node as you've
drawn, and the convection one, which is tied from each cap node and a
voltage source representing ambient temperature.

--
Thanks,
Fred.- Hide quoted text -

- Show quoted text -

Yeah, That's right Fred.  I was just simulating a step change in the
temperature.  Simulating the other losses is an interesting question.
It looks to me that other resistive losses, will just be a loss of
voltage and not a change in the time response.  No that's not right.
I'll loss more near the hot end than at the far end...

I can simulate that too.  Say is there a way to get numbers out of an
LTspice simulation?

George H.

You could use the "ltline" lossy transmission line model with L==0, or
there's a couple of pure RC line models, urc and urc2. The
metalization lines in ICs are so lossy that L is essentially zero, so
they behave like thermal conductors.
Click to expand...
There are special cases (I think) with ballistic phonons, that travel
as some 'coherent thing'. There must be something like inductance
(..momentum), but it's rare. I don't really understand (never
measured) second sound in helium

George H.
There are no thermal inductors, which is why temperatures don't
overshoot in purely thermal-conductive systems. There *is* a tricky RC
circuit that has above-unity gain at some frequencies... I wonder if
the thermal version would be interesting.

John- Hide quoted text -

- Show quoted text -
Click to expand...
 
On Jan 20, 11:13 pm, Phil Hobbs
<pcdhSpamMeSensel...@electrooptical.net> wrote:
John Larkin wrote:
On Thu, 20 Jan 2011 18:57:00 -0800 (PST), George Herold
gher...@teachspin.com>  wrote:

On Jan 20, 7:03 pm, Fred Bartoli<myname_with_a_dot_inbetw...@free.fr
wrote:
George Herold a crit :

So from a previous thread,

https://groups.google.com/group/sci.electronics.basics/browse_thread/....

it was suggested by John L. that I try some RC modeling.  I did that
today and the results were not what I expected.  I sent a step voltage
into a single RC (R=1 C=1) and then into a string of 10 R/C's.  (R=0.1
and C=0.1).  With the long string the voltage was a bit slower at the
beginning, but it reached a 1/e value quicker than the single RC.  My
colleague says I have reproduced the error function (erf).  And this
looks about right.

Very interesting,

Thanks John

George H.

Heat flux (power) is more seen as a current which is integrated by the
thermal mass (cap). Then you have two thermal resistances: the one from
your heated structure between each blob of capacitance node as you've
drawn, and the convection one, which is tied from each cap node and a
voltage source representing ambient temperature.

--
Thanks,
Fred.- Hide quoted text -

- Show quoted text -

Yeah, That's right Fred.  I was just simulating a step change in the
temperature.  Simulating the other losses is an interesting question..
It looks to me that other resistive losses, will just be a loss of
voltage and not a change in the time response.  No that's not right.
I'll loss more near the hot end than at the far end...

I can simulate that too.  Say is there a way to get numbers out of an
LTspice simulation?

George H.

You could use the "ltline" lossy transmission line model with L==0, or
there's a couple of pure RC line models, urc and urc2. The
metalization lines in ICs are so lossy that L is essentially zero, so
they behave like thermal conductors.

There are no thermal inductors, which is why temperatures don't
overshoot in purely thermal-conductive systems. There *is* a tricky RC
circuit that has above-unity gain at some frequencies... I wonder if
the thermal version would be interesting.

John

Yes it would--as in, breaking the second law of thermodynamics by making
heat flow from cold to hot.

Unfortunately there's no way to wire the two thermal masses in series--a
thermal mass is always a _grounded_ capacitor.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058

email: hobbs (atsign) electrooptical (period) nethttp://electrooptical.net- Hide quoted text -

- Show quoted text -
Click to expand...
Yeah, the charge Q is the thermal energy stored in the thing. It
seems like it would make sense to use the absolute temperature scale
(K). Then it's easy to see that discharging a thermal capacitor
involves tieing one end to absoulte zero.

George H.
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top