[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Help-gnucap] Re: NTC and PTC

From: nbs_r
Subject: Re: [Help-gnucap] Re: NTC and PTC
Date: Fri, 09 Jun 2006 23:15:05 +0900
User-agent: Thunderbird 1.5 (X11/20060317)

Hello Markus,

I don't know Gnucap syntax and all capabilities so I can't give you a
final solution. The example I gave you is likely to work in Hspice,
which has some extensions to the spice2/3 syntax.

If I understood you correctly, you want to simulate the effect of
current flow through an NTC device, causing power dissipation, operating
temperature change and then resistance change. Is it OK? If so, it is
the very thing that _cannot be directly done_ in spice(like) simulators,
ie. by using built-in temperature control (because these controls are

Instead, you can map thermal effects into electrical domain (as in
solution #2). Just map the operating temperature to the voltage node,
heat flow to the electrical current, thermal resistance to electrical
resistance and thermal capacitance to electric capacitance. Look at the
following circuit (sorry, hspice-like description again):

iNTC n1 n2 cur='v(n1,n2)/(resistance_value+tc1_param*v(temp_val))'
iheat 0 temp_val cur='p(iNTC)'
Ctemp temp_val 0 'thermal_capacitance'
Rtemp temp_val 0 'thermal_resistance'

Where v(temp_val) represents temperature difference between the NTC
device and environment.

I am looking forward to see working NTC model for Gnucap. :-)

Markus Feldmann wrote:
> Does this 2. Method work if i don't sweep the temp, but the current changes?
> In my simulation i don't sweep the temp but the current will be raised,
> and therefore the resistance of the NTC goes down while the temp of this
> NTC is growing up.
> Does your circuit above work for me?

The new circuit should be more useful as it models temperature change
caused by current flow.

> Another thing i don't understand your circuit.
> I_ntc1 is a current source.

It is actually a resistor with a temperature controlled resistance.
(note that the current value is proportional to voltage across nodes).

> v_temperature is a voltage source but i didn't see nodes.

Nodes are 'temp_val' and '0'.

> The cur=... doesn't exist in gnucap.

Well, I'm fresh to gnucap so I don't know how to obtain similar

reply via email to

[Prev in Thread] Current Thread [Next in Thread]