summaryrefslogtreecommitdiffstatshomepage
path: root/src/lib/netlist/documentation/mainpage.dox.h
diff options
context:
space:
mode:
Diffstat (limited to 'src/lib/netlist/documentation/mainpage.dox.h')
-rw-r--r--src/lib/netlist/documentation/mainpage.dox.h122
1 files changed, 61 insertions, 61 deletions
diff --git a/src/lib/netlist/documentation/mainpage.dox.h b/src/lib/netlist/documentation/mainpage.dox.h
index 053271f64ae..686a77913e9 100644
--- a/src/lib/netlist/documentation/mainpage.dox.h
+++ b/src/lib/netlist/documentation/mainpage.dox.h
@@ -17,15 +17,15 @@
The following example shows a typical connection between several devices:
- +---+ +---+ +---+ +---+ +---+
- | | | | | | | | | |
- | D | | D | | D | | D | | D |
- | | | | | | | | | |
- +-O-+ +-I-+ +-I-+ +-T-+ +-T-+
- | | | | |
- +-+---------+---------+---------+---------+-+
- | rail net |
- +-------------------------------------------+
+ +---+ +---+ +---+ +---+ +---+
+ | | | | | | | | | |
+ | D | | D | | D | | D | | D |
+ | | | | | | | | | |
+ +-O-+ +-I-+ +-I-+ +-T-+ +-T-+
+ | | | | |
+ +-+---------+---------+---------+---------+-+
+ | rail net |
+ +-------------------------------------------+
A rail net is a net which is driven by exactly one output with an
(idealized) internal resistance of zero.
@@ -39,29 +39,29 @@ The system differentiates between analog and logic input and outputs and
analog terminals. Analog and logic devices can not be connected to the
same net. Instead, proxy devices are inserted automatically:
- +---+ +---+
- | | | |
- | D1| | D2|
- | A | | L |
- +-O-+ +-I-+
- | |
- +-+---------+---+
- | rail net |
- +---------------+
+ +---+ +---+
+ | | | |
+ | D1| | D2|
+ | A | | L |
+ +-O-+ +-I-+
+ | |
+ +-+---------+---+
+ | rail net |
+ +---------------+
is converted into
- +----------+
- | |
- +---+ +-+-+ | +---+
- | | | L | A-L | | |
- | D1| | D | Proxy | | D2|
- | A | | A | | | |
- +-O-+ +-I-+ | +-I-+
- | | | |
- +-+---------+--+ +-+-----+-------+
- | rail net (A) | | rail net (L) |
- +--------------| +---------------+
+ +----------+
+ | |
+ +---+ +-+-+ | +---+
+ | | | L | A-L | | |
+ | D1| | D | Proxy | | D2|
+ | A | | A | | | |
+ +-O-+ +-I-+ | +-I-+
+ | | | |
+ +-+---------+--+ +-+-----+-------+
+ | rail net (A) | | rail net (L) |
+ +--------------| +---------------+
This works both analog to logic as well as logic to analog.
@@ -74,61 +74,61 @@ and transistors again.
Instead, the following approach in case of a pure terminal/input network
is taken:
- +---+ +---+ +---+ +---+ +---+
- | | | | | | | | | |
- | D | | D | | D | | D | | D |
- | | | | | | | | | |
- +-T-+ +-I-+ +-I-+ +-T-+ +-T-+
- | | | | |
- '+' | | '-' '-'
- +-+---------+---------+---------+---------+-+
- | Calculated net |
- +-------------------------------------------+
+ +---+ +---+ +---+ +---+ +---+
+ | | | | | | | | | |
+ | D | | D | | D | | D | | D |
+ | | | | | | | | | |
+ +-T-+ +-I-+ +-I-+ +-T-+ +-T-+
+ | | | | |
+ '+' | | '-' '-'
+ +-+---------+---------+---------+---------+-+
+ | Calculated net |
+ +-------------------------------------------+
Netlist uses the following basic two terminal device:
- (k)
- +-----T-----+
- | | |
- | +--+--+ |
- | | | |
- | R | |
- | R | |
- | R I |
- | | I | Device n
- | V+ I |
- | V | |
- | V- | |
- | | | |
- | +--+--+ |
- | | |
- +-----T-----+
- (l)
+ (k)
+ +-----T-----+
+ | | |
+ | +--+--+ |
+ | | | |
+ | R | |
+ | R | |
+ | R I |
+ | | I | Device n
+ | V+ I |
+ | V | |
+ | V- | |
+ | | | |
+ | +--+--+ |
+ | | |
+ +-----T-----+
+ (l)
This is a resistance in series to a voltage source and paralleled by a
current source. This is suitable to model voltage sources, current sources,
resistors, capacitors, inductances and diodes.
\f[
- I_{n,l} = - I_{n,k} = ( V_k - V^N - V_l ) \frac{1}{R^n} + I^n
+ I_{n,l} = - I_{n,k} = ( V_k - V^N - V_l ) \frac{1}{R^n} + I^n
\f]
Now, the sum of all currents for a given net must be 0:
\f[
- \sum_n I_{n,l} = 0 = \sum_{n,k} (V_k - V^n - V_l ) \frac{1}{R^n} + I^n
+ \sum_n I_{n,l} = 0 = \sum_{n,k} (V_k - V^n - V_l ) \frac{1}{R^n} + I^n
\f]
With \f$ G^n = \frac{1}{R^n} \f$ and \f$ \sum_n G^n = G^{tot} \f$ and \f$k=k(n)\f$
\f[
- 0 = - V_l G^{tot} + \sum_n (V_{k(n)} - V^n) G^n + I^n)
+ 0 = - V_l G^{tot} + \sum_n (V_{k(n)} - V^n) G^n + I^n)
\f]
and with \f$ l=l(n)\f$ and fixed \f$ k\f$
\f[
- 0 = -V_k G^{tot} + sum_n( V_{l(n)} + V^n ) G^n - I^n)
+ 0 = -V_k G^{tot} + sum_n( V_{l(n)} + V^n ) G^n - I^n)
\f]
These equations represent a linear Matrix equation (with more math).
@@ -136,7 +136,7 @@ These equations represent a linear Matrix equation (with more math).
In the end the solution of the analog subsystem boils down to
\f[
- \mathbf{\it{(G - D) v = i}}
+ \mathbf{\it{(G - D) v = i}}
\f]
with G being the conductance matrix, D a diagonal matrix with the total