[theory] help with diode math

[gaussmarkov]

hi,

i am in the process of figuring out some math describing how the VCO in the electric
mistress works and i think i have isolated one part of what i am missing down to the
following simpler question:

what is the differential equation describing the adjustment of a diode from one
equilibrium to another?

i can simulate what looks like an exponential path from one current-voltage state
to another current-voltage state, each consistent with the Shockley diode equation.
so i figure there must be something simple like the differential equation for capacitors,
but i have not been able to find it.

i suppose this is an unusual kind of forum question and i am sure there is another
forum set up somewhere that takes this kind of question regularly.  but this is the forum
that i know and there are definitely folks here who know the answer.

it's not OT because it's about the electric mistress.  the information will be useful for
understanding how to debug the LFO/VCO combination when it is not working properly,
which seems to happen to a some folks.

cheers, gm

[gaussmarkov]

i think i have stumbled upon the answer.  it looks like diodes have a capacitance that can be modeled as an ideal diode in parallel with a capacitor.  for example, the Vishay datasheet for a 1N914 says that the diode capacitance is 4pF.  in ltspice, the 1N914 model and the ideal diode in parallel with a 4pF cap give similar, though not exactly the same, transient responses.  so qualitatively and reasonably accurately, that seems to be how to think about it.

perhaps someone would explain the difference in capacitances reported in two datasheets?  the 1N914 has a diode capacitance parameter that is given the "symbol" Cd and the value 4pF.  the Fairchild datasheet for a 1N400x diode lists something called "total capacitance" with the "symbol" Ct and the value 15pF.  are these comparable or do they mean different things?

cheers, gm

[MR COFFEE]

gm,

Yes they are comparable. The difference in capacitance between the two diode data sheets is related to the differences between the two diodes.

The 1N914 is a small signal Si diode (200ma max iirc), designed for speed (yeah, I know it's not a superfast or Shottkey, but relative to a 1N400x series, it's fast).

The 1N400x series is a rectifier diode and is designed for use in power supplies and is intended for use at line frequencies. The higher capacitance is related to the higher current-carrying capacity and surge-current capability (1 amp max continuous iirc).

The reverse recovery time is another pertinent characteristic when you are trying to understand the finer points of the diode transfer curve. Seems there are some good tutorials on diodes on the net if you google for them.

Seems like Jack or RG did a piece on the knee characteristics some time back, too, addressing the "Ge knee is softer" claims. Apologies if it was somebody else.

mr coffee

[puretube]

not diode, but transistor...

[gaussmarkov]

@ mr coffee:  thanks!  the "reverse recovery time" phrase helps a lot in locating the right sort of tutorial using google.

not diode, but transistor...

ton, i was trying to make the mechanisim that i am trying to understand as simple as possible. hence the substitution of a diode for a transistor.  i was trying to understand why the output had an exponential rise when there was no cap involved in that part of the cycle.  so i was just looking at that phase and the diode action of the transistors in the LM311 comparator seems to be the source.  or am i over-simplifiying?

thanks for the link!  it confirms that i figured out the basic VCO mechanism properly, which is satisfying.