1
\$\begingroup\$

Saturation Current is one of parameters in Ebers Moll Model, but I can't find it in datasheets. I explore SC1627 and BC856 datasheets. How to find it?

As opposed to How do I saturate an NPN transistor? I want to know influence of saturation current for value of collector current at real conditions in compliance with Ebers Moll model

\$\endgroup\$
  • \$\begingroup\$ See electronics.stackexchange.com/a/13082/65586 which seems to address your specific question (read the comments, too.) \$\endgroup\$ – user65586 Mar 15 '16 at 15:27
  • 3
    \$\begingroup\$ Possible duplicate of How do I saturate an NPN transistor? \$\endgroup\$ – JRE Mar 15 '16 at 15:32
  • \$\begingroup\$ The question refers to finding a SPICE model parameter, not to operation of the transistor, so it's not a duplicate (at least not of those questions). \$\endgroup\$ – Spehro Pefhany Mar 15 '16 at 15:44
5
\$\begingroup\$

The easiest way to find reasonable numbers for the Ebers-Moll parameters is to look at existing SPICE models.

For example, the BC856 ISC = 1.633E-14 (NXP model).

ISC is base-collector saturation current.

There's also IS transport saturation current and ISE base-emitter leakage saturation current.

Deriving a reasonably accurate model from a BJT datasheet is non-trivial. You can find a quite lengthy description here.

\$\endgroup\$
  • 1
    \$\begingroup\$ Wow! That's very interesting! Thank you, Mr. @Spehro Pefhany \$\endgroup\$ – amste_av Mar 15 '16 at 17:10
1
\$\begingroup\$

You cannot find it because there is no "Saturation current" in a real BJT. There will be many mode parameters in an Ebers-Moll model which you will be unable to find in a datasheet.

Also note that there is no fixed point at which a BJT suddenly enters / goes out of saturation. It's more of a gradual thing. This behavior is not in the Ebers-moll model.

It's the same with diodes, you know the formula for a diode I guess, it has this current \$Is\$. It's not in any datasheet. Also, if you would try to measure it you will find that it varies over almost everything like forward current, dopings of the diode, temperature, whatnot. What you do in practice is assume a certain value for \$Is\$ such that we get a realistic forward voltage at a certain forward current.

It is probably the same with that Saturation current in the Ebers-Moll model. When modelling a transistor we tweak it such that it gives a transistor which has a realistic behaviour.

\$\endgroup\$
  • \$\begingroup\$ I understand you, but if we haven't any definite value of saturation current then how we can use Ebers Moll Model? In this case it's only theoretical abstraction which are applicated for ideal cases, isn't it? \$\endgroup\$ – amste_av Mar 16 '16 at 9:43
  • \$\begingroup\$ You have to read my answer carefully, you still couple this "saturation current" to a physical property of the transistor. In reality it is just like Sphero says in his answer and what I say about the diode equation's \$Is\$: it is a number to make the model match reality. This does not make the model useless, in fact it makes the model usable ! \$\endgroup\$ – Bimpelrekkie Mar 16 '16 at 10:13
  • \$\begingroup\$ Also: trying to match model parameters (like "saturation current" and Is) to physical properties of a transistor is a useless waste of time, like Sphero writes: model parameter extraction is a not trivial, this is because most model parameters are not directly related to the transistor's properties. \$\endgroup\$ – Bimpelrekkie Mar 16 '16 at 10:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.