MPP

Kean, the curve tracer can be setup in many ways. I have it running now on my new HP Pavilion g7 machine and the cluttering is gone. Here is a re measurement of the BC140, 160 i found fascinating.
 

Attachments

  • BC140-160.JPG
    BC140-160.JPG
    124 KB · Views: 304
Sorry, i do not have any BC550, 560 with me but at home i have a lot of cause.
They are most popular here in Europe, more or less in every German HiFi amplifier since the 70th. I have the experince that they work well in mirrors. A problem in very high impedance mirrors can be the high Hfe although on the other hand high impedance is what we want. BCs come of cause in all varieties of voltges and Hfes.
 
I seem to recall NXP BC3x7 had worse quasi-saturation than the OnSemi or Fairchild parts.

It doesn't seem quasi-saturation is a parameter manufacturers deem important, so it would not surprise me if parts from one were inferior in this aspect. NXP, OnSemi and Fairchild all seem to give different Cob for the BC327.

Do you know who made your BC3x7?

Vcesat for the BC3x7 up to 50mA or even 100mA would be useful. We need to be able to apply these metrics to the measurements:

1: Vce point where Ic falls to 90%. For the BC3x7 this appears to be about 130mV.
2: Vce at which the first metric falls significantly, again by say 90%. For the BC5xx this would be somewhere around 20mA I think.

These metrics define the minimum Vce and Ic for high collector impedance. Knowing this information will be liberating to those that are paranoid of quasi-saturation. There will be no need to fear running at low Vce as long as you pick the right transistor.

For TO92 devices, I would request an Ic range from 500uA to 100mA, and a Vce range from 0V to 2V.
 
Last edited:
I buy from Fairchild and Diotec and found that the Diotecs matched better in Hfe, PNP to NPN and also had a tighter match in the same gender, whatever that means.
Tell our friend Sergio that On Semi gives the Cob of the BC327 goes up from 8pF at -10V, Vce to 20pF at -0.1V Vce, so the tendency is that Cob goes up when Vce goes down.
Is that new news ? I do not know.
 
Quasi-saturation could also be defined directly in terms of collector impedance. For instance collector impedance after several V will be relatively constant. We can define the beginning of quasi-saturation as the region where collector impedance falls to 90%. I think, if not too tedious to determine, this will be a more effective metric for #1.

The Kmultiplier is an interesting case. It seems that Early affect modulates Vbe directly, because my Kmultiplier sees a low base source impedance. This was unexpected when Early affect is usually talked about in terms of collector impedance or base current. But in the Kmultiplier we have a situation where collector impedance and base current should be negligible, yet quasi-saturation has a large impact on filtering performance. Early voltage is often described as B-C leakage, but it does not behave like this. Rather it seems to be that Vce directly modulates Vbe and produces altered base current and Hfe as a byproduct.