Early voltage

Thread Starter

Nik

Joined May 20, 2006
55
{ I'm not at college so this isn't really a home-work question, but topic resembles a college assignment...}

How many data-points should I manually collect to make a reasonable estimate of a BJT's Early voltage ?

I know the Early voltage is the X-axis intercept extrapolation of 'active' region of Collector current vs Collector Emitter voltage for multiple Ib.

Snag is the BJTs I hope to evaluate are the Q2s of 6n139 optocouplers' split darlington.

Their data-sheet treats the device as a 'black box', for which no Q1/Q2 info or Spice model is available. Took me a LOT of research to learn that LED-->photo-diode CTR is ~0.0016 (~0.16%). I must assume that Q1 & Q2 may be optimised as pre-amp and driver, so are non-identical. I cannot do much about the 'buried' Q1 beyond back-calculating its Beta(hFE) but Q2's EBC are all accessible.

My guess would be six data-points minimum per chip, all at one Ib or as prials, with errors dropping as the sq_root of the count. I was planning to take a dozen or so readings per chip, but where should I stop ?

Measuring a dozen chips' Beta(hFE) values would give a fair guide to typical gain, but each chip's Early voltage must be extrapolated into next county...

Suggestions welcome...
 

Thread Starter

Nik

Joined May 20, 2006
55
FWIW, I've hooked the chips up to an Atlas component mini-analyser. There's a slight drift in Beta (hFE) readings if repeated, but usually no more than +/- 1%.

Number in sample = 25.
Ic = 2.5 mA. BE = 0.82 V

Mean = 101.1
Range 92 ~~110
SD(n-1) = 5.0
.......
Given the LED --> PhotoDiode CTR = ~0.0016 (~0.16%)
{ OptoElectronics Applications Manual (c)1977 Hewlett-Packard p3.7 and their 6n136 model ;-}

With 6n139 LED cathode connected to Q2 emitter and both collectors linked, the Atlas accepted it as an odd darlington, reported the beta. Mean of 25 = 7.3, range 6~~9. Calculating each Q1 beta gave Mean of 25 = 45.1, range 37~~53 with SD(n-1) = 3.6.

Disclaimer: These are based on single measurements points per chip. Collecting data over a range of LED currents and Vcc may give significantly different results. I'll have to do that to get the Early voltage for Q2...

My head hurts...
 
Last edited:

Thread Starter

Nik

Joined May 20, 2006
55
I got non-convergent tangents, giving VAF ranging from 70~~170. After much hair-tearing and deep searching, I've learned that this is a 'well-known' issue in devices with small areas and/or self-heating.
;-(
 

Thread Starter

Nik

Joined May 20, 2006
55
The basic problem with modelling the 6n139 is that Q1's base is not accessible, so *only* the product LED==> photo-diode CTR x Q1 hFE is available...

(CTR = Current Transfer Ratio ;-)

Or so I thought...

Last night, I had a rush of blood to the brain: What if I fed the LED a few milliamps and put my DVM across Q1's emitter and collector ? No Vcc, just the photo-diode's output ? Would its photo-current flow through Q1's base/emitter PN ?

So, today, I tried it. I bridged Q1 emitter to Q2 emitter as latter is connected to substrate. I put an LED in series with the opto's LED to see current was flowing. A 9V battery and 680R, 1K0 or 1K8 fed 9.05, 6.21 or 3.49 mA by DVM. I measured 53.4, 35.4 and 18.6 uA photo-current across Q1 emitter and collector.

These gave CTR of 0.0059, 0.0057 & 0.0053, equivalent to ~0.5 %.

This is approx three (3) times the CTR from the old H-P reference, thrice the quoted figure for the data-sheet and model of 6n136...

If this ratio is representative, it means that a Q1's hFE is only ~15, in marked contrast to the Q2s' measured hFE of ~100...

D'uh, can you see any hole in my methodology or logic ??
 

Thread Starter

Nik

Joined May 20, 2006
55
Some-one kindly pointed out that I'm only seeing part of the photo-current as it will divide between the two PN junctions. So, my data is nonsense. Why so much current went that-away is a puzzle. Also, the gains of the two stages should be similar, so ~100 per my measurements of Q2. This means that my researched '0.15~~0.16%' CTR is nonsense, too.

I've just found a 6n140 model at http://www.irtc-hq.com/pspicemodellibrary.html
The 6n140 has a photo-diode driving a Darlington, but the Q2 has no accessible base. Instead, an internal EB resistor ~5k aids pulse recovery.

The model uses a tweaked LED, two generic 2n2222 NPNs and a polynomial transfer function for curve matching. Yup, it is not a 'physical' model...
 

Ghar

Joined Mar 8, 2010
655
One issue you're going to have is that CTR isn't constant with LED current.

The specification will be for a specific drive current at a specific temperature and with a specified VCE voltage on the phototransistor.
Change any of those and your CTR can be way off.
Most optocouplers aren't specified very well either, since the part is several components glued together and it's not very well controlled.
The CTR can range between individual parts by huge factors, like from 200% to 2000%...
 
Top