150
poorly
Sure, if the swing is small enough. Though I suppose care mostly near zero amps anyway.
Cheers
Phil Hobbs
The thermal compensation is almost independent of swing. Since the diff-pair transistors cool by the same amount (from nominal at "maximum heat"), it's good to close to saturation or cutoff.
about 150
poorly
It looks like it can stand an input swing up to maybe 50 mV before it gets seriously nonlinear. And V- can be low, like -1V, and the max pair current can be low, 1 mA or less.
Interesting problem, for such a simple circuit.
about 150
poorly
And so easy to calculate. ...Jim Thompson
heat"),
It only works for small signals, though, because once you get away from V_CE = VCC/2, the differential dissipation is no longer quadratic in deltaV_BE. It does help over a considerable range.
Cheers
Phil Hobbs
mv,
If Vi is the voltage due to the sensed current and Iv is the current applied to the diff pair due to the drain voltage, the differential signal current, Isig, is...
Isig = Iv*TANH(Vi/(2*kT/q))
So a significant TC, if the diff pair sees any heating at all.
So, given the fist-full of parts, why not just buy an off-the-shelf cheapy multiplier like a 1496 (or modern equivalent) and lose the TC? ...Jim Thompson
mv,
OK, why multiply when adding is easier?
This behaves like a classic foldback current limiter. Since SOAR curves are generally not true constant-power curves (less power is allowed at high voltages in most mosfets) we can pull down the X1 endpoint a bit to accommodate that.
May as well add the cap to account for thermal taus (different time lines on the SOAR graph) and toss in heatsink temperature, too.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required