1
\$\begingroup\$

I have an issue related to Baker Clamp application for PNP Transistor. I know how to implement the Baker Clamp for NPN Transistor. However I do not know how to do it for PNP. So could anyone can suggest me a schematic?

Secondly, can the Baker Clamp reduce the turn-off time of PNP transistor?

Thank you very much in advance.

\$\endgroup\$
  • \$\begingroup\$ to convert NPN to PNP just reverse everything. (all voltages, all diodes etc) \$\endgroup\$ – Jasen Feb 26 '18 at 18:42
4
\$\begingroup\$

How about this one?

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
2
\$\begingroup\$

Standard 2 diode version of a Baker Clamped NPN transistor on the left.

Switch all the arrows round, and viola, the PNP version on the right.

Omit D1/D3 and replace D2/D4 by shottky or germanium for the one transistor version. This is the normal way the technique is used today.

Add another diode anti-parallel to D1/D3 for the 3 diode version. This is faster than the two diode version, as the extra diode provides a path for base charge to be abstracted by the driving circuit.

schematic

simulate this circuit – Schematic created using CircuitLab

The clamp will speed up switch off of a PNP, as it prevents collector saturation and so excess charge storage.

\$\endgroup\$
1
\$\begingroup\$

The Baker clamp uses carefully selected Schottky to raise Vce >Vce(sat) and Silicon diodes to raise Vce(max) and to lower BC pF somewhat like a Schottky Transistor used in 74Sxx and 74LSxx but at the expense of Pd drop like a Darlington. This raises the VI product while also raising the transition frequency.

Most recent IC to use this I know is here. http://home.mira.net/~gnb/audio/lme49810.html

It is important to realize when rCE drops to conduct more current, the equivalent Collector capacitance Cce and storage charge increases which is an RC=T (approx. limitation) . The same is true in cascading power FETs with T=RonCoss and RgCiss so there becomes an optimal critical ratio of Ron/Rg for fastest yet efficient speeds . In slow SMPS it is common to see Ron/Rg be 1000:1 where in fastest high power FET designs this sometimes reduces to 10:1 in Ron Input driver/Output driver ratios.

This is also apparently true in optimized Baker designs where the Ic/Id has been suggested around 4:1 which to me implies a component power ratio of each.

Perhaps as well just as ESR*Pd(max) =1 is somewhat a design constant in all diodes with some reduction from 1 to 1/2 due to improved SMT thermal design and some rising to 1.5 due to poor bulk resistance which accounts for all the tolerance spreading of Vf/If at rated current tolerances often to +-50%.

I realize this is more than you asked for, but I thought others may find useful. If not voted as useful, I may reduce my answers in future. (Typing on an iPod is hard,with frequent incorrect spelling subs. but then I can be resting on the beach!)

\$\endgroup\$

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.