High Power TL432?

What's wrong with the application note for using a higher current pass transistor as a series regulator? App note 10.3 fig 10-5 in my copy.

Nothing wrong, just trying to avoid plan B. A is TL432, C is MJ11011. Need B in between.

MJ11011 needs 5V base voltage and 1A base current. Unless there is a triple Darlington available.

I figure 500mA should drive around .5A base and 10A collector.

Good question. The TL43x High-Current circuit's shown in Figure 29 on my own datasheet, last revised in November 2018:

It seems we talked about the circuit's 2N222 misnomer in years past. A couple of open questions:

1. What's the name for two transistors topologically connected as such?
2. What's the circuit's maximum "high current?" Danke,

BTW, page 3 of this data sheet is wrong: 11/13/15 should be PNP and 12/14/16 should be NPN.

I ordered PNP (TIP127) last month and they shipped me NPN (TIP122). Didn't check before. Too late to protest on ebay. Waiting for another order to come.

Double Darlington. I need Triple Darlington for 10A to 20A.

Assuming that's a typo and they missed a '2'. 2N2222 is 0.6A.

Paralleling shunt regulators is a generically bad idea.

Boost it with a transistor.

But why? If they are closely matched, would be any worst than parallel MOSFET.

If you are intending to make a high-current shunt regulator, and you put 5 small shunt regs in parallel, the one with marginally lower voltage will hog all the current and fry. Paralleling zeners has the same problem.

Paralleled mosfets don't share current well in linear mode. As switches, their similar Rds-on gives pretty good current sharing.

How about a small (couple ohms) current limiting resistor?

I am trying to make a small PCB under the TO-3 case. A (SOT23) and C (TO-3). B between them would need to be a big TO-220.

Ideal case is a Tarlington (Triple Darlington) with smaller base current.

More ideal case is a 20A TL432 in TO-3.

The dynamic impedance and voltage tolerance determine current sharing. It looks bad for hard-paralleled parts. A couple of ohms per chip would help a lot, as would using the 0.5% parts. I assume the 1% parts are bin-outs.

But exactly what do you want to do? Build an X? volts 500 mA shunt regulator?

It looks easy to boost the TL with a transistor, or better a mosfet if there's enough voltage available for the gate drive.

A schematic of your situation would help. Too many possibilities.

That is way too cryptic to make any sense of. What are you trying to do?

You are confused. Darlingtons drop more voltage across the device but at 10A collector current I'd expect Vbe(sat) ~ 2v and Vce(sat) ~ 2.5v.

That is a darlington pair with a min gain somewhere between 200 (@ 30A) and 1000 (@ 20A). It should be fine at 10A with a 100mA base current.

The device gain should be around 3000 at 10A (it would be even more nearer 8000x if you used an NPN version). See DC current gain vs Ic.

We can sort them out with matching reference resistors.

Drive 5V 1A of the MJ11011

Just the classic Triple Transistors circuit, usually with TO-92, TO-220 and TO-3, but i want to use SOT-23 and TO-3 only.

I explain a bit more in other posts.

That's the small signal DC current gain. We still need a substantial base current. The spec says 1A base current, no data on min or typical.

The 432 looks topologically like an NPN transistor with very high beta (*). To get a higher-current shunt regulator, the OP can just turn it into a Darlington:

Cathode to collector, anode to base, resistor from base to emitter to take the min anode current from the 432 feedback pin used as normal.

Watch out for the transistor dissipation!

Cheers

Phil Hobbs

(*) The LM4041 and LM385 look like PNPs instead--you connect the feedback to the anode to get 1.2V. That's sometimes very useful.

If you add voltage dividers er reg, their tolerances maks things worse. Sharing a common divider is bad too.

Well, I said "exactly." Can't help without understanding the issue.

Almost all with 100mA limit.

I guess the question is what's needed to drive the MJ11011 types.

And do it twice for Tarlington.

Probably need a TO-220 for 0.5A to 1A between the SOT23 and TO-3.

(Sorry for the worse than usual spelling. I just had cataract surgery and can't see well through this silly perforated shield. And I think they got my focal length wrong.)

Data sheet, fig 29.

Yes, assuming the SOT-23 can handle 1/4W. Over 12V range, it might only be able to sink 20mA (0.25/12). I might need to parallel a few buffering 2907s (0.5A over 5V, which is the MJ1101X base voltage range)