I have a dumb question on the use use of the 'Szikliai' pair in the switching mode.
The particular project i am working on would benefit from the low turn on voltage for the switching and the high current gain of this setup.

My question is ,,,, In the arrangement with the PNP power device and the NPN input device, is this the more suitable pair for switching the positive rail to the load. (The emitter of the power device would be connected to +Vcc. The collector to load)
I seem to be being told that the pairs take the form of the input device, but i am having trouble getting my head around that.
The input drive being high or low going is no concern for the circuit and is easily inverted.

Any help/explanations in an understandable fashion would be great.

Thanks
Norm

 

For high-side switching at the rail, the PNP driving an NPN would likely work better.
But if you want a high-side switch with low ON voltage and high impedance (no current) control you can't beat a P-MOSFET.

A Szikliai pair has a minimum ON voltage of more than 0.7V (one base-emitter drop plus the ON saturation voltage).
A MOSFET can have low milliohm ON resistance.

 

Thanks for the reply, the NPN power device is the way i will go then, it just seems backwards to me..........

I would like to use a PMOS but the switch on voltage is too high/low. The 0.65v is a benefit in this app.

Norm

 

................
I would like to use a PMOS but the switch on voltage is too high/low. The 0.65v is a benefit in this app.

What is the voltage?
What do you mean it's "too high/low"?
How could the 0.65V be a "benefit"?

 

Hey, don't get cranky, the object of the circuit is to 'self excite' a dynamo field coil from scratch.
The residual magnetism will allow the dynamo to generate about 0.7 volts when rotated, this can then switch on the field control via either a high gain PNP device or a complementary pair. The dynamo generates 7.2 volts when fully excited/controlled by the field coil.
To get a PMOS device to function here i would need to generate a voltage of about 5v or so to get the device to switch on wouldn't i?

The 0.65v turn on is therefore a huge advantage isn't it?

I realise i can, with greater complexity get a PMOS to work but i wanted to keep it simple.
As this was my first post on here, i don't expect people to be so stroppy!

Norm

 

As this was my first post on here, i don't expect people to be so stroppy!

Norm

Asking for clarification is not being "stroppy"
It's hard to provide really useful help unless you can get into a poster's head, people often forget to include essential context in their questions.

The only way to do that is to ask more questions.

 

I wasn't cranky, just a little frustrated, but sometimes I do live up to my self-imposed title.

You didn't tell me what the application was so I didn't understand what your limitations were.
How would I know that 0.65V is a benefit, when I had no knowledge of what the transistor was for?

It's common for newbies to ask questions about a project without telling what the project is, which makes it more difficult to give good answers.
Another annoyance is to ask questions about a pre-supposed solution to the problem instead of asking how best to solve the problem.
In the future if you fully explain what you are doing (including schematics) and what problems/questions you have with that, then we can give you better and faster answers without a lot of "20 questions" to get there.

 

I wasn't cranky, just a little frustrated, but sometimes I do live up to my self-imposed title.

You didn't tell me what the application was so I didn't understand what your limitations were.
How would I know that 0.65V is a benefit, when I had no knowledge of what the transistor was for?

It's common for newbies to ask questions about a project without telling what the project is, which makes it more difficult to give good answers.
Another annoyance is to ask questions about a pre-supposed solution to the problem instead of asking how best to solve the problem.
In the future if you fully explain what you are doing (including schematics) and what problems/questions you have with that, then we can give you better and faster answers without a lot of "20 questions" to get there.

Fair enough, as a new user of this forum i take your point. I should have been more explicit.
I'll have to look into how to post drawings too, that always makes life easier.

All the best
Norm

 

To the original poster, and any who come here after

Here is a good primer on the use of transistors as switches. If you scroll down past halfway point, you will find a section about H bridges using darlington TIP120's.
I know it doesn't directly address your specific situation but perhaps you will find it useful otherwise.

http://www.ermicro.com/blog/?p=423