Well it might seem to work in simulation but it shouldn't
Hi and thanks a lot indeed for the always kind and fundamental support
But this is a limit for sure of the user but also of the SW
In a ideal situation the check that you have performed should be also carried out in automatic by the SW itself
The PNP in the way i have connected if i understand well it cannot work at all ... then why the sim runs ? even the FFT is nice ... very nice
Do you mean upside down ? because the signal should enter from the base
The idea is to replace the second bjt with a mosfet that they say works very nicely in class A.
I would have used a P-channel mosfet with a NPN input ... but i have not found the complement of the 510 ... the only mosfet i know of and that could work fine i suppose
The fact that a circuit can simulate and then not work is dramatic for me
Do you mean upside down or what ?
Thanks again, gino
Last edited:
Swap the emitter and collector over
The models in LTSpice (as I understand it... I'm no real expert on the intricacies of LT) are what might be termed first order models in that they model the basic properties of the device in normal use. You could make the models ever more complex but for this type of application its not needed. For example a transistor in LT will happily run while dissipating several kilo watts of power. It won't suddenly conduct and break down if you apply 100 volts to a 30 volt model. They are the sort of limitations that you as a designer have to know first, and realise that just because it works in simulation, it doesn't automatically mean it will be OK for real. You still need that practical knowledge of circuit design.
Did you know that many transistors will actually work when reverse biased and still exhibit gain. They only work like that at very low voltages though. Take the voltage higher and they conduct and would fail. Muting transistors are one type where that property is exploited, and special devices are made to operate that bit better under those special conditions than ordinary devices.
The models in LTSpice (as I understand it... I'm no real expert on the intricacies of LT) are what might be termed first order models in that they model the basic properties of the device in normal use. You could make the models ever more complex but for this type of application its not needed. For example a transistor in LT will happily run while dissipating several kilo watts of power. It won't suddenly conduct and break down if you apply 100 volts to a 30 volt model. They are the sort of limitations that you as a designer have to know first, and realise that just because it works in simulation, it doesn't automatically mean it will be OK for real. You still need that practical knowledge of circuit design.
Did you know that many transistors will actually work when reverse biased and still exhibit gain. They only work like that at very low voltages though. Take the voltage higher and they conduct and would fail. Muting transistors are one type where that property is exploited, and special devices are made to operate that bit better under those special conditions than ordinary devices.
Not always (although it does for your amp). Look up common base amplifiers.
Common base - Wikipedia, the free encyclopedia
These are common in very high frequency amplifier designs such as tuner front ends. The signal is applied to the emitter while the base is kept at a fixed voltage.
Swap the emitter and collector over
Hi and thanks a lot indeed. So the condition for a pnp to work is Ve > Vc
Good ! i take a note on this
I see but this is easy ... i can always check the Pw on each active/passive device. Wrong connections is what concerns me most.
Clearly it is not an idiot proof SW
Some sort of basic knowledge is really needed.
Thank you ! message received.
This is difficult.
I am obsessed by simple circuits like this, both for preamps and amps.
I found some interesting chapters in the Bob Cordell's book, where he talks about open loop bandwidth, a very popular issue also around here.
These simpler circuits are also interesting because, if i understand well, they are inherently wide open loop bandwidth.
Instead i have got the feeling that many designers here have a soft spot for extremely complex topologies
I have always the KISS principle in mind.
Of course i should get a basic knowledge first.
Then i will start simulating some commercial amps of which i have the schematics.
Thanks a lot again.
Kind regards, gino
Last edited:
Hi and thanks again. A good point to start.
Then i do not understand why the sw proposes by default the different arrangment ... anyway
I have found something here
That is great indeed ! i wonder if any of his projects has become a commercial product, just out of curiosity
But i have to read more the book ... a great resource indeed.
Thanks again, gino
Hi and thanks and yes i have now the evidence
Just want to say that i am a supporter of idiot-proof things ... where it is almost impossible to make some mistake or better where the chance is ALARP.
More than idiot i should use the term ignorant
Thanks but i guess it is too early for this.
I have to study first. Thanks a lot, gino
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.