hello to every one,
What does a transistor Ft means when applied to an audio
power amplifier specially at the outputs? Does this thing do something about distortion, sonic quality, whatever?
this might be crazy little questio but it bother a lot to me..
This is a number that shows how the transistor gain falls of with frequency. The Ft = transit frequency is frequency where the current gain = 1, IIRC.
It doesn't directly work on distortion as such. But you will see that if you have an output transistor with low Ft working at high frequency, you have to put in a lot of base current to get some output current. The driver stage has to deliver more current and therefore will distort more, so the whole amp distorts more. There may be other effects I am not aware of like phase shifts in the output transistor having an effect on stability.
Jan Didden
It doesn't directly work on distortion as such. But you will see that if you have an output transistor with low Ft working at high frequency, you have to put in a lot of base current to get some output current. The driver stage has to deliver more current and therefore will distort more, so the whole amp distorts more. There may be other effects I am not aware of like phase shifts in the output transistor having an effect on stability.
Jan Didden
ft
😉 😀
this time its a bit clear for me, how Ft's affect power amplifiers.
so is it rather preferable to use output devices with high Ft's?
so as to
prevent so much phase shifts or other kind of distortions ?
😉 😀
this time its a bit clear for me, how Ft's affect power amplifiers.
so is it rather preferable to use output devices with high Ft's?
so as to
prevent so much phase shifts or other kind of distortions ?
Yes, but...
The Ft depends much on the manufacturing process. If you use a process that produces high Ft, it is difficult to have at the same time high power, especially at higher voltages. Google "safe operating area" (SOA), and you will see that a power transistor of, say 100W, CANNOT dissipate this 100W at high voltage levels, otherwise it self-destructs. 5A at 20V will be OK, but 1A at 100V not (assuming a 100V transistor). The thing is that with (very) high Ft, you also get very small hi-voltage SOA. That is why it is difficult to find high power, high voltage, high Ft transistors, and the ones that exist are expensive.
Jan Didden
The Ft depends much on the manufacturing process. If you use a process that produces high Ft, it is difficult to have at the same time high power, especially at higher voltages. Google "safe operating area" (SOA), and you will see that a power transistor of, say 100W, CANNOT dissipate this 100W at high voltage levels, otherwise it self-destructs. 5A at 20V will be OK, but 1A at 100V not (assuming a 100V transistor). The thing is that with (very) high Ft, you also get very small hi-voltage SOA. That is why it is difficult to find high power, high voltage, high Ft transistors, and the ones that exist are expensive.
Jan Didden
Yes transistors with really high Ft's as much as 30 to 50Mhz which at the same time high voltage, high power dissipations really are very
expensive. Sanken's 2SC2922 and 2SA1216 are some examples.
From my experience, and as far as my crave for deep punchy bass
is concerned or even anywhere in the audio spectrum is concerned,
I do love this devices but I do hate their prices
Another thing, transistor output capacitance (Cop) does this thing
goes with Ft?
🙂
HIENRICH THIELE
expensive. Sanken's 2SC2922 and 2SA1216 are some examples.
From my experience, and as far as my crave for deep punchy bass
is concerned or even anywhere in the audio spectrum is concerned,
I do love this devices but I do hate their prices
Another thing, transistor output capacitance (Cop) does this thing
goes with Ft?
🙂
HIENRICH THIELE
Well, for deep crunchy bass your fast transistors are a waste of money, assuming you use a sub amp. There are very rugged transistors in the <5Mhz Ft region that are perfect for bandwidths up to 1kHz or so.
Jan Didden
Jan Didden
Perhaps not always true.
High Ft transistors at output stages can sometimes pose a serious problem in making the amplifier work in class AB. The Quiescent current may suddenly shoot up making it impossible to control to desired value. Under such circumstances it may be necessary to use transistors of lower Ft.😕
High Ft transistors at output stages can sometimes pose a serious problem in making the amplifier work in class AB. The Quiescent current may suddenly shoot up making it impossible to control to desired value. Under such circumstances it may be necessary to use transistors of lower Ft.😕
Obviously NOT ...sorry but this is nonsense IT is not the transistor's problem The problem originates form the one that designed a bad circuit behind them😡
This is a 10 years old Thread...!!
source : BJT quirks : Bipolar Junction Transistors - Electronics Textbook
bigger transistor mostly have higher Cob than the smaller one, because it has larger area.
not sure its(Cob) correlation to fT. but it might, cause higher Cob(Miller effect) mean lesser gain at high freq', as to fT, lower fT has lower gain available at high freq'
here's my two cents.
source : BJT quirks : Bipolar Junction Transistors - Electronics Textbook
bigger transistor mostly have higher Cob than the smaller one, because it has larger area.
not sure its(Cob) correlation to fT. but it might, cause higher Cob(Miller effect) mean lesser gain at high freq', as to fT, lower fT has lower gain available at high freq'
Last edited:
Yes transistors with really high Ft's as much as 30 to 50Mhz which at the same time high voltage, high power dissipations really are very
expensive. Sanken's 2SC2922 and 2SA1216 are some examples.
From my experience, and as far as my crave for deep punchy bass
is concerned or even anywhere in the audio spectrum is concerned,
I do love this devices but I do hate their prices
Another thing, transistor output capacitance (Cop) does this thing
goes with Ft?
🙂
HIENRICH THIELE
2SC5200/2SA1943 are not expensive.
A 5MHz Ft bipolar with a DC Hfe of 100 would have its gain drop 3dB by about 60kHz, so 1kHz is being a bit mean 🙁
Since the output tranies are usually the lowest ft in an amp they will create the first pole for the entire amp ( the freq that the open loop gain is down 3db ) until compensation is added. This freq in most audio amps is below 20khz. Sometimes around 1 kHz , even with ft's in the MHz. Since your open loop gain is dropping with f , your distortion is increasing. The low ft also introduces phase shift which causes instability, so you have to compensate by adding a pole below the one ft causes, which further increases HF distortion. So higher ft output tranies increase the open loop bandwidth which helps stability and HF distortion. This is all however CIRCUIT DEPENDENT. This is the general case for most feedback amps.
Not sure
Though I am not a designer myself, My experience relates to Not a complementary symmetry circuit but a transformer coupled output stage driven by an input transformer for phase split and pre driver transistors. The power transistor emitters were loaded by output transformer primary. I had added base resistances. The overall freq response is -3db at 50 hz to 15khz with a fair hf distortion. Overall feedback used is only 10 db. It is the pre drivers that appear to lead to instability. Unfortunately very little material is available on this type of construction though it has many great advantages. Any suggestion, welcome!
Though I am not a designer myself, My experience relates to Not a complementary symmetry circuit but a transformer coupled output stage driven by an input transformer for phase split and pre driver transistors. The power transistor emitters were loaded by output transformer primary. I had added base resistances. The overall freq response is -3db at 50 hz to 15khz with a fair hf distortion. Overall feedback used is only 10 db. It is the pre drivers that appear to lead to instability. Unfortunately very little material is available on this type of construction though it has many great advantages. Any suggestion, welcome!
- Status
- Not open for further replies.