Active load on a differential BJT Amplifier

[inass bouhichia]

Hi guys,

So i'm trying to design a differential amplifier using BFU660F transistor, and i'm choosing fot its bias Ic=30mA and Vce=2.5V. and i'm trying to get a 40dB gain from this amplifier and i'm using a 5V tension source. so with these values the best solution is to use an active load. and i wanted to use a current mirror. however on my first simulations it doesn't work, and i don't understand why.

Does the transistor need to be saturated? and generally how are they biased? do i need to add a resistor at the base of the transistor of the differential?

here below the circuit i did.


I hope i was clear enough and thank you in advance.

 

[Harald Kapp]

Your current mirror is incorrct. Base should be connected to collector, not emitter. Also you should use PNP transistors for the current mirror in the upper legs of your circuit. See e.g. this information.

 

[WHONOES]

Also, the chosen current through your transistors is much to high at 30mA each. At that value, the 300Ω resistors will have 9V across them, much more than the 5V supply which of course is impossible. A more suitable value would be 1 - 2mA in each transistor. Harald is correct, the current mirror transistors should be PNP's with the emitters connected to the + supply.
Also, you don't say what you wish to amplify, ac or dc.
Your scheme is probably not the whole solution for the gain you are after.

 

[inass bouhichia]

The 300 resistors was just a dumb idea i was trying, i took them off. right now my circuit is looking something like the picture below.
I don't understand why i should use PNP transistors, i thought for a current mirror, i just needed to use the same transistor.

I got the 30mA Value from the datasheet. it's what the transistor need to get an optimal gain.

I still have some little problems but i think i'll be able to fix them. but feel free to point out anything that looks off in my circuit.

Thank you.

 

[WHONOES]

Because with the NPN arrangement for the current mirror as shown on your 2nd sketch, the top 2 transistors will be turned hard on resulting in the collectors of the lower 2 transistors being pulled up to a few millivolts below the 5V supply. Also, the current source at the bottom doesn't look right either. Are you going to use a single 5V supply or a ± 5V supply?

 

[inass bouhichia]

i'm sorry i still didn't fully understand why i should use PNP, i actually added a resistor between the collector of the lower transistor and the emitter of the current mirror transistor, and now i have the Vce i wanted. is this wrong?

And i'm using a ± 5V. at the bottom it's -5V.

You asked this on your first reply but i forgot to answer it, I want to amplify AC, i'm just starting first by biasing the transistors.

 

[WHONOES]

Simple explanation. Because the current flowing into the collector of one transistor in a current mirror pair will be reflected in the second transistor. With the pair connected as in your circuit, the collector current in either transistor is to all intents infinite therefore, both transistors will be saturated. With the PNP pair this does not happen because the current flowing into a current mirror collector is the result of the collector current from one of the transistors of the differential pair.

 

[WHONOES]

What sort of bandwidth are you after. I see that the device you have chosen is a μwave device.

 

[inass bouhichia]

i need to amplify up to 100MHz. Using a common Emitter i was able to get 30dB gain, buit i wanted to chanllenge myself and do it this way.

Now that i added the AC port i have no gain at all even if i thought i had a good bias point, i'm going to try with PNP transistor now.

Thank you for your help

 

[WHONOES]

If you are getting 30dB gain with a common emitter circuit why don't you just use a second stage to give you the other 10dB that you need?

 

[inass bouhichia]

i could do that, As i said i just want to learn new things and new technics. I havent done this in years so i have gotten pretty bad at it.
Plus a differential amplifier has less distortion than a common emitter so i wanted to see that too.

But to this moment i didn't get any good results.