Mixing signal at high frequency

Thread Starter

captoro

Joined Jun 21, 2009
20
Hello,

I am in the conceptual stage of creating an amplifier that will boost mixed signal input.
I tried using a class A amplifier with a 2n4401, I get good amplification , but agter 60khz, the amplitude decreases, at 130Khz, I get 50% of the desired amplitude.
How can I fix this ? I need the full amplitude up to 20mhz. Any special circuit design ? opamp ?
any help would be appreciated.

ken
 

Papabravo

Joined Feb 24, 2006
12,978
Hello,

I am in the conceptual stage of creating an amplifier that will boost mixed signal input.
I tried using a class A amplifier with a 2n4401, I get good amplification , but agter 60khz, the amplitude decreases, at 130Khz, I get 50% of the desired amplitude.
How can I fix this ? I need the full amplitude up to 20mhz. Any special circuit design ? opamp ?
any help would be appreciated.

ken
How much gain do you require and at what frequency? Transistors are characterized with a gain bandwidth product. There is a transition frequency at which the current gain drops 3 dB
 

Thread Starter

captoro

Joined Jun 21, 2009
20
What sort of amplitude and output impedance you do need?
One idea:

View attachment 199263

Here is my circuit. The 2N4401 works at low frequencies, but not at higher 130khz, I need to get 20Mhz.
I changed the supplied voltage to 18V, so I could get at 12v output. With that input voltage at 130Khx, I get max 4v pk-pk, its worst at higher frequencies.
I think you have the information.

Ken
 

Attachments

Papabravo

Joined Feb 24, 2006
12,978
Hi,

I need this to work for frequencies from 0hz to 20Mhz
I can change the capacitor to a smaller value, but this is restricting me from having all the frequencies I want.

Ken
You need a different transistor and possibly a different circuit. You still have not defined your requirements for gain, only for bandwidth. Also if you want it to work at DC why are you using coupling capacitors? Forgive me for observing that you seem clueless about what you actually need, and you are going about it in the wrong way.
 

Janis59

Joined Aug 21, 2017
1,061
4401 have f(t)=250 MHz, thus for K(u)=100 the problems ought to begin at 2 MHz. Or if the signal form is important then around 100 kHz.
 

danadak

Joined Mar 10, 2018
3,912
Your ci
Hi,

I need this to work for frequencies from 0hz to 20Mhz
I can change the capacitor to a smaller value, but this is restricting me from having all the frequencies I want.

Ken
Your circuit is ac coupled at input and output, so will not work down to 0 Hz.


Regards, Dana.
 

Thread Starter

captoro

Joined Jun 21, 2009
20
You need a different transistor and possibly a different circuit. You still have not defined your requirements for gain, only for bandwidth. Also if you want it to work at DC why are you using coupling capacitors? Forgive me for observing that you seem clueless about what you actually need, and you are going about it in the wrong way.
what you mean by requirements?, yes I am not an expert in amplification of ac Signal.
The signal is generated by and AD9850 which has an output of 1v pk-pk AC. The 2N4401 acts as an ac amplifier to obtain 12v AC pk-pk. But now not working .
WOuld using OPamp (LM741) be able to have a stable output for frequencies up to 20Mhz ?
Maybe not 0Hz, but 5Hz will be the minimum. still very low frequency.

Ken
 

danadak

Joined Mar 10, 2018
3,912
LM741 totally inappropriate.

http://www.ti.com/amplifier-circuit/op-amps/high-speed/products.html

What is G you want ? So pick a part whose GBW = 20 Mhz x G for starters.
Look at the reference design tab as well.

Keep in mind BW very dependent on source Z you are presenting. And layout
of these fast OpAmps very stringent to keep from oscillating. Contributing
factor to this is bypass cap performance. Ceramics key in order to get adequate
ESR at operating freq. Consider ferrite beads on supply pins as well.

https://www.electronicdesign.com/technologies/analog/article/21771753/proper-design-techniques-solve-highspeed-opamp-stability-problems

https://www.analog.com/media/en/technical-documentation/application-notes/AN-257.pdf

http://www.tij.co.jp/jp/lit/an/sloa046/sloa046.pdf

1581972944686.png

https://www.mpdigest.com/2016/05/23/esr-losses-in-ceramic-capacitors/



Regards, Dana.
 
Last edited:

Papabravo

Joined Feb 24, 2006
12,978
what you mean by requirements?, yes I am not an expert in amplification of ac Signal.
The signal is generated by and AD9850 which has an output of 1v pk-pk AC. The 2N4401 acts as an ac amplifier to obtain 12v AC pk-pk. But now not working .
WOuld using OPamp (LM741) be able to have a stable output for frequencies up to 20Mhz ?
Maybe not 0Hz, but 5Hz will be the minimum. still very low frequency.

Ken
OK. Requirements are essential to how circuits are designed. You want a gain of 12 from a single transistor stage using the 2N4401. First problem is that you cannot get 12V Pk-Pk from a single +12VDC power supply. I would shoot for a bipolar supply of at least ±15 Volts. Second problem is to determine from the datasheet or from an experiment if the 2N4401 has a gain of 12 from DC to 20 MHz. with no unusual behaviors inbetween. You're obviously going to need a new circuit to ditch the coupling capacitors. Looks kike you got some work cut out for you. BTW -- off the top of my head I have no idea if this will work -- I'm away from home and my resources.
 

DickCappels

Joined Aug 21, 2008
6,199
It looks like you are driving the gate of that MOSFET through the 2.5 mH inductor, but the ooutput impedance the 2N4401 circuit is 2.2k + whatever capacitance you have there. I second the opinion that you need a new circuit.
 

Thread Starter

captoro

Joined Jun 21, 2009
20
It looks like you are driving the gate of that MOSFET through the 2.5 mH inductor, but the ooutput impedance the 2N4401 circuit is 2.2k + whatever capacitance you have there. I second the opinion that you need a new circuit.
Yes I agree a new circuit is necessary.
I saw a few links that was given to me. I'll see if I get anywhere with that ?
Any suggestions ?

Ken
 

DickCappels

Joined Aug 21, 2008
6,199
This is not going to be an easy project by any means. It looks like you are trying to drive the MOSFET as a linear amplifier. You probably can't get enough drive out of the 2N4401 to obtain much. You might try adding an analog buffer between the 2N4401 and the MOSFET gate.

Is this a radio transmitter, if not what it it's purpose. What is the signal source driving the 2N4401 (which, by the way is a very respectable high frequency transistor)?
 
Top