The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Imperative Task: Independently Amplify Voltage or Current at 2GHz
https://designers-guide.org/forum/YaBB.pl?num=1298586239

Message started by jianke on Feb 24^th, 2011, 2:23pm

Title: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by jianke on Feb 24^th, 2011, 2:23pm

Hi all,

Recently I met an interesting question for RF amplifier design. In the design, at a frequency of about 2 GHz (e.g. 2.4 GHz), we want to design an amplifier which can independently amplify "current" or independently amplify "voltage".

At low frequency, we know that amplification circuits can be classified into three categories: "power" amplification circuit, "current" amplification circuit, and "voltage" amplification circuit. For example, a BJT Common-Emitter (CE) amplifier can amplify current and voltage simultaneously. A Common-Collector (CC) amplifier mainly amplify current (i.e. amplifier voltage gain is about 1). A Common-Base (CB) amplifier mainly amplify voltage (i.e. amplifier output current almost has no change). But at RF frequency, typically at about 2 GHz, I cannot find such amplification circuits which can independently (or mainly) amplify the "current" or "voltage".

At about 2 GHz, are there such current amplification circuits which can mainly amplify the current (with small small voltage gain), or voltage amplification circuits which can mainly amplify the voltage (with small small current gain) ?

Doesn't anyone have an idea or any comments? It is an imperative duty! Thanks!

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by rfidea on Feb 25^th, 2011, 8:57am

As long as the fT of the technology you have chosen is high enought the stages, CE, CG and CE, you suggested will work. There are a lot of available bipolar technologies with fT:s higher than 25GHz that can do the job at 2GHz.

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by aaron_do on Feb 25^th, 2011, 4:38pm

I agree with rfidea. The technologies now are so fast that a lot of your low frequency designs can work into the low GHz range. It will really depend on what kind of power consumption and NF you want to achieve.

cheers,
Aaron

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by loose-electron on Feb 27^th, 2011, 2:47pm

As you get to RF frequencies, the technique of defining the amplifier as being a driver into a defined impedance becomes prevalent.

With a defined load, it sort of becomes irrelevant.

Also, I agree with what others have said, yesterdays RF techniques now can be done using op-amps...

Consider the AM broadcast band - 500Khz to 1.5MHz (give or take) - that can all be dealt with using feedback amplifiers these days.

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by jianke on Mar 1^st, 2011, 10:26am

Thank you so much for your reply. Those days, I try to construct a BJT amplifier base on Common-Collector (CC) and Common-Base (CB), respectively. The BJT chosen for this design is AT41411, which has a cut-off frequency (ft) of 7 GHz. But, the ADS simulation results are strange. Take the CC amplifier for example, the emitter current (IE) is bigger than the base current (IB), but the output current (I_out) becomes lower than the input current (I_in). According to the characteristics of a BJT CC amplifier, the output current should be amplified. I don’t know why. (Of course, what is correct is that the output voltage becomes a litter lower than the input voltage, because the amplifier gain is between 0 and 1)

The simulation circuit and results are attached for your review. Looking forward for your reply. Best regards!

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by rfidea on Mar 1^st, 2011, 11:22am

You are trying to drive 1.8G in an 7GHz maxFt device in emitter follower topology. That is a very high frequency for that device.

I do not agree that an emitter follower is a current gain device. It is more of a lousy unity gain voltage buffer with bad linearity, large current consumption and a tendency to be unstable. But otherwise it is fine.

By the way, the DC operating point of your circuit will have a large dependency of transistor beta.

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by jianke on Mar 1^st, 2011, 12:40pm

Hi, rfidea. Thank you for your reply. The idea of using CC and CB to as a current and gain device, respectively, is from the low-frequency analogy circuit. To be honest, I am not sure whether it can work or not at RF. The problem of a CE (or CG) amplifier is that the current and voltage will be amplified simultaneously, and it is not what I want. For a CE (or CG) amplifier, do you know whether it can mainly amplify current or voltage, while has low gain for the other one.

Could you suggest a BJT device (and a FET device) ? If I want my amplification circuit to work at 2GHz, whether the device’s maxFt should be higher than 25GHz as you suggested first ? I am sorry I don’t have much experience in RF circuit design. Cloud you give me a hand ? Thanks and Best Regards!

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by loose-electron on Mar 1^st, 2011, 5:46pm

rfidea wrote on Mar 1^st, 2011, 11:22am:

Well stated, I fully agree.

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by loose-electron on Mar 1^st, 2011, 5:52pm

jianke wrote on Mar 1^st, 2011, 12:40pm:

A suggestion -

Please tell us what you are trying to do, and perhaps we can offer suggestions.

Not the circuit you are trying to create, but rather what you are trying to do. (example, you have a 2.5GHz signal with XX amplitude, coming from a 50 ohm cable and you need to amplify it to XX amplitude, and feed it into another cable at 50 ohms, you need a linearity of IP3 = XXdb, and your Noise Figure has to be XX)

At these frequencies, you have to know what you are doing and parasitics and many other factors will effect what you are doing.

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by jianke on Mar 2^nd, 2011, 1:40pm

Hi, Jerry. Thanks for your suggestion. Now, it is hard to say the final objective or some other performance parameters, such as IP3 or NF. But I think they should bot be very bad. Do you have some idea about Op-Amps ? I wonder whether an Op-Amp (e.g. VOA, OTA, CFA, CC) can do this job or not. That is, independently (or mainly) amplify voltage or current at 2GHz.

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by loose-electron on Mar 2^nd, 2011, 4:04pm

At the frequency mentioned op-amps are generally not used.

If you can't define what you need to do, its not going to be easy to help.

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by rfidea on Mar 6^th, 2011, 1:08am

I have looked at your voltage and current plots. The output signal voltage is somewhat lower than the input signal voltage. Probably due to limited frequency response and limited beta. Both input and output current is set by Ohm's law, V=R*I. The input impedance in your circuit is probably capacitive since you are operating the transitors close to maxFt. The output impedance is you load, 50ohm. Apparently those impedances is giving you the currents you are simulating. I do not understand your statement that an emitter follower is a current gain device, can you please describe why.

Since I personally work with RFIC design I do not have any knowledge of a suitable discrete transitor for you. Since you have not described you real spec or problem it is also hard for someone else to help you on that topic.

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by rfidea on Apr 25^th, 2011, 12:54pm

Some clarification, because of a direct question from jianke, about the suggestions of using technologies with maxFt > 25GHz for this problem at 2GHz.

I did not mean that there is a strict requirement that the technology used must be higher than 25GHz. I only mean what I said, that there is a lot of technologies available with maxFt > 25GHz that can be used.

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by jianke on Apr 27^th, 2011, 8:21am

Thanks so much again for your reply. As you suggested, "As long as the fT of the technology you have chosen is high enought the stages, CE, CG and CE, you suggested will work."

I want to know why, with high-enough MaxfT, CC or CB amplification circuit can work at 2GHz, as I suggest current amplification or voltage amplification, respectively.
Because of "Transistor Interelectrode Capacitance" is very low with so high fT, or something else ?

Title: Re: Imperative Task: Independently Amplify Voltage or Current at 2GHz
Post by rfidea on Apr 27^th, 2011, 1:17pm

Ft is the frequency where the transitor has unity gain. If that frequency is higher than your operating frequency it is possiblilty to achive gain in your circuit.