Choosing a transistor for wideband amplifier

Status
Not open for further replies.

wallander

Newbie level 2
Joined
Feb 21, 2015
Messages
2
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Visit site
Activity points
18
Hi,

i'm relatively new in RF design and i have to deal with design of wideband amplifier. It should work from 5kHz up to 1.5 GHz. First i focused on choose of proper transistor. I found pHEMT ATF531P8, which have good parameters for my design (P1dB, NF and OIP3). But, in datasheet stays, that it's designed for frequency range 50MHZ to 6GHz. I'm pretty confused, will it work under 50 MHz ?

Thanks for any advice
 

It will work but it will not able to satistfy the specifications ( especially NF because FETs have higher Flicker Noise compare to Bipolars) as above 50MHz.If your amplifier will work 5kHz, I don't recommend you to use FETs because their NF problems under 50-100MHz.Why don't you consider Biploar ? NXP,Avago,Infineon,CEL and Renesas have very wide biploar portfolio.Think about that..
 

That's a question. Transistor will be used as a new final stage amplifier of a signal generator. So, NF is not critical parameter. P1dB and OIP3 are. Original stage had 35dBm OIP3 and 27.5dBm P1dB. NF was about 5dB.

But it's true, that in original stage, which was designed in 1995, was bipolar transistor. I tried to find proper bipolar transistor, but i haven't found any with those OIP3 and P1dB values.
 


At which frequency these characteristics have been obtained ? It's not possible from 5kHz-1.5Ghz..There should be a reference frequency.
Have you ever looked at NXP Wideband Bipolars ??
 

At which frequency these characteristics have been obtained ? It's not possible from 5kHz-1.5Ghz..There should be a reference frequency.
Have you ever looked at NXP Wideband Bipolars ??

Designing a wideband amplifier is not easy mainly at the lower-frequency limit.
Years ago I used a RF 2326 integrated circuit made for CaBLE tv APPLICATIONS, quite successfully.
This IC required a high inductance DC-bias network to cover 3 kHz to 1.5 GHz bandwidth. For DC bias at the output, I used a series of low-inductance chokes (10 uH in series with 1 mH and 1 H) chokes, to achieve >13 dB gain over the bandwidth, and the noise figure was < 3 dB.
Designing such amplifier with a BJT or MESFET transistor alone is difficult due to wide-band matching and low-frequency noise problems.
Use what is offered on the market, it makes the design much easier and working well.
 

Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…