I've been working on learning transistor 101s, for a 900MHz transmitter.

I'm to the point, I can work the basic algebra, and intuitively understand changing the passive components in the transistor circuit to boost an RF signal.

I came across a wide bandwidth RF amplifer from NXP, the BGA2817.

Here is the BGA2817 Datasheet.

At a cost of $0.08/USD at volume, what is the downsides to using BGA2817 as an amplifier, versus going through all the development and trial and error of building up some transistor stages and amplifying your signal for a transmitter?

The one thing that sticks out to me is the 20mA supply current. Not really a deal breaker for a transmitter, but clearly a deal breaker for a receiver.

Are people moving over to this type of IC for RF amplification?

Why does this thing draw 20mA in the first place, when you get down to how this IC is built. I have no idea.

It doesn't get any easier to implement one of these, all the voltage step-up, impedance matching, and whatnot is just baked into the IC.

  • \$\begingroup\$ I strongly suspect that if you managed to build an amplifier with comparable performance then it too would draw a fair amount if current. \$\endgroup\$ – JRE Aug 9 '17 at 19:22
  • \$\begingroup\$ It would be a lot of trial and error (read: whole lotta board turns) to build something comparable at my EE skill level... it seems almost too good to be true for what you are getting in this package at that price point. Is there some gotcha here? \$\endgroup\$ – Leroy105 Aug 9 '17 at 19:37
  • \$\begingroup\$ The advantage of this kind of IC is that it just does a lot for you - matching, low noise biasing, etc. As you get to higher and higher frequency, doing it with discretes becomes more and more impractical or even impossible \$\endgroup\$ – Joren Vaes Aug 10 '17 at 6:29

For small low power amplifiers, this is by far the easiest way to do the job.

The 20mA bias current is 'paying for' the P1dB output power of 6dBm and the distortion performance output TOIP of 18dBm. If you don't need these values, then you can save power by going through the catalogues (NXP aren't the only people making these things) and tabulating power consumption versus output specs. You might find examples with better ratios, you will find examples with lower power.

Have a look at minicircuits, avago, rfmicrodevices, for monlithic broadband amplifiers.

  • \$\begingroup\$ Neil, appreciate the input! This NXP line seems to just be the cheapest game in town. In this case, that extra DBM puts the 900MHz TX IC right at the FCC Reg Limit of 30dBm for a spectrum device hopping device. I'm coming at the EE end of design, starting in 2017 -- you read so much about building your own amplifier using transistors, and then you look at these ICs and have to scratch your head if you're missing something.... We source passives for fractions of pennies in China, but adding in the PCBA costs, I seriously doubt we could build this amplifier cheaper! \$\endgroup\$ – Leroy105 Aug 9 '17 at 21:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.