Post Go back to editing

Hi,

I'm just dipping my toes into an HF design and could do with some gain, the ADL5530 got my eye,

A quick calc (please check, I'm slightly out of my comfort zone):

• The ADL5530 is specified to have a 1dB compression point of 21.7dBm = 148mW.
• P = V*V/R so in a 50R system this yields V = 2.72V rms.

Hmm, now here's how I know I've gone wrong,

• 2.72V rms for a sinusoid equates to 3.85v pk and so 7.7v pk-pk, how with just a 5v supply?
• Working backwards, if it could swing to the rails, thats only 5Vp-p, 2.5Vpk, 1.77Vrms, 62.5mW so 17.9dBm

Any pointers in the right direction would be most welcome.

Project background:

I'm designing a test pulse generator for 15ns wide, fast rise/fall pulses with variable amplitude from mV to +/-2.8v pk

The 2.8v pk pulse would I thought need a gain stage capable of 2Vrms hence 80mW rms hence 19dBm rms output.

I've +/-300mV to +/-500mV pulses from an ADCMP606 as a starting signal, a digital step attenuator for level control.

Allowing for DSA insertion, RF switch, connectors, some coax and PCB losses I need about 22dB of gain.

I was thinking of an ADL5530 before and another after the DSA.

I rulled out a conventional opamp as I'd be looking for 1500v/us and large signal GBP > 2.3GHz

• Think of the RF output as the collector of a transistor. This is connected to the 5 V supply through a bias inductor. So the output has a dc bias level of 5V. However from an ac perspective, is an open between the supply voltage and the RF output. That allows the collector of the transistor (i.e. the RF Output) to go where ever it needs to go to deliver (ac) output power.

If the connection from the RF output to the 5V power supply was made using a resistor, then what you are saying would be correct, that is, the RF output would be limited to voltage excursions betewen 0V and 5V.