2
\$\begingroup\$

I want to design an amplifier that can take an 8MHz signal (even square) with an amplitude of only a few volts (1-3V) and amplify it up to 160V but I only need it to deliver about 60mA. I also want to avoid using a high voltage input voltage.

Does anyone have any ideas where to start? I was thinking of a small class A amplifier in conjunction with a transformer. Would the use of a transformer be impossible due to the high frequency? Or should I just use a boost converter and then use it to supply my class A/class AB amplifier?

Any ideas are welcomed.

\$\endgroup\$
15
  • 3
    \$\begingroup\$ Have you computed the slew-rate this is going to require? (You write "even square" and this could imply a rather fast edge.) \$\endgroup\$
    – jonk
    Jul 10 '18 at 20:05
  • \$\begingroup\$ It's not going to be a perfect square as that is impossible (infinite slew-rate). But for a triangular wave at 8MHz I would need about 4800V/us, Am I right? (for a 150V output from zero) \$\endgroup\$ Jul 10 '18 at 20:19
  • \$\begingroup\$ Could I maybe use something like this? I don't mind the floating output. ti.com/lit/an/snva516/snva516.pdf \$\endgroup\$ Jul 10 '18 at 20:21
  • \$\begingroup\$ The very first sentence in that datasheet says it has +/- 40 V maximum power supply, so that won't get you to 160 V. \$\endgroup\$
    – The Photon
    Jul 10 '18 at 20:23
  • 2
    \$\begingroup\$ If you need a square wave of a fixed amplitude, forget about amplifiers, switch a 160V supply instead. Fast enough switching is challenging enough. \$\endgroup\$ Jul 10 '18 at 20:57
1
\$\begingroup\$

As an overall approach, I would start at the high level solution architecture first, then progress to the specifics and exact components.

In this case, you need up to a 160 x amplification of a 1v p-p HF (high frequency) RF signal. 8 MHz is radio frequency. 160 v p-p RF brings with it some other challenges around RFI / EMI that need to be considered (you didn’t explain how this 160V RF signal is to be used).

160 x amplification is 40 dB of gain. You will need at least a couple of amplification stages. This sounds similar to “linear amplifiers”, where one would begin with a Class A single transistor amp stage, followed by a Class AB push-pull stage. Then there’s the matter of matching the input and output impedances, properly biasing each stage and ensuring no RF gets back up into your power supply with proper filtering.

At the low current levels you need, it shouldn’t require highly expensive power transistors as you’re only talking 10 watts output. With efficiency in the 60%+ range you can probably do fine with 20 watt finals in the PA.

That’s how I’m framing out a potential solution based on the limited requirements available at this stage.

PCB layout and shielding will also need some consideration, along with your power supply.

\$\endgroup\$

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.