HF Amplifier Help!! / General Chatting / Forums

Forums

4hv.org :: Forums :: General Chatting

« Previous topic | Next topic »

HF Amplifier Help!!

Move Thread

LAN_403

Hazmatt_(The Underdog)

Tue Jul 22 2008, 01:09AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

I was wondering if any of you guys could recommend a Book on HF amplifier design and construction. I don't usually build amplifiers and I usually can't get very far with their construction because I don't have much experience in this area.

What I want to do is to get some power behind my signal generator, rather then buying one of those exotic lab signal generators, I simply don't have the money. So I'm sorta stuck between Audio and CB stuff since my desired range is 50KHz to 2MHz. If anyone has some suggested reading for this I'd be very appreciative.

I have built a couple of variants of a push-pull mos amp now. It wasn't amplifying efficiently because the phase splitter was not balanced, so I'm rebuilding the amp phase splitter. I just characterized it and it works okay, it has a resonance at 80KHz and works up to 10MHz. After I get that situated I'm going to go back to the power stage and try to get that set back up with another set of IRF840's and see if I can't get it amplifying efficiently again.

Matt

Dave47

Tue Jul 22 2008, 02:37AM

Registered Member #84 Joined: Thu Feb 09 2006, 01:06PM
Location: Dallas, TX
Posts: 47

How much power do you need? Something like this would probably at least be in the ballpark of what you are looking for as a signal generator amplifier with a couple of watts of power...

If you are talking 100W with the IRF840s, then it might at least make a decent preamp for an even higher powered output stage. You might just get away with two even higher power transistors enclosed in the feedback loop.

David

Hazmatt_(The Underdog)

Tue Jul 22 2008, 04:30AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

I just finished up re-characterizing the amp and It's working pretty good from 200KHz to 2MHz. It has some problems from 91K-250K because of the input transformer, so I'm going to try to replace that with a phase splitter/driver. I don't need much for input drive to get output power, and the amp is idling at ~800mA. I'm setting the input bias voltage to 3.85Vdc to keep the amp cooler then it was at 4.1V. I'm not sure what the output power is just yet because it varies quite a bit over the spectrum. I've got a current probe on it right now and I'm getting 160mA RMS and 1.4V RMS, so it looks like I'm getting 0.224W RMS at 540KHz. I suppose that means I need to lower my bias and increase my input drive because the main supply is burning up 16W to power the amp, or 1.4% efficiency, ouch!

Here's some pics:
The big sine is 5V/div into a 50R load

GeordieBoy

Tue Jul 22 2008, 11:40AM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

If you want a workhorse RF power amplifier for testing stuff over a reasonably large frequency range then go for a Class A linear amplifier. Depending on frequency range and power requirements, one or more IRF510 devices should do the job.

Efficiency of Class A is poor but it's tolerance of bad loads is superb. It already dissipates about 85% of it's input power as heat so even with complete reflection of all output power it's unlikely to break into a sweat. That's why Class-A RF power amps are used in RFI and EMC compliance testing where the load is "variable."

The more efficient classes like Class-C, Class-D, Class-E, D-E etc are all great for delivering kW's of power at one frequency into 50 ohms with excellent efficiency. Outside the design envelope they fall apart pretty quick though.

A typical Class E amplifier might be 95% efficient. So a 95W amplifier is only designed to dissipate 5W of power into its meagre heatsink. Under mismatched load conditions this can rise to 100W. That's an increase in dissipation of 20 times, and spells certain heat-death if the amp is not protected.

-Richie,

Dave47

Tue Jul 22 2008, 12:03PM

Registered Member #84 Joined: Thu Feb 09 2006, 01:06PM
Location: Dallas, TX
Posts: 47

Cool! You got it to work. I do understand the desire to make the amplifier with parts on hand, even though there is probably a suitable component already on the market, though.

I'll just set the bar for your (eventual) performance with the amplifier I posted

.

That amplifier has a quiescent current of 13mA per driver. So, 26mA for both if you go for differential output drive.

Figure 38 has 20V output into 25 ohms at 4MHz (Gain of 5) single ended. Differential, you could get 40V into 50 ohms with a giant heatsink on the the IC. It has a thermal shutdown circuit, so it probably won't kill itself if you abuse it.

Figure 20 shows a gain of 100 with flat AC performance to 10MHz. The plot shows 100KHz at the low end, but DC coupled it would be flat to DC.

Digikey has it for $7.32 for a single. Add some supply bypass caps and some resistors and you are done.

David

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob