Few watt RF amplifiers

You might search around for CATV amplifier modules. Especially the backhaul types. There may be some good deals on surplus modules.

What's the voltage and load impedance? There are opamps that might work.

Can you resonate the load?

--

John Larkin         Highland Technology, Inc 
picosecond timing   laser drivers and controllers 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

I searched ebay using 'rf power amplifier hf vhf uhf' without the quotes.

eBay has two models that give 1.5W, but the power drops off either side of the optimum frequency. These have fairly high gain - 0dBm max input for around 32dBm output. Here's the two results. Funny thing, the first one gives a different price every time I click on the link.

RF 20--500MHZ 1.5W HF VHF UHF Power Amplifier For Ham Radio US $38.81 to US $40.94

1MHz--500MHZ 1.5W HF FM VHF UHF RF Power Amplifier for ham radio + Heatsink US $53.29

If you really need more power, maybe two or more could be combined using a Mini-Circuits power splitter:

Most of these are 0.5W to 1W max, which would be fine for the input side. I could not find any that would handle 3W or more for the output. It might take finding an Amidon core and winding your own output combiner. Joerg would probably have a lot more info.

Mini-circuits has some app notes on combiners and spltters:

I don't know how well the amplifiers output phase would match. Presumably a phase error of 10 degrees or less would have a small effect. The amplifiers might be able to hold that over the part of the frequency range that you need.

You certainly have all the equipment needed to do it yourself!

It's an octave-band acousto-optic deflector (45-90 MHz). Pretty much 50 ohms, apart from a 2:1 VSWR--most of the power actually goes into the acoustic wave.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

Thanks. What I don't have is the time, and the customer is very nervous about schedule risk. Since it's a demo, and this amp is a small part of a big system that I'm building by myself, something from an onshore vendor with a good track record would be very comforting.

I've gotten in touch with an outfit previously unknown to me, RF Bay Inc., who claim to sell a nice-looking 4W unit for $470 in onesies (p/n MPA-40-40), which is more in the price range I was expecting.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

In case that doesn't pan out, here's another offshore eBay item that might be interesting as a backup. You could ask them to ship via DHL or air freight and get it next week.

New 2MHz - 80MHz 5W RF Wideband Amplifiers / Frequency amplifier power amplifier US $63.55

On a sunny day (Tue, 06 Jan 2015 00:15:03 -0500) it happened Phil Hobbs wrote in :

Well, that is real wallpaper.

Long time ago I did some spice slimulations with NPN/PNP complementary output stages. With the right transistors you can get to 100 MHz. What sort of voltage and current do you need? Else small wideband transformers can be used. Push pull on transformer with center tap should work too.

4 W is nothing...

Phil, I left you a voicemail. I have a couple of digital Neos Technology Drivers that are FM Capable with a internal DDS for fast sweep as well as analog AM.

Amps used for this task are transformer coupled, push pull stages. Small units use CATV bricks, ie stuff from RFPARTS.com and other vendors. Large units use broadband RF stages right out of Helge Granberg's Motorola App Notes.

Steve

There is also

They have a FM band amp for 375$ that may be of use.

Steve

That's true, particularly when the application is narrowband, as in a ham transmitter, and there isn't too much else going on in the system.

For this job I need reasonable gain and power output flatness (say +-1 dB) over an octave band, to drive acousto-optic cells. The gain needs to be stable, because you get best deflection efficiency right near the point where the AOD roasts.

I'm already doing the theory, the optics, the mechanics, the RF section, the nulling amplitude and phase digitizer hardware and firmware, and software for navigation, scanning, control, deconvolution, and display. For this I have some PC board layout help, but am otherwise a one-man band.

We got the contract in place just before Christmas, and it all has to ship by April and be fully acceptance-tested by May. Plus I have a

2-week hiatus in February to testify in a trade-secret trial, as well as a couple of other smaller projects.

So I'm a busy guy at the moment, and the client would rather write cheques than suffer delay. Five bills isn't outrageous for a 4W instrumentation amp, especially with 200:1 bandwidth, but I was pretty surprised by Mini Circuits' jump from $220 for 1W (ZX60-100VHX+) to $535 for 2W (ZHL-1-2W-S+) to $1k for 5W (ZHL-5W-1X).

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

On a sunny day (Tue, 06 Jan 2015 09:09:43 -0500) it happened Phil Hobbs wrote in :

I was talking about wideband DC - 100 MHz amps.

Well the complementary pair itself has a gain of about 1 :-) How you drive it is what limits things, but quite standard stuff.

High voltage versions of this, and totem pole versions using only NPN, were common in CRT rgb drivers to >>100 MHz (pixel frequency VEGA).

Look up the LM2406 and LM2427 datasheets for example, although those do not go so high, the 27 can dissipate up to 10 W or so. I tried making my own with discrete transistors, that is why the slimulation. If you needed only a high voltage / low current that could be a solution, there are chips for this that go much higher than those 80 MHz.

That is why I asked about voltage / current.

High power RF transistors usually are low voltage types (unless you use MOSFETs) and driving a transformer in push pull mode with 2 is common place. Above a few MHz you probably can use air coupled coils, but there are plenty ferrites for that frequency. I have a 30 MHz wide band 100W push pull transistor linear upstairs.

Have you checked with ENI in Rochester, NY? Never used their stuff personally but I've seen many of their amplifiers built into various mass spectrometers and other analytical instruments and never heard any complaints. Web site is

no online pricing but worth a look and a call or email.

Regards, Carl Ijames carl.ijames aat deletethis veriz> On a sunny day (Tue, 06 Jan 2015 00:15:03 -0500) it happened Phil Hobbs

That's true, particularly when the application is narrowband, as in a ham transmitter, and there isn't too much else going on in the system.

For this job I need reasonable gain and power output flatness (say +-1 dB) over an octave band, to drive acousto-optic cells. The gain needs to be stable, because you get best deflection efficiency right near the point where the AOD roasts.

I'm already doing the theory, the optics, the mechanics, the RF section, the nulling amplitude and phase digitizer hardware and firmware, and software for navigation, scanning, control, deconvolution, and display. For this I have some PC board layout help, but am otherwise a one-man band.

We got the contract in place just before Christmas, and it all has to ship by April and be fully acceptance-tested by May. Plus I have a

2-week hiatus in February to testify in a trade-secret trial, as well as a couple of other smaller projects.

So I'm a busy guy at the moment, and the client would rather write cheques than suffer delay. Five bills isn't outrageous for a 4W instrumentation amp, especially with 200:1 bandwidth, but I was pretty surprised by Mini Circuits' jump from $220 for 1W (ZX60-100VHX+) to $535 for 2W (ZHL-1-2W-S+) to $1k for 5W (ZHL-5W-1X).

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

Quote: "> With the right transistors you can get to 100 MHz."

Change that to "With the right transistors you can go up to 600 Mhz"

I have bricks that cover .5 Mhz to 550 Mhz for CATV, with a linearity that is unbelievable.

Steve

Thanks, Steve, you're a brick yourself. ;) I'm going to roll my own DDS board, because it has to have the amplitude and phase digitizers on it anyway, so everything needs to be phase coherent. I can smooth out the amplifier and LPF variations by dorking the DDS amplitude.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

Hi Phil, lurking... I was hoping you would decide to roll your own. And I could hear /tell of a good RF power transistor. I've got a(n) Rb discharge lamp with a 60-80 MHz few watt oscillator. It uses a 2N3375 (an old part that we get from advanced seimconductor.) I expect it to go away some day, and I'll have to find a replacement.

George H.

George, call up the guys at RFParts, and ask them what will be stable for a few years. Phone is (800) 737-2787

Steve

Ahh a second source! Thanks Steve. (octoparts doesn't know everything.)

George H.

The RF Bay Inc MPA-40-40 has impressive documentation, but it doesn't talk about one spec that is very important for lab work. There is no data on load VWSR, which could range from a short to open and everything in between. The reflected power could blow the output MOSFETS. It might be a good idea to talk to them about this.

In case it doesn't handle the load, there is another option. Thanks to Steve for the information on Motorola's H.O. Granberg. He wrote numerous articles on HF and VHF wideband amplifiers which are very informative.

One article is AR313, a 10MHz to 150MHz 300 watt amplifier.

You certainly don't need 300 watts, but the amplifier will operate over a wide range of supply voltages. This means you can drop the supply voltage and it will survive any VWSR you can give it.

One option might be to run it at 30 watts into a 10 dB attenuator to drive the acousto-optic cell. This should reduce the harmonic distortion and provide immunity to any reflected power. Ebay has a 100 Watt 10 dB

2GHz attenuator for US $29.95:

Communication Concepts, Inc, in Beavercreek, Ohio, has a nice business selling amplifiers based on Granberg's designs. The AR313 is $291.50 fully assembled. It is shown on their page at

They also state:

"As the MRF141G is basically made for operation from a 28 volt supply, lowering the voltage down 20 volts or lower would make the unit almost indestructable against load mismatches such as open coax or a broken antenna."

With lower supply voltage and a 10 dB attenuator, your lab amplifier would be bulletproof.

Oops - it is a kit. But it should only take an hour or so to solder the few parts and mount it in a suitable enclosure. Maybe you know someone who could do it for you.