High-Power RF Multiplexer

Do you have a question? Post it now! No Registration Necessary

Translate This Thread From English to

Threaded View


I would like to be able to switch one 25 Watt 50-Ohm RF source between
two 50-Ohm loads, at up to about a 10 kHz switching rate,
continuously.  The RF will be at up to 2 GHz.  I would like the
switching between the loads to take about 3 to 5 us, but not more than
about 10 us, and give at least 20 or 30 dB of attenuation toward the
unselected load when fully switched.  And the insertion loss between
the source and the selected load should be as low as possible, and no
more than 3 dB, but preferrably less than 1 dB.

Is it possible?

If it's possible, does anyone know of a commercially-available unit
that already does that?  Or, how difficult would it be to design and
build one?  I assume it's not going to be as simple as just a couple
of MOSFET switching circuits.  What types of devices might be used?

Also appreciated would be pointers to any datasheets or appnotes that
might be relevant.

Thanks,

Tom



Re: High-Power RF Multiplexer


Quoted text here. Click to load it

Without fries? Comin' right up, sir:

http://www.microsemi.com/datasheets/UM9400_Series.pdf

Which diode to use depends on the lower end of spectrum you have to switch.

--
Regards, Joerg

http://www.analogconsultants.com /

We've slightly trimmed the long signature. Click to see the full one.
Re: High-Power RF Multiplexer


On a sunny day (Sat, 05 Dec 2009 10:54:06 -0800) it happened Joerg

Quoted text here. Click to load it

Those guys do not allow you to use wget to get the pdf:
 wget http://www.microsemi.com/datasheets/UM9400_Series.pdf
--20:04:21--  http://www.microsemi.com/datasheets/UM9400_Series.pdf
           => `UM9400_Series.pdf'
Resolving www.microsemi.com... 208.19.99.70
Connecting to www.microsemi.com|208.19.99.70|:80... connected.
HTTP request sent, awaiting response... 999 No Hacking
20:04:22 ERROR 999: No Hacking.

Had to use Opera webbrowser and that stupid adobe plugin to read it...

Hope their diodes were made by people with better brains then those who build
teh website.

Re: High-Power RF Multiplexer


Quoted text here. Click to load it
teh website.


That's the case with many vendors and the reason I still use the buggy
Acrobat Reader.

--
Regards, Joerg

http://www.analogconsultants.com /

We've slightly trimmed the long signature. Click to see the full one.
Re: High-Power RF Multiplexer



Quoted text here. Click to load it

Hi Jan,
Use Opera to grab the file but change the file handler to "Save to Disk"
now you can use what you like to read it.

--
Best Regards:
                     Baron.

Re: High-Power RF Multiplexer


On a sunny day (Sat, 05 Dec 2009 21:24:02 +0000) it happened Baron

Quoted text here. Click to load it

Yes that is a logical way to do it, but it does require people to start a
browser.
I saved the file from the pdf.

Re: High-Power RF Multiplexer



Jan Panteltje wrote:
Quoted text here. Click to load it
browser.
Quoted text here. Click to load it


   No, it doesn't. I routinely save PDF files from my Netscape 4.78 news
reader program that I'm using to read this group . Right click and save
link as.  I've used it for over 10 years with no problems.


--
The movie 'Deliverance' isn't a documentary!

Re: High-Power RF Multiplexer



Quoted text here. Click to load it

Yes Michael, you're right !  I'm so used to just clicking the link I'd
forgotten that works too. ;-)

--
Best Regards:
                     Baron.

Re: High-Power RF Multiplexer



Quoted text here. Click to load it

No !  The browser is started when click the link.  You do have to close
it afterwards, assuming you don't want to use it for something else.

Just direct Opera to put the pdf in a directory reserved for that
purpose.  If its useful then I keep it, if not...

Open SuSE 10.3 KDE 3.5.10

--
Best Regards:
                     Baron.

Re: High-Power RF Multiplexer


On a sunny day (Sun, 06 Dec 2009 17:29:08 +0000) it happened Baron

Quoted text here. Click to load it

Dunno, different newsreader I suppose..
I just cut and paste the link, although NewsFlex is able to get it by itself.
Indeed, kust tried, I can just add it to the URL list, mark it,
and select "get marked URls", and it actually works.
Would you believe that? Programmed that 10 years ago, forgot all about it.
My newsreader:
 http://panteltje.com/panteltje/newsflex/index.html
Much better then anything else really.
It can even download websites, but in the last 10 years so many new http things
have appeared,
that I think that no longer works reliably.

Re: High-Power RF Multiplexer



Quoted text here. Click to load it

Opps, Should have said my news reader is Knode !

--
Best Regards:
                     Baron.

Re: High-Power RF Multiplexer


Quoted text here. Click to load it

tried to set the user-agent in wget to something like ie or firefox?

-Lasse

Re: High-Power RF Multiplexer


On a sunny day (Sat, 5 Dec 2009 13:33:15 -0800 (PST)) it happened
Quoted text here. Click to load it
Hey, great idea!
wget --user-agent=opera http://www.ieee.li/pdf/pin_diode_handbook.pdf
works!
Had forgetten that option.
Thank you.

Re: High-Power RF Multiplexer


Quoted text here. Click to load it

Sorry, Joerg.  I didn't want to irritate people by under-specifying
it.

Quoted text here. Click to load it

Thanks!  I also found this:

http://www.ieee.li/pdf/pin_diode_handbook.pdf

Quoted text here. Click to load it


Re: High-Power RF Multiplexer



[...]

Quoted text here. Click to load it

That's a great place to start if you have never used PIN diodes before.
Agilent also has good app notes (from the HP days).

Once you get the hang of it PIN diodes are really cool and
cost-efficient RF switches. Just keep rectification effects in mind,
especially at higher power levels. I'd try to find some with long
carrier lifetimes in the microsecond range, but that depends on what
your lowest frequency will be.

--
Regards, Joerg

http://www.analogconsultants.com /

We've slightly trimmed the long signature. Click to see the full one.
Re: High-Power RF Multiplexer


Quoted text here. Click to load it

It looks like this should work:

http://www.jfwindustries.com/documents/50S-1821_specsheet.pdf

http://www.jfwindustries.com/catalog/High_Power_Solid_State_50_Ohm-78-1.htm =
l

---------------------------------

Now, another QUESTION:

That switch requires +15Vdc at 80mA and +5Vdc at 40 mA; about 1.4
Watts total.

It would be very nice if I didn't have to run any power leads to the
switch.

If I'm willing to sacrifice 2 to 3 Watts of the RF, does anyone see
any problem with making a small RF-to-DC power supply?

If the RF is a 1-2 GHz sinusoid of 25 Watts in 50 Ohms, that should be
about 35.36 V RMS, or 50.0 V Peak (100V P-P).  What would be a good
way to get that down to 15 VDC/80 mA and 5 VDC/40mA?  The only two
even-possibly-reasonable ways that I can think of, off the top of my
head, are: 1) rectify, smooth, and regulate it to DC and then feed
that to a switch-mode supply to produce the 15V and 5V, or, 2) use a
step-down RF transformer to lower the peak voltage and then rectify,
smooth, and regulate.  Option 2 could involve two secondary windings,
for both the 15V and 5V, or I could just do 15V and then also regulate
it down to 5V, since it's only at 40mA.

I've never done a DC power supply design using an RF input frequency,
and have just now started thinking about this.  Would there be any/
many "gotchas"?  What would be the best path to try to take, here?

IF the required RF step-down transformer would be small, and available
off-the-shelf somewhere, then that way sounds simpler.

Now I'm suddenly wondering if there are even any suitable smoothing
caps, for those frequencies and voltages.  Maybe a motor-run
polypropylene?  Does anyone have any ideas?  Or is there some entirely-
different approach that I should look at?

Maybe I should have done some homework, before asking.  But I thought
that a quick reality check, first, might be very helpful.  All ideas
and opinions will be greatly appreciated!

Thanks,

Tom

Re: High-Power RF Multiplexer


Quoted text here. Click to load it

Or, maybe, at GHz frequencies, the smoothing will be much, much easier
than at 60 Hz?  Maybe it could be done with simple LC (or RC) low-pass
filters?

Re: High-Power RF Multiplexer


Quoted text here. Click to load it

Looks like mucho Dolares :-)


Quoted text here. Click to load it

Nope, can be done, you just have to make sure that it bridges any gaps
that might be there when RF drops out for a moment if that's a concern.
Also, keep in mind that rectifying means the switch will be "in limbo"
until the power supply comes up (which will not be instant). You don't
want to fry the PIN diodes in there during the first millisecond of so.
Best to talk to the manufacturer about any "grace periods" here but
chances are they won't know because they may have never tested that.

It is possibly to make switchers that can bring the output up to spec
voltage in under 200 microseconds. But that wouldn't help much if that's
enough to make the PIN diodes in your switch module go phssst ... *PHUT*
because they saw RF power before it was switched to a defined state.


Quoted text here. Click to load it

At gigahertzes I'd forget transformers. Yeah, you could do
stripline/microstrip stuff but that gets old when your frequency range
is wide. You could just rectify and then use two buck converters to get
from your 40-50V down to 5V and 15V. Use some automotive rated switcher
chip that is ok up to 100V. Linear Technology etc. Rectifier diodes for
that voltage level and GHz-range is another matter.

Make sure nothing "spills back" into the RF. When the load is varying
you will effectively AM-modulate your RF source via load changes. So
think about whether that could cause some grief and how much AM you can
tolerate. If the switch control draws fairly constant current that can
be ok but there may be a load glitch when it swings.


Quoted text here. Click to load it

You need very small ceramic RF caps and they need to be rated 100V. 63V
would be a white-knuckle ride. Haven't looked but try Johanson. If you
get stuck on the caps post again. Then, if you fear load changes or RF
drop-outs, follow that with a choke and an electrolytic.


Quoted text here. Click to load it

I think this can be done.

--
Regards, Joerg

http://www.analogconsultants.com /

We've slightly trimmed the long signature. Click to see the full one.
Re: High-Power RF Multiplexer


Quoted text here. Click to load it

Thanks, Joerg!

I am still having trouble getting my head around the interface between
the 25W of (say 2 GHz) RF and the RF-to-DC converter circuitry.
Everything else will already be 50 Ohms, with N-type connectors.
Would I have to use something like a directional coupler?  Or am I
missing something obvious?

Re: High-Power RF Multiplexer



Quoted text here. Click to load it

A directional coupler may or may not be a good idea.  In either case i20%
would expect not more than 6 db, maybe less.  The advantage is less AM
caused by your leach device.

Site Timeline