1. Home
  2. General Electronics
  3. Microwave Circuits OK with FR4? (5GHz)

Microwave Circuits OK with FR4? (5GHz)

For a graduate project, I'd like to design a simple small signal amplifier, centered at 5GHz. I would like to etch this at home, and FR4 .031'' copper clad boards are easily available. (FYI- I am deciding between Toner Transfer method, or Photofabrication.) I understand FR4 isn't even spec'd above 1GHz. It's lossy and the inconsistent dielectric constant might cause discrepencies from simulation.

As this is for college/learning only, I can live with imperfections. Losing a couple dB of gain is fine, and I'm only gonna operate at room temp. If the inconsistent dieletric centers my design at 5.2GHz instead of 5.0GHz, I can live with that too.

Yes, I'm aware Rogers or Teconics have sample programs. Will look into that too.

So is FR4 ok for my purposes of proving that I can design an amp with actual gain? (Hopefully positive gain!)

bill I cross posted this msg to .design and .misc. hopefully I understand their charters correctly.

Reply to
bwillis88
Loading thread data ...

Unless you have very little money and lots of free time, I'd have the board fabricated.

Even for high quality materials the dielectric constant tends to vary more than you might "like." But you're correct that FR-4 is quite lossy... the loss tangent is in the ballpark of 0.02.

Sure; you seem well aware of FR-4's limitations, and since I imagine your entire PCB isn't going to be much larger than a couple inches on a side, you can definitely get away with building an amp and losing a bit of gain.

Were you planning to route signals with striplines or microstrips?

Reply to
Joel Kolstad

Hello Bill,

It's certainly not ideal but I have seen lots of 2.4GHz circuitry on FR-4 and even some 5GHz wideband stuff. Those were consumer products where every penny counts. Even my 802.11b/g wireless card is done with FR-4. So IMHO I'd give it a shot if it's not for anything commercial.

Get the best quality board material you can buy. There are some papers about this topic but I don't have links. Except this one which requires registration:

Regards, Joerg

formatting link

Reply to
Joerg

Your biggest problem will be to few vias, as they need a lot of space and a lot of work when done manual, and the ground plane being too far away from your component side resulting in a long way for the ground vias, resulting in excessive inductance for the vias...

--
Uwe Bonnes                bon@elektron.ikp.physik.tu-darmstadt.de

Institut fuer Kernphysik  Schlossgartenstrasse 9  64289 Darmstadt
--------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------
Reply to
Uwe Bonnes

You should go ahead and do it, document everything to the hilt, measure all of the parameters you possibly can in your prototype, and do a paper on it - you might even break new ground! :-)

Good Luck! Rich

Reply to
Rich Grise

FR-4 is OK up there, given your needs. Just try to keep everything zero-length! Try to keep the backside mostly ground plane. Flange-type SMA connectors, with the flange soldered to grounds on both sides, make excellent launchers.

Coplanar waveguide works well, with lots of vias, just thru-holes with wires. Flange-type SMA connectors, with the flange soldered to grounds on both sides, make excellent launchers.

You could sweet-talk the laminate guys out of some samples of rigid microwave laminates, but it's hardly worth it.

What are you building?

John

Reply to
John Larkin

positive gain is easy, getting it above one is the hard bit.

--

Bye.
   Jasen
Reply to
Jasen Betts

Gore just introduced a new material that is intermediate to existing technologies, SPEEDBOARD C Prepreg:

formatting link

Reply to
Fred Bloggs

At 5GHz getting it beyond 0.9 is often hard (~1dB loss), even across short distances. :-)

Reply to
Joel Kolstad

Hello Fred,

formatting link

Thanks for that link. Spec'd to 40GHz, wow! Wonder what it costs.

Regards, Joerg

formatting link

Reply to
Joerg

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.