Hi. Pls, what could be the loss tangent of FR-4 (Lossy) at 3.8 GHz? Is there any formula to determine this at varying frequencies?

Link to old thread:
https://forum.allaboutcircuits.com/threads/fr4-loss-tangent.165820/post-1469585

 

FR-4 at frequencies up to 100 MHz gives tan delta about 0,015. Woeful, byt so it sad to tell, is.
I dont know precise value of transition frequency for FR4, but guess it may happen be around 100 MHz.
Then applying the standard formula tan delta effective is tan delta at low freq multiply with sqrt (work freq divide to transition freq) gives 0.015*sgrt(3800/100)=6^0.016=0.1 Tragic result. My cauncel - apply the Rogers biclad teflon based PCB having 0.00003 at frequencies up to 40 GHz and never You will fail. It is yet bit more expensivish, about 200-400 USD per A3 size, but it is worth to pay for to turn to be a convinced alpha male.

 

Hi. Pls, what could be the loss tangent of FR-4 (Lossy) at 3.8 GHz? Is there any formula to determine this at varying frequencies?

Link to old thread:
https://forum.allaboutcircuits.com/threads/fr4-loss-tangent.165820/post-1469585

I did some searching and found a paper stating 0.0476 at 4.04 GHz.
https://www.researchgate.net/figure...alues-of-FR-4-and-BV-Substrate_tbl2_318573849

Other papers on FR-4...

Wideband frequency-domain characterization of FR-4 and time-domain causality
https://www.semanticscholar.org/pap...ilji/007068629033a7ac017f2b2dcfaccf50912cad4c

FR-4 Dielectric Constant https://resources.pcb.cadence.com/blog/2023-fr4-dielectric-constant

Limits of FR-4 in High-Speed Designs https://www.ieee802.org/3/10G_study/public/jan00/sayre_1_0100.pdf

Understanding when to use FR-4 laminates or High Frequency Laminates https://www.ipc.org/system/files/technical_resource/E6&S21_03.pdf

 

RE:"" Is there any formula to determine this at varying frequencies? ""
In the zone where resonant frequency is very far away, the tan delta is stable and not varies within the frequency. In the range where resonant frequency is one order of magnitude (guess 10%....100%) then formula is tan(delta)eff=tan(delta)low.freq *SQRT [f(work)/f(reson)].

 

RE: Dovo: - thank You very much. I didn`t met those articles before. It are useful for my daily work.

 

The "transition frequency" of FR-4 was mentioned. If by transition frequency you mean where dielectric losses exceed copper losses we can calculate that with paper and pencil. Copper losses are proportion to the square root of frequency whereas dielectric losses are proportional to frequency (assuming loss tan is the same at all frequencies). I will post this if anyone is interested.