RF Attenuator and connector behavior after exceeding frequency rating.

[robert1111]

Hello.  What is the result when the applied frequency greatly exceeds the frequency rating of a RF attenuator?  For example, applying a swept signal from 3GHz to 40GHz for a 3GHz rated N type coaxial attenuator?  How about for a 18Ghz rated coaxial SMA attenuator?  What happens when a swept signal from 3GHz to 40GHz is applied to a N type connector that is only rated for 11Ghz or 18Ghz?  What is the physical difference between the 11Ghz and 18Ghz N type connector?  What happens when a swept signal from 3GHz to 40GHz is applied to a SMA connector?  Thank you for your help.

[xmo]

There are no 40 GHz 'N' connectors.  Exceeding the connector's rated frequency can introduce “modes” in the frequency response. A “mode” occurs when the connector structure becomes a weakly resonant cavity. This resonance would appear as a very narrow band “suckout” in the connector insertion loss.

[robert1111]

There are no 40 GHz 'N' connectors.  Exceeding the connector's rated frequency can introduce “modes” in the frequency response. A “mode” occurs when the connector structure becomes a weakly resonant cavity. This resonance would appear as a very narrow band “suckout” in the connector insertion loss.

Does this mean the attenuation can be greatly reduced in the mode frequency?  By "suckout" do you mean increase or decrease in insertion loss value?

[TheSteve]

Often bad things happen. You can exceed the frequency of the connector itself, you may experience moding as a result. The bigger issue is that attenuators often perform very poorly beyond the specified frequency.

Let's take a look at a Mini Circuits 10 dB SMA attenuator from 300 kHz to 26.5 GHz. This is part # VAT10+, with a rated frequency of 6 GHz
You can see it performs just fine to 6 GHz. Beyond that all parameters are less then ideal.

[TheMG]

With connectors, the result would usually be an increase in insertion loss at certain frequencies worst than others. It won't be a nice gradual increase as you go up in frequency, there will be all sorts of dips and bumps.

As for RF attenuators, these can go both ways, the attenuation value may go both above and below the attenuator's spec.

[ejeffrey]

The difference between an 11 GHz and 18 GHz connector of the same type is mostly precision manufacturing.  A lower spec connector, one that has been abused, or not torqued correctly will have worse insertion loss and worse return loss at high frequency because it will not present a constant impedance and may have more material loss.

If you go higher in frequency the other factor is multi-mode behavior when the frequency exceeds the waveguide cutoff for the ground shield.  This is what happens if you try to use an N connector at 40 GHz, and why 40 GHz connectors have to be so small.  This causes resonances and beating between the modes.  Any bend in the cable will cause mixing of the modes so the performance is not just bad, it is unstable as slight changes can cause a dip to turn into a peak or vice versa.

When talking about coaxial attenuators and filters, multi-mode behavior also degrades performance significantly since the waveguide modes typically bypass the attenuator or filter component unless you take special precautions.  For instance, a segment of coaxial transmission line with the dielectric filled with eccosorb can be made that will heavily attenuate from ~20 GHz to visible light (literally!) including waveguide modes.  These filters can't really be made with precise or sharp cutoffs but you can cascade them with a conventional lowpass or bandpass filter and make a filter that has no higher order transmission.

[robert1111]

Thank you for the replies.  Thank you TheSteve for measuring the VAT10+.  Do you or anyone have a 18GHz SMA attenuator they can test?  Or a 6GHz SMA Weinschel attenuator they can test?  The VAT10+ has high VSWR of 1.9 typical from 5 to 6 Ghz.  This is not typical for attenuators made by companies such as Weinschel.  Thanks.

[robert1111]

"If you go higher in frequency the other factor is multi-mode behavior when the frequency exceeds the waveguide cutoff for the ground shield."

Hello.  What formula do I use to calculate the "waveguide cutoff for the ground shield"?  Which modes are present for a N connector at 30GHz?  Which modes are present for a N connector at its rated frequency?  Thank you.

[ejeffrey]

"If you go higher in frequency the other factor is multi-mode behavior when the frequency exceeds the waveguide cutoff for the ground shield."

Hello.  What formula do I use to calculate the "waveguide cutoff for the ground shield"?  Which modes are present for a N connector at 30GHz?  Which modes are present for a N connector at its rated frequency?  Thank you.

You can find the formula and more information here: https://www.microwaves101.com/encyclopedias/coax-cutoff-frequency along with cutoff limits for various precision connectors.  I believe that N connectors are similar size to 7 mm, which according to the table there puts the TE11 cutoff at around 19 GHz.

[djidji]

an very very good and very very very expencive N connector will have satisfactory characteristics up to the 21,5GHz, but if you need or require very low insertion loss and excellent return loss, than N connector is usable up to the 12GHz only. for higher frequencies you must use the SMA (up to the 26GHz or APC7 (up to the 18GHz) then K (good up to 40GHz) or APC3.5 (good up to 26GHz). for frequencies above 40GHz comes the RPC-1.85, RPC-1.35 and RPC-1.3 (110GHz) connectors.

[TheUnnamedNewbie]

Thing to keep in mind is that you can sometimes operating things above the higher mode frequency. Just because they can exist, does not mean they will be excited, and if the propagation constants are sufficiently different very little energy will transition from the desired mode into the undesired mode, provided there are very little defects or other structures that can excite this mode.

This is how you get N-connectors going up to 26 GHz, and BNC connectors going up to like 10 GHz on some Keysight scopes.

[Joel_Dunsmore]

an very very good and very very very expencive N connector will have satisfactory characteristics up to the 21,5GHz, but if you need or require very low insertion loss and excellent return loss, than N connector is usable up to the 12GHz only. for higher frequencies you must use the SMA (up to the 26GHz or APC7 (up to the 18GHz) then K (good up to 40GHz) or APC3.5 (good up to 26GHz). for frequencies above 40GHz comes the RPC-1.85, RPC-1.35 and RPC-1.3 (110GHz) connectors.

There are "grades" of connectors: commercial, precision and metrology.    N connectors have a 7 mm outer conductor and so have the same limitations as APC7 in a precision version.  SMA is really only good to 18 or 20 GHz due the solid dielectric (teflon) and has a 3.5 mm outer conductor inner diameter.  The air dielectrice 3.5 mm connector is good past 26.5 and pretty usable up to mid 30's.   Commercial versions don't  have well defined transitions and so form kind-of low-pass filter structures (might have, for example, too wide of a center conductor causing excess low impedance like a shunt capacitor).   2.92 mm (K connector) is good to 40 or a bit more; 2.54 mm is common up to 50 GHz and is the Keysight standard.  then 1.85 to 67 (but works OK to 72 ish), then up to 1 mm for 110 or 120 GHz (depends on the bead).

For your attenuator question: it depends all on the design of the series and shunt resistors.  Keysight uses a thin-file "patch" resistor which is quite broad band but quite expensive. Low cost attenuators use lumped SMT parts that have higher parasitics and fall apart out of band.