Spec Differences Between Type N and 3.5mm/SMA

Started by Hawaii596, 10-29-2009 -- 10:49:07

Previous topic - Next topic
Print
Go Down Pages1
User actions

Hawaii596

10-29-2009 -- 10:49:07
I don't recall if it was here or on the elsmar forum where a comment was made about the difference between N and 3.5mm (or SMA?).

There was some input about type N having some better bandwidth capabilities than 3.5 mm.

Anyone know of any good documents with those spec differences?  I've always had some presuppositions that may be wrong.  I'd like to correct my thinking if that is the case.
"I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind."
Lord Kelvin (1824-1907)
from lecture to the Institute of Civil Engineers, 3 May 1883

RichieRich

#1
10-29-2009 -- 11:45:15
Some good information about 3.5 / SMA here.  http://www.gore.com/MungoBlobs/671/522/Intermateability_Microwave%20Journal_Pino.pdf
It doesn't cover N, but I think you had some questionable information.  3.5 MM should have better vswr and frequency response than N.  I think the main advantage to N is mechanical ruggedness compared to any of the smaller connectors.  SMA is better suited to internal connections as it was never intended for repetitive use.

Hawaii596

#3
10-29-2009 -- 13:49:13
I'm a self-confessed "Jack of All Trades" (i.e.: I have moderate experience in a lot of different disciplines, but not highly expert  in very many of them - worked in a lot of different labs including a few one-man labs). 

And I always thought I understood the spec differences bandwidth-wise between SMA, 3.5mm and Type N (including the differences between air dielectric and various forms of teflon, etc.). 

Well, as it turns out, I DID understand correctly.  I need to go back and find the post.  I was corrected by someone who said I was wrong in stating that 3.5 mm has a higher bandwidth than Type N.

Thanks for those links.
"I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind."
Lord Kelvin (1824-1907)
from lecture to the Institute of Civil Engineers, 3 May 1883

Wilk

#4
10-29-2009 -- 18:48:36
3.5mm does have a larger bandwidth than type N.  3.5mm is very good up to about 26.5 GHz, 32-36 GHz in some cases on very high end connections. 

N started out as a 4 GHz connection, then 12 GHz, and now 18 GHz after some modification by Amphenal in the 80's.  So you have to be very careful when looking at N conenctions as they are not all built the same.  On the female side, 4 fingers = 12 GHz, and 6 fingers = 18 Ghz.  It is quite a bit more complicated than that, but thats the easyist way to identify them.  On the male side, 4 fingers on the outer conductor = 12 Ghz, and solid = 18 GHz.

The one place N can out class 3.5mm is actually at low frequencies.  When dealing with precision connections, N is the more stable connection at frequencies under about 800 Mhz.  This is due to the fact that all RF connectors center to outer conductor sizes are based on the 1/4 wavelength of the desired carrier wave.  3.5mm is much smaller than the 1/4 wavelength of 1 GHz so you can end up with as much of the signal traveling on the outer conductor as you do the inner.  This can cause major phase issues and has large effects on the effective VSWR.   

SMA is a entirely different beast, and is a 18 GHz low cost connection that relies on dielectic loading of the center conductor VS accuracte construnction to provide a decent VSWR.  It does physically mate to 3.5 mm, but you can damage both connections in doing so.  Thus you mainly only see it used inside equipment where connections are never broken, and power losses can be compensated for.

As far as written papers go, Maury Microwave has the best stuff I have seen.

RFCAL

#5
11-07-2009 -- 15:01:02
the previous post is right on!! 100% correct

Hawaii596

#6
11-10-2009 -- 19:02:40
Good stuff!!

And I can attest to VAST difference in type N's.  I have a set of the HP N-3.5 adapters (I found them for $20 each on ebay, versus sticker price of around $179 for one and $329 for the other).  I have two of each kind...

Anyway, There is a big difference when checking type N 8490 series stick attenuators with different type N adapters.  The HP one has a solid ring around the center conductor, and some othe cheap brand ones like Pasternack, etc.  have four sections to that ring (if you know what I'm referring to), and I think it acts like stray capacitances; not to mention the poor quality and dimensional accuracy of the center pin..

Well, gotta go home and do my duty watching Dancing With The Stars.
"I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind."
Lord Kelvin (1824-1907)
from lecture to the Institute of Civil Engineers, 3 May 1883

jimmyc

#7
12-04-2009 -- 14:10:22
funny statement in there about 7mm connector

"APC-7 stands for "Amphenol precision connector", 7mm. Developed in the swinging 60s, ironically a truly sexless connector, which provides the lowest VSWR of any connector up to 18 GHz.

a couple other statements on http://www.microwaves101.com/encyclopedia/connectors.cfm were kind of scary.  talks about SMA to 3.5 and 2.9 mating.  i can't count the number of destroyed 3.5mm connectors (esp.on splitters for some reason) i have seen due to mating the two. i say don't do it.

Bryan

#8
12-04-2009 -- 20:00:45
Unfortunately if they'll thread together it's going to happen.
I have noticed Tek publications imply that SMA to 2.92 is fine, Agilent says gage them first.  Wiltron/Anritsu on the other hand made thier own 3.5 & called it K, ditto 2.4, they call it V

Duckbutta

#9
12-04-2009 -- 22:27:59
3.5 and 2.92 don't "thread" together. You'd have to be a complete moron to think it was mating correctly.

Anritsu's K connector does mate with 3.5. The difference is the K is rated to 40 GHz, whereas the 3.5 is only rated to 26.5 GHz, so they really aren't interchangeable.

jimmyc

#10
12-07-2009 -- 10:47:42
http://www.mwjournal.com/journal/article.asp?HH_ID=AR_4000

here is a good article about mating, but i still wouldn't do it if i had other choices.

RFCAL

#11
12-07-2009 -- 15:12:07
N and SMA=18gHz ; 3.5mm = 26.5 gHz ; 2.4 mm = 50 gHz ; 1mm = 110gHz. Generally, you get what you pay for.And there will be damage if you try to cross them-guarnteed.--Pasternak is cheaper,but only good for accessories. Lab connectors and adapters should be high quality,high $$

Talbot_ESI

#12
12-08-2009 -- 06:35:38
I read that mating 3. 5 to SMA is fine as long as you torque to 5 inch pounds

RFCAL

#13
12-08-2009 -- 12:04:16
You should NEVER mate an SMA to a 3.5 mm.! This is hoe degradation of cables and connectors / adaptors occurs
Print
Go Up Pages1
User actions