Much less so because they are not the same characteristic impedance. Coaxial cable acts more like a typical transmission line. IDC Cable is more like (well is) parallel straight wires. 🙂 Of course it does have a characteristic impedance, but it is much more like a twisted pair then coax cable. Still it is fair to note that neither is a perfect transmission line. We do stagger signal and GND for the IDC.
Also while we supply IDC, I do not propose it is the best possible interconnect. It is just very practical while still working quite well. If you want the best wiring then you should solder wire direct. There is no beating that when routed well.
Also while we supply IDC, I do not propose it is the best possible interconnect. It is just very practical while still working quite well. If you want the best wiring then you should solder wire direct. There is no beating that when routed well.
Hey Russ,
another worth checking out is the Microdot connector. It's between SMA and the wifi connectors size-wise and have the possibility of termination without too many proprietary tools.
another worth checking out is the Microdot connector. It's between SMA and the wifi connectors size-wise and have the possibility of termination without too many proprietary tools.
I prefer soldering the lines directly to the board not using any connectors other than the ones at the back panel.
Adapters would be great, so I can avoid them... ;-)
I prefer to solder the wires directly or via the pin headers so no need for special connectors or adapters. Keep it simple.
I prefer to solder the wires directly or via the pin headers so no need for special connectors or adapters. Keep it simple.
FYI here is the data sheet for the w.fl in question:
http://www.hirose.co.jp/cataloge_hp/e32124820.pdf
It would indeed be important to stress that diy'ers leave it to the last minuite to make the connections.
The board mount jacks will do ~20 mate-unmate cycles no worries (for our usage this would probably be a lot more in actual practice).
But as Russ said, the connector assemblies will start to go "bad" after as little as 6.
This can be mitigated somewhat by removing them properly... if you pull them up by the cable then you're DOING IT WRONG!
http://www.hirose.co.jp/cataloge_hp/e32124820.pdf
It would indeed be important to stress that diy'ers leave it to the last minuite to make the connections.
The board mount jacks will do ~20 mate-unmate cycles no worries (for our usage this would probably be a lot more in actual practice).
But as Russ said, the connector assemblies will start to go "bad" after as little as 6.
This can be mitigated somewhat by removing them properly... if you pull them up by the cable then you're DOING IT WRONG!
Russ & Brian,
The adaptors, as far as I understand, would be optionals and are therefore optional🙂
Offering different choices has always been a strength of TPA so please do make them available - but do it soon....
I would use them with my BII, BIII and teleporter module🙂
Cheers,
Nic
The adaptors, as far as I understand, would be optionals and are therefore optional🙂
Offering different choices has always been a strength of TPA so please do make them available - but do it soon....
I would use them with my BII, BIII and teleporter module🙂
Cheers,
Nic
Hi NicMac,
I hear you.
What I am trying to determine if its even worthwhile. 🙂
I would really like to get some more input - especially the empirical type. 🙂
I hear you.
What I am trying to determine if its even worthwhile. 🙂
I would really like to get some more input - especially the empirical type. 🙂
Sorry to have to disagree but SMA connectors are rated for only one mating/unmating cycle before their impedance versus frequency plot starts to degrade (usually at the high end first).
This is according to an Agilent instructor I had, for a week-long course in RF and Microwave Measurement Fundamentals (great course, by the way). He spent a couple of hours just on connectors.
SMA has only one available grade: "crap", whereas the N-type, 3.5mm, and even BNC types (and many other types) have different grades available (usually three or four grades), from crap-grade up through metrology-grade.
Agilent uses SMA only for internal PCB connections, since those are only mated once, usually. If you like the SMA form factor, definitely use 3.5mm instead, if there is to be more than one mating/unmating cycle. (BTW, never use SMA and 3.5mm together, even though they might look identical, because 3.5mm has a strict standard for the male pin length and SMA does not, and using a non-conforming SMA male can often ruin the RF characteristics of the 3.5mm female.)
As a somewhat-shocking side note: Even the best N-type RF connectors are rated for only 100 mating/unmating cycles! After that, their impedance/frequency response starts to degrade. So if you have any expensive Agilent or other RF equipment, it's a very good idea to install a M-F adapter on every external RF connector, when the equipment is new. Then replace the adapter every 100 mating/unmating cycles. Notice that you still only get 10,000 mating/unmating cycles (after using 100 adapters) before you need to have the original external connectors replaced.
Cheers,
Tom
Last edited:
Coris,
You'e welcome.
Regarding SMA connectors, the degradation effects might stay mostly in the upper part of their frequency range. I haven't seen much data on that. But it might mean that they are fine for repeated lab-type use if we are only talking about going up to a few GHz or so. Still, it's a good idea to check every cable and path with a network analyzer, if it might matter. (And a few GHz in frequency would be for sinusoids, not for the repetition rate of analog pulses, of course. For pulse or digital signals, it could be a limit for the rise time, which means 1 / ( Pi f ) for the required highest frequency content, which would limit the rise time to something like 64 ns or more, if we could assume that they were good for up to 5 GHz. But, for plain digital signals, and in cases where pulse edge shapes shouldn't necessarily matter, maybe the sinusoid frequency limit could be OK for the repetition rate.)
Another very-important piece of really-basic RF connector information, which I learned the hard way (by doing it wrong in front of a senior engineer on my first day in a real RF lab, years ago), but which the Agilent instructor also stressed, is that, when mating connectors, one should NEVER allow the male pin to turn, inside the female connector, while mating them and screwing their threaded portions together. That is especially easy to do incorrectly when one is mating an adapter to something. (And that was especially embarrassing when done in an RF lab where almost everything was metrology grade and each adapter's price was in the hundreds of dollars, or more.) You have to be sure to hold the right part of the connector or adapter in a fixed position and rotate ONLY the outer threaded portion.
The torque used, when tightening the threaded portion during mating, is also important. Too much or too little is bad. Too much is worse, in terms of the long-term quality of the RF characteristics of the connector. Ideally, a calibrated "handle-break" torque wrench, designed for use with a specific RF connector type, should be used. Each connector type has a specified torque and torque tolerance, for tightening, in order for the resulting connection to be within the specifications of the connector type.
Regards,
Tom
Last edited:
I found an interesting page on this subject:
RF Connectors / test equipment connectors from WA1MBA
I'm not technically knowledgeable enough to have an opinion on the subject. Personally, I like terminal blocks/plain old solder, unless there's a good reason to use a connector.
RF Connectors / test equipment connectors from WA1MBA
I'm not technically knowledgeable enough to have an opinion on the subject. Personally, I like terminal blocks/plain old solder, unless there's a good reason to use a connector.
- Status
- Not open for further replies.