[TowerTalk] Long Term Connector Reliability Re: TowerTalk Digest, Vol

[Lux, Jim]

On 7/25/22 9:17 AM, Drew Vonada-Smith wrote:
> All,
>
> Before anyone decides to engineer a full-scale change from UHF to N connectors, whether it be due to "electrical superiority", "water resistance", "reliability", or any other reason, I suggest that you poll the experiences and opinions of a few large station or multi-multi users who have worked with both.  You are likely to exit that conversation with a very different opinion.
>
> The devil is in the details.  Feel free to email me if interested.
>
> 73,
> Drew K3PA


This is very much so. I would say that one might also be influenced by 
the following:

Am I doing 1 connector or 100?

Am I doing the work myself, or am I ordering cables with the connectors 
assembled?

Generically, for pretty much all RF connectors, there's multiple 
workflows to put the connector on the cable - some are "one at a time, 
using simple hand tools" , some are "use specialized tooling (crimpers, 
etc.)" and that has a lot of effect on the ultimate cables.

Even between crimping there's a difference between "assembled one by one 
using a hand crimper hanging from a harness on a tower" and "assembled 
as a lot of 100 cables, using a air powered crimper in a factory with 
rotary strippers and assembly jigs".


I will note (having bought hundreds of cables at work over the years) 
that there are foul ups, even in a spaceflight qualified* assembly line, 
so another factor might be "do they test the cables before shipping" and 
"what's their return policy" - everyone I've ever worked with is happy 
to replace a defective cable, but there are differences in "how quick" 
you might get it.  If you have to wait 3 months for the next production 
run of that kind of connector that's different than "we'll ship it to 
you tomorrow".

*a space qualified process might actually be worse than a big mass 
production, since space stuff tends to be artisanally hand crafted with 
lots of inspection steps and handling.  Give me a cable made as a lot of 
a million going into a mass production item with high 
servicing/replacement costs - There's an incentive to "get the process 
right".   For instance, automotive engine control units have very low 
failure rates in service.  A 0.1% failure rate would be crippling to a 
car maker.