Molex vs Terminal Block Connector

Help Support GroupDIY:

drtechno

Member
Joined
Feb 3, 2014
Messages
17
I don't think that's derating with temperature. Switches or contacts should have close to zero resistance or they will heat up. The heating up is a function of the contact failing rather than the other way round. IMHO the contacts have either lost their springyness & are no longer tight or they have corroded. Either way there has been resistance introduced & that combined with a high current is what causes the heat. P=(IxI)/R.
That would be nice if everything was cut and dry. But it isn't. I have repaired thousands of things in the past 30 years from TV's to Arcade games, to the pro audio equipment and intelligent lighting. And the auto-self destructing nature of the Molex connector is constant and the major factor was temperature environment. The contacts themselves de-rate, heat up and burn plastic and is probably from the cheap metal they are using.
 

Rob Flinn

Well-known member
Joined
Jun 3, 2004
Messages
4,541
Location
Sussex, UK
That would be nice if everything was cut and dry. But it isn't. I have repaired thousands of things in the past 30 years from TV's to Arcade games, to the pro audio equipment and intelligent lighting. And the auto-self destructing nature of the Molex connector is constant and the major factor was temperature environment. The contacts themselves de-rate, heat up and burn plastic and is probably from the cheap metal they are using.
You are not the only person round here who has repaired thousands of things over the last 30 years. When they start burning up it is caused by them developing resistance on the contacts probably because of tarnishing. If you look at the pic you posted you can plainly see the burning is localised to one of the pins which has developed a poor contact, This may have been caused by the environment, but if it were the environment alone it would probably be more than just the one pin, since they are all in the same environment.
 

Newmarket

Well-known member
Joined
Oct 10, 2016
Messages
735
Location
Brighton Sussex UK
My "two pennies" on this. Molex KK 2.54mm pitch used with polarised / friction lock headers.
It's all "in the crimp". No solder - as Rob says a properly crimped KK contact doesn't have anywhere to solder it and doing so gives problems. And use the correct crimp contact for the wire size. IIRC correctly there are two sizes for different wire diameter ranges.
I've designed and specified with these connectors over several companies and product areas including pro-audio and also industrial/scientific analytical instrumentation.
Problems - rare. A few instances where the crimp contact was not initially made properly and failed in use. This very probably due to the hand tooling of the crimp. Where crimping was carried out by a subcontractor - automatic wire stripping and crimping - no issues. Some units had been in the field >5 years so likely any issues would have come to my attention.
Signal types used - analogue audio and (mainly) logic level signals and dc power (12/24V).
Can't recall right now the contact material option used but not gold.

Also used larger pitch - 3.96mm and 5mm versions for power connections. AC and DC All good.
I did switch to a different type - still Molex - that had greater contact area and pull out force but is still compatible with the same headers. Or with a more specific header type (foot print compatible) that seemed a better mechanical match. All combinations problem free.

Don't scrimp on the crimp. Yeah - the proper tooling does cost a bit.
 

Gold

Well-known member
Joined
Jun 23, 2004
Messages
2,747
Location
Brooklyn
I did switch to a different type - still Molex - that had greater contact area and pull out force but is still compatible with the same headers. Or with a more specific header type (foot print compatible) that seemed a better mechanical match. All combinations problem free.
Just because it says Molex KK and you have the right pitch doesn't mean the pins, header and housing are compatible. I've had all sorts of problems with pins not seating and headers not making good contact. My crimping skills are fine. I've come to the conclusion that unless you are very careful that you have pins, housings and headers that match they are unreliable.
 

fragletrollet

Well-known member
Joined
Sep 19, 2010
Messages
382
Location
Oslo
My limited experience says the same, I always get pins, headers and housing from the same brand as I've also experienced housings that only fit "their own" headers snugly.
 

Newmarket

Well-known member
Joined
Oct 10, 2016
Messages
735
Location
Brighton Sussex UK
Just because it says Molex KK and you have the right pitch doesn't mean the pins, header and housing are compatible. I've had all sorts of problems with pins not seating and headers not making good contact. My crimping skills are fine. I've come to the conclusion that unless you are very careful that you have pins, housings and headers that match they are unreliable.
Yes - not disagreeing there. The parts I used were advised as compatible by Molex. Agree that you shouldn't just mix'n'match regardless. IIRC there are square pins and round pin options- apart from anything else. And tbh I'd always go with the friction lock option with the taller nylon face.
Reminds me that some years ago Molex transitioned part numbers from old "Engineering" numbers to new "Catalogue" numbers - it was a "challenge" to say the least to identify the correct part to specify. Fortunately I had a good Molex sales contact to advise.
Still, if everything is lined up properly they are reliable. Of course, if dealing with existing equipment you don't get to dictate that.
And there is there is always the issue of non-Molex manufactured copies that aren't manufactured to the same tolerances.

Now if you want unreliable connector systems - look at the the tiny connectors (I forget the range atm) from JST. Their larger crimps are fine and arguably have better hand tools available but with the tiny tiny stuff - endless problems if you cycle the connection more than once or twice. We asked our contract wire people about it and the response was basically - "Yeah - they don't really work in production and everyone ends up dropping some superglue or similar in to secure them".
Meanwhile the Molex picoFlex connections used were not problematic (assembly was subcontracted).
 
Top