Cable assembly crossed vs straight

Go To Last Post
22 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi all

In a customer project we have two PCBs that I intend to connect with a cable. The PCBs have Molex 53047-0410 PicoBlade connectors and I want to connect pin 1 of one connector to pin 1 of the other connector so I need a straight cable.

 

I found this on Molex and ordered:

https://www.molex.com/molex/prod...

 

But once I started using them I realize that they are crossed cables; not straight. So pin 1 goes to pin 4, pin 2 to pin 3, pin 3 to pin 2 and pin 4 to pin 1. I cannot find any notion of this in the product information. This was quite surprising. Is this the expected type of cable, would you say?

/Jakob Selbing

Last Edited: Sat. Oct 24, 2020 - 07:08 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Well, it seems to be a matter of "straight, relative to what?" 

 

If it is a ribbon, and same connectors at opposite ends, then the first pin at each end will connect to the last pin at the other end. That is  inherent in ribbon. I've always had to design boards assuming that is the case. Sometimes, cables with insulation-displacement connectors can be built with a half-twist, making pin1 connect to pin 1, but I recall there being something odd with that arrangement (maybe its pin 1 to pin 2). 

 

Now, if it is twisted cable, it is a different ball game. There I typically specify green at each end goes to pin 5, red goes to pin ... . Or, you take it the way it is. Sometimes there are choices, as with ethernet cable and RS232 cable. 

 

In short, you need to do your due diligence up front. More than once, I have forgotten to do so!

 

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'll bet when you laid out the connector on the pcb, you arbitrarily assigned pin numbers to your connector. thus the connector on the right and left boards look like this when facing each other:

 

1    1

2    2

3    3

4    4

 

looking at each connector from the front you will see you changed the numbering order of the connector on one of them, ie, you labeled one 1-2-3-4, while the other is 4-3-2-1...

How ever, if you place the connectors side by side, and mark the same side as pin 1,  turn them to face each other, you will see they look like this:

 

1   4

2   3

3   2

4   1

 

So the cable is correct, in that pin 1 on one end, connects to pin 1 on the other end, looking from above, it looks like they cross, but in fact it is a 1-1 cable.

 

Hope that helps

Jiim

 

 

 

 

(Possum Lodge oath) Quando omni flunkus, moritati.

"I thought growing old would take longer"

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0


 

It's the old mirror image problem,  & who gets to name which pin is which.  Do you want straight through mechanically or numerically? 

 

note A & B are specified as the exact same part & the cable itself appears "straight wire" mechanically  

I can't tell from the dwg if a & B are rotated 180 from each other, but if they can only plug into the mates one way, then they must be spun 180 to match the pin numbers of the C"s.

 

When in doubt, get the ohm-meter out!!!

 

 

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

Last Edited: Sat. Oct 24, 2020 - 07:55 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I diagnosed a crazy system with many connectors and cables (many wires!!) some cables split apart and went to different boards, just to add to the excitement.  The tech had it all wired up & I was called in to investigate. Many signals daisy chained onto to multiple boards.  Ohming out everything between beginning and the very end, every single line seemed fine.   Intermediate points were not so accessible at the board connectors themselves (until wwe tore more and more of the machine apart). 

But very mysteriously, boards were not seeming to get the right data, or other signals, but on another board things seemed fine & yet other specific control lines seemed screwy on different boards, but ok on some of them???!?

All signals seemed to be generated, but many looked rather similar, making 100% certain identification haphazard, even if there were no issue.

 

You see the proper signal on board E, you can look at the schematics and see it arrives from board B to C.  The schematics also clearly shows C delivers it to E.  But how doesn't it show up at the chips on C???  Impossible!!!

 

In the end what happened, was there was mirroring on certain connectors, but when remirrored onto the next board it became "fixed" or mirrored again and "broken".  So looking at the endpoints, all seemed mysteriously well, yet at different places in between it was mayhem since the signals appeared scrambled (and some scrambles were further sent out onto other branching cables).  The final fix was to wire the cables (as needed) to not be straight through & account for mirrors wherever they occured.

 

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:

Sometimes, cables with insulation-displacement connectors can be built with a half-twist, making pin1 connect to pin 1, but I recall there being something odd with that arrangement (maybe its pin 1 to pin 2). 

I don't agree. Typically a ribbon cable with IDCs that are oriented the same way will have straight pin mapping (pin 1 to pin 1, etc). It is obvious if you look at e.g. this picture: https://upload.wikimedia.org/wik...

 

/Jakob Selbing

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ki0bk wrote:

I'll bet when you laid out the connector on the pcb, you arbitrarily assigned pin numbers to your connector. thus the connector on the right and left boards look like this when facing each other:

 

1    1

2    2

3    3

4    4

 

looking at each connector from the front you will see you changed the numbering order of the connector on one of them, ie, you labeled one 1-2-3-4, while the other is 4-3-2-1...

How ever, if you place the connectors side by side, and mark the same side as pin 1,  turn them to face each other, you will see they look like this:

 

1   4

2   3

3   2

4   1

The connector is a library component so it's the exact same pin numbering on all boards. The signal mapping is assumed to be identicaL - e.g. if pin 1 is 5V voltage supply (outgoing) on one PCB, on the other PCB I also assume pin 1 is 5V supply (incoming).

 

ki0bk wrote:

So the cable is correct, in that pin 1 on one end, connects to pin 1 on the other end, looking from above, it looks like they cross, but in fact it is a 1-1 cable.

I don't quite understand your conclusion. The cable I linked to obviously connects pin 1 on one side to pin 2 on other side. Take a loot at the mechanical drawing provided in post #4.

/Jakob Selbing

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

avrcandies wrote:

It's the old mirror image problem,  & who gets to name which pin is which.  Do you want straight through mechanically or numerically? 

I want pin 1 on one PCB to connect to pin 1 on other PCB, etc.

 

avrcandies wrote:

note A & B are specified as the exact same part & the cable itself appears "straight wire" mechanically  

I can't tell from the dwg if a & B are rotated 180 from each other, but if they can only plug into the mates one way, then they must be spun 180 to match the pin numbers of the C"s.

 

When in doubt, get the ohm-meter out!!!

 

 

 

That drawing shows my problem: pin 1 on side A connects to pin 4 on side B. What I wanted is pin 1 to pin 1, pin 2 to pin 2, etc.

 

So my problem is: how do I find the type of cable that I want? Should I look for "crossed" cable when I would rather call it "straight"? I guess I just have to tryu, but then also check mechanical drawing...

 

BTW as a side-note, initially I had no problems with the cables. When I discovered that they were in fact "crossed" (in my sense) I was baffled since it should not have worked. But then I found that the connector on one of the PCB's had accidentally been assigned the reverse pin numbering (and we had different connector types on those PCBs). So it worked only because of 2 mistakes cancelling out each other and because we had changed the connector type after designing the first PCB :)

/Jakob Selbing

Last Edited: Sun. Oct 25, 2020 - 06:40 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

FYI I looked around and found this page:

https://uk.rs-online.com/web/p/w...

 

It seems to be the same crossed cable but still says this:

"The pin to pin single map configuration allows each signal to be assigned to a pin without having to design the cable layout."

 

This has made me really confused...

/Jakob Selbing

Last Edited: Sun. Oct 25, 2020 - 06:49 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

BTW as a side-note, initially I had no problems with the cables. When I discovered that they were in fact "crossed" (in my sense) I was baffled since it should not have worked. But then I found that the connector on one of the PCB's had accidentally been assigned the reverse pin numbering (and we had different connector types on those PCBs). So it worked only because of 2 mistakes cancelling out each other and because we had changed the connector type after designing the first PCB :)

Sounds like our troubleshooting session..except they had more links in the chain of reversals...so if you picked the wrong board a signal would be gone (at some component), but would show on other boards--which seemed impossible, since the "missing signal boards" were sending the signal onwards.  Reason: Say it was supposed to come in on 3 but came in on 15.  Then was supposed to leave on 3 (on another connector) but went out on 15 (reversal fixed)...however the components attached to 3 on such boards never saw the signal, so probing was very confusing!!   Schematically, it made no sense whatsoever (since the schematics are reversal free). 

 

 

The problem is, they don't (or cant') make clear which faces of A & B are showing (maybe they look the same).  But in order to match the C's as drawn,  A & B must be showing opposite faces (for example, you could pull off B, rotate it 180 & put it back on the wires).  

So the wires are physically straight across on the cable, but the pin numbers on the cables themselves are swapped.   

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

Last Edited: Sun. Oct 25, 2020 - 09:16 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jaksel wrote:

FYI I looked around and found this page:

https://uk.rs-online.com/web/p/w...

I don't recommend using that cable - it was designed by a f&*kwit. Doing other than 1-1 will spread confusion for evermore and every technician that works on your board will hate you.

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

avrcandies wrote:
The problem is, they don't (or cant') make clear which faces of A & B are showing (maybe they look the same).  But in order to match the C's as drawn,  A & B must be showing opposite faces (for example, you could pull off B, rotate it 180 & put it back on the wires). 

The two faces are not identical so in the drawing you see the same face of both connectors. Hence it is not the opposite faces. The C view shows the connector from the mating side.

/Jakob Selbing

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

N.Winterbottom wrote:

jaksel wrote:

FYI I looked around and found this page:

https://uk.rs-online.com/web/p/w...

I don't recommend using that cable - it was designed by a f&*kwit. Doing other than 1-1 will spread confusion for evermore and every technician that works on your board will hate you.

 

I totally agree. I would never redesign the PCB to accomodate for that cable.

/Jakob Selbing

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

N.Winterbottom wrote:

I don't recommend using that cable - it was designed by a f&*kwit. Doing other than 1-1 will spread confusion for evermore and every technician that works on your board will hate you.

 

Truly antique floppy drives used to have a section of their ribbon cable sliced out, half-twisted, and put back in.

 

When you have a connector on your board, make the "Pin #1 Pad" slightly different than the others - square instead of round is traditional.  Then just by running your finger down the edge of the cable and looking at the boards you can be sure that pin #1 mates with #1.

 

avrcandies wrote:

 

 

 

 

 

Also note that in the above picture, the connectors 'C' are not the same way around!  For a straight cable, you are not allowed to randomly flip the connectors over.  Or install them on the back of the board.  Or any other gymnastics.

 

Also note that the connectors on the ends of the cable will (should have, anyhow) some indicator of 'pin 1' molded into the plastic housing, as will the plastic housings of the connectors on the PCB.  If you're using keyed connectors, they should always take care of that on their own.  Again, unless you get f***wit cables... 

 

S.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0


this part of the drawing seems odd, can you provide any detail?

Whats in box D, (see Detail A)!!!

 

 

 

(Possum Lodge oath) Quando omni flunkus, moritati.

"I thought growing old would take longer"

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ki0bk wrote:

 

 

this part of the drawing seems odd, can you provide any detail?

Whats in box D, (see Detail A)!!!

 

The drawing says "Blank label 12.7x35.5mm (WHT:12.7mm) HF".

 

 

/Jakob Selbing

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Scroungre wrote:
Also note that in the above picture, the connectors 'C' are not the same way around!  For a straight cable, you are not allowed to randomly flip the connectors over.  Or install them on the back of the board.  Or any other gymnastics.

 

Also note that the connectors on the ends of the cable will (should have, anyhow) some indicator of 'pin 1' molded into the plastic housing, as will the plastic housings of the connectors on the PCB.  If you're using keyed connectors, they should always take care of that on their own.  Again, unless you get f***wit cables... 

These connectors are keyed. You can only mate in one orientation. I already concluded that these cables are indeed crossed (so not a 1-to-1 pin mapping). The mechanical drawing is correct and it clearly shows this. The cables I received look like in the drawing.

 

I just cannot understand why is this not clearly stated anywhere in the product info. In fact if you look at Mouser (https://uk.rs-online.com/web/p/w...) the product info says stuff like "The pin to pin single map configuration allows each signal to be assigned to a pin without having to design the cable layout".

/Jakob Selbing

Last Edited: Tue. Oct 27, 2020 - 03:02 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 

jaksel wrote:
the product info says stuff like "The pin to pin single map configuration allows each signal to be assigned to a pin without having to design the cable layout".

 

Fee-fi-fo-fum,
I smell the blood of a Marketing Man,
Be he alive, or be he dead
I'll grind his bones to make my bread

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It's the old wordplay---what they say can be construed many ways.

 

In fact when saying you want straight through connections, that could be taken more than one way...for example if the two different type boards are facing each other, you could say that means the topmost pin of board A hooks to the topmost pin of board B.

However if the same two different type boards are pointing in the same direction, straight through might mean something different, like pin 1 of A (wherever it is)  connects to  pin 1 of B (wherever that is).  

 

Seems like a great making for a 3 stooges movie!

https://www.youtube.com/watch?v=...

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

Last Edited: Tue. Oct 27, 2020 - 09:42 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I've had the same problem with the Molex Pico Blades and also with a SAMTEC cable. Pin1 on one board goes to Pin4 on another.  SAMTEC made the cables in both ways so correct cables were available--just didn't understand the spec sheet when ordering because it never occurred to me.  For the pico-blade cables, it is not hard to remove the wires and reverse them. Just have to carefully lift a plastic tab. Fixed it on the next board rev.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Fixed it on the next board rev

Better watch which cable gets plugged in!   Billy, this cable is chafed, go get me one outta that loaner unit.

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Cables with insulation displacement connectors can be connected with half-twisted wires, so that pin 1 is connected to pin 1, you can operate and see the result.