A4Dtux Mistral 69 Jouef and A8u PBA Trix coupled with the 4-pole electric coupling.
For a long time, I have been testing electrical couplings for (almost) non-separable trains: suburban trains, TEE. I have made almost as many attempts as I have different trains: RIB, VB2N, RGP, TEE PBA and Mistral… The time has come to move on to a serious realization.
This coupling must have four contacts. In fact, it must be suitable for train sets in which a vehicle has a function decoder, which controls:
With the addition of the common power supply, we get a total of four conductors.
It can be asymmetrical, that is to say not allowing a coach to be turned over in a train.
It must be suitable for NEM boxes, simple to manufacture and as little bulky as possible.
However, coupling will not be automatic if this poses too much difficulty. The fact that the targeted trains are non-deformable allows this functionality to be dispensed with.
Finally, it must be enough inexpensive.
Based on my experiences, I have retained the principle of a Ribu (Viessmann) type drawbar fitted with 1.27 mm pitch headers. I’ve already done this with such bars, or with home-made circuit boards. In the latter case, I only used single sided PCB, for convenience; machining swallow-tail shapes was tedious and time-consuming. The big news is to call on a service provider who can provide me with professional circuits, this time on a double-sided basis.
Here is the 3D drawing of this coupling.
The coupling combines two functions:
Note that the hooks are different between the two ends: when pulling on two coaches to uncouple them, the bar must always yield on the female connector side, otherwise there is a risk of tearing the wires on the male side. For this, the inclines are softer on the female side than on the other.
Thanks to the double-sided PCB, you can wire 2 to 4 wires at will. The wires are connected at the ends of the tails, therefore without any particular problem of passage. From experience, the female connector should preferably be always 4-point, otherwise it is too fragile, because it is cantilevered on its circuit. Even like that, you will have to handle the coupling gently…
The tail of the female part is extended by 1.5 mm, so that wires can be connected underneath, avoiding a conflict with the male part.
Soldering is facilitated by the fact that the areas to be soldered, at the end of the tails, are further apart than at the connectors: no risk of making solder bridges between contacts.
The PCBs, which I ordered from JLCPCB, apparently posed difficulties, as the manufacturing time was four days instead of the usual one. After an email exchange and some corrections, the drawings were accepted, but the result is not perfect. The outline shows excessive attacks of the cutter in the corners. This defect is not as apparent in all copies. However, we’ll see that the operation is almost correct.
In addition to the defect noted above, there is another one of which I am this time solely responsible, which is that the marks on the pads to be soldered are both too small and truncated. However, they are very useful in avoiding connection errors. I corrected this defect in the files by moving the texts on the large part of the circuits.
The 1.27 mm pitch connectors are available in several numbers of points from TME.
The thickness of the male and female connectors must be as equal as possible, to facilitate alignment. Thus, in the brand chosen (Connfly), the DS1031 male strips have a thickness of 2.5 mm, while there are two thicknesses of female cases: 2.4 (DS1065-07) and 1.8 (DS1065-01 ). It is therefore better to choose the DS1065-07 series which, moreover, has more robust pins.
Small problem: the male headers are not available at TME in angled form (DS1031-02 series). Apparently, they are not easy to find. We will see a parade in the following. Here is for all intents and purposes the manufacturer document (Data sheet) in PDF.