1 Like
Sorry, but I just can’t stop fiddling with this. Just ordered 5 pcs where I’m experimenting with larger pin-holes (Ø1.4). The idea is to mount the MCU-board (pill) using Swiss Machine Pin-headers. So the spacing between pcb’s can be reduced to 3 mm (same as height of usb).
1 Like
What do you plan to use on the pill? Those pins normally also have a spacer to hold the pins together and keep clearance. Is that included in the 3mm? The square pins you used for the 2.54mm connectors might be too thick to fit the swiss maschine pin headers on the power board.
Yes, the spacer is 3 mm, and then I made the pinholes 1.4 mm - same as the thickness of the widest part of the pins. So the spacing between the boards will be kept by the plastic spacers, at 3 mm. Don’t know if it’s a good idea, but it’s the only way I see a solution where the “pill” can be attached to the foc-board through a push on/pull off connector, and still not exceed the “height” of the mosfets, which is 5mm.
3 mm (spacer) + 1.7 mm (pcb) < 5 mm (mosfet)
Which means that an aluminum heat-sink plate with some goodwill and insulation can cover the whole bottom part of the pcb.
1 Like
This also means that the 2.54 jst connectors on the top side have to go… So the only remaining jst connectors will be the 6-pin HALL and 2-pin CAN, which both will be at 2.0 mm pitch.
Those inputs and outputs where there isn’t a 2.54 jst connector any longer, we can access by making a connector out of…? Swiss machine pinheaders?
3 Likes
It’s just an idea that I want to try out.
Shambler, love ya! Manager of Test Department 
3 Likes
I use these capacitors:
You don’t have to use the same. Two 330uF might also do, or what you have available.
I tried to use the jst connectors themselves for connecting some of the pinholes of the boards. I managed to screw-up one board because of this. Still my advice would be to use standard pin headers for ALL pinholes, and still do all tests to catch any errors early.
Also after you have attached pinheaders to the MCU Pill, check if everything still works. Don’t cut the pins with pliers too far down. Leave as much of the pins so you’re still able to pull them out if you need to detach the pill again (if something doesn’t work). And by the way, I’m an idiot at soldering.
I guess you can see why I’m trying to find a push-on/pull-off-connector-like solution for all of this. Hope female machine-pins will do the trick.
I soldered the pos,neg power wires before soldering on the capacitors. Think this is the easiest way. Top side is negative, bottom is positive.
AND, the circuit for the on/off push-button can have a life of it’s own. The first time you connect it to a power supply or a battery, you might need to push the button for a whole minute before it turns off. This is just the first time. After that it only takes 4-5 seconds.
If you DON’T have a push-button, you can by-pass the circuit simply by bridging the exposed pads (which are there for that exact purpose).
1 Like
40 pin IC sockets. Why haven’t I considered that an option?
1 Like
Even better option - with nice long legs 
Basically, a female machine pinheader
A thats a great idea. With this headers. I will get some to test. For the caps im using some with 100V
Sounds like streetface under heavy vibration. I don’t think all the connectors and headers is a good idea. Too many things to vibrate apart.
1 Like
I agree, and that is why it is always an option to just solder the parts together. Connectors just make it a lot easier to locate stuff that doesn’t work, and you will only need to replace the part with such stuff. The struggle with jlcpcb could change things though.
Why are they called “machine” pinheaders. Because they are “machined”? Or because they are made for vibrating machines?
Anyone from Switzerland?
1 Like
Even soldered headers can fail under vibration. If I recall, a commercial ESC has already failed because of this.
That must have been a cold solder joint, else all components will also fall off. Eventually, they will under extreme circumstances.