Hey
I wanted to consult the wisdom of the crowd about the chamfering of a Pcbs connector edges. I typically use the 30degree bevel because it’s a longer ramp and a sharper leading edge so in my mind, this is better protection of the female connector than a 45degree bevel. Is my thinking accurate or is there a better reason to use 45degree ?
I’ll include a few pictures for reference….. though I doubt people that have made Pcbs will need them.
Thanks!
Pcb beveled edges
Moderator: B00daW
Forum rules
1. NO BLATANT PIRACY. This includes reproducing homebrew less than 10 years old, with the exception of free software.
2. No advertising your reproductions, with the exception of free software.
3. Be nice. See RFC 1855 if you aren't sure what this means.
1. NO BLATANT PIRACY. This includes reproducing homebrew less than 10 years old, with the exception of free software.
2. No advertising your reproductions, with the exception of free software.
3. Be nice. See RFC 1855 if you aren't sure what this means.
-
Markfrizb
- Posts: 631
- Joined: Sun Dec 02, 2012 8:17 am
- Location: East Texas
Pcb beveled edges
You do not have the required permissions to view the files attached to this post.
-
segaloco
- Posts: 913
- Joined: Fri Aug 25, 2023 11:56 am
Re: Pcb beveled edges
If I had to wager a guess, it may have something to do with the expected insertion velocity. In my mind, if you have a sharper slope of the edge connector, the bump from insertion on the individual pins is more likely to result in just a little more give and flexibility vs a smoother slope. As you described, this could then result in a tad bit more wear-per-insert on the mating pins in the socket.
However, a factor here may be the thickness of the gold fingers. If they've got a noticeable "edge" on the beveled end of the connector, then a smooth slope means less overall outwards force, meaning once the tip of the gold finger hits it, there is less force pushing outwards towards the edges of the socket, more pushing forward towards the tips of the gold fingers.
A parallel may be in the concept of a "bump key". These are special keys used to defeat conventional pin-based locks. They have a series of bumps at sufficient height and angles to strike the bottoms of the pins of a lock, transferring the motion through the lower pin to the upper pin, bumping it into place. Couple with torsion on the key and the pins will stick against the chamber and the lock is defeated. In this case, too smooth of a slope on a bump and you're transferring more of the motion parallel to the insert of the key rather than perpendicular, so it "rattles" the pin in its slot rather than transferring motion perpendicular upwards into the shaft. Too sharp of a slope and you get a lot of physical resistance and wear on the pins you wouldn't otherwise.
So yeah, this may become a factor with quick (or not so quick) inserts, one or the other slope may introduce more laminar stress on the gold fingers, especially at the tip, if they're not sufficiently controlled to a certain thinness. Granted, it is very possible with current planing technologies this is no longer a concern, but I could see those sorts of micro-variations mattering quite heavily in the 50s and 60s.
Edit: I realize I didn't make the velocity connection. Like the bump key, the velocity of inserting the edge connector does have a minute effect on how much force is being directed towards the connectors and at what angles. Couple this with the slope of the edge connector and you could probably derive a function for how much "bump" is given to a pin based on the velocity and angle of the edge contacting it.
However, a factor here may be the thickness of the gold fingers. If they've got a noticeable "edge" on the beveled end of the connector, then a smooth slope means less overall outwards force, meaning once the tip of the gold finger hits it, there is less force pushing outwards towards the edges of the socket, more pushing forward towards the tips of the gold fingers.
A parallel may be in the concept of a "bump key". These are special keys used to defeat conventional pin-based locks. They have a series of bumps at sufficient height and angles to strike the bottoms of the pins of a lock, transferring the motion through the lower pin to the upper pin, bumping it into place. Couple with torsion on the key and the pins will stick against the chamber and the lock is defeated. In this case, too smooth of a slope on a bump and you're transferring more of the motion parallel to the insert of the key rather than perpendicular, so it "rattles" the pin in its slot rather than transferring motion perpendicular upwards into the shaft. Too sharp of a slope and you get a lot of physical resistance and wear on the pins you wouldn't otherwise.
So yeah, this may become a factor with quick (or not so quick) inserts, one or the other slope may introduce more laminar stress on the gold fingers, especially at the tip, if they're not sufficiently controlled to a certain thinness. Granted, it is very possible with current planing technologies this is no longer a concern, but I could see those sorts of micro-variations mattering quite heavily in the 50s and 60s.
Edit: I realize I didn't make the velocity connection. Like the bump key, the velocity of inserting the edge connector does have a minute effect on how much force is being directed towards the connectors and at what angles. Couple this with the slope of the edge connector and you could probably derive a function for how much "bump" is given to a pin based on the velocity and angle of the edge contacting it.
-
krzysiobal
- Posts: 1221
- Joined: Sun Jun 12, 2011 12:06 pm
- Location: Poland
Re: Pcb beveled edges
From my little experience with different PCB factories (China's pcbway and then jlcpcb) I can san that chamfers are quite random and not something you can expect a repeatable quality.
First of all, they don't make chamfers if you choose the "cheap" prototyping project. (PROTOTYPING = 2$ for 5pieces or 5$ for 10 pieces AND your project is smaller than 10x10cm and quantity is 5 or 10). So you need to make bigger order something larger.
Second, despite dwo distinct options of chamfering (30* and 45* degree) in the ordering process I remember when I placed once a 45 degree option, I was asked by the staff if 30 can be because they dont make 45
Third, the real length of chamfer is really random. One time you place order you get nice long chamfer, sometimes it is short or even missing and then you need to prove that you`re not an elephant to get some refund or getting the pcbs to be made again for free
Or if you are extra unlucky, you can get PCBs with burrs othe edge that is worse than even no chamfer and you need to dremel the edge on your own to remove it.
First of all, they don't make chamfers if you choose the "cheap" prototyping project. (PROTOTYPING = 2$ for 5pieces or 5$ for 10 pieces AND your project is smaller than 10x10cm and quantity is 5 or 10). So you need to make bigger order something larger.
Second, despite dwo distinct options of chamfering (30* and 45* degree) in the ordering process I remember when I placed once a 45 degree option, I was asked by the staff if 30 can be because they dont make 45
Third, the real length of chamfer is really random. One time you place order you get nice long chamfer, sometimes it is short or even missing and then you need to prove that you`re not an elephant to get some refund or getting the pcbs to be made again for free
Or if you are extra unlucky, you can get PCBs with burrs othe edge that is worse than even no chamfer and you need to dremel the edge on your own to remove it.
My website: 
My NES/FC flashcart: -
segaloco
- Posts: 913
- Joined: Fri Aug 25, 2023 11:56 am
Re: Pcb beveled edges
Yikes...it didn't occur to me the QA had gotten so bad but that makes sense. One other factor that did occur to me is if the slope is more gradual, you either need a longer taper or it can't taper to as thin of an edge. This does have a minute effect at the small scale on how rigid that edge is. Could be a factor in wear over time.