bkenobi:
Really? If the headers are shifted in/out, they won't line up with standard shields.
If they're offset, it just means the upper and lower headers on each side are offset sideways from each other. As long as both sides are offset in the same direction, the distance between them is the same and they'll match the standard Arduino pin layout. You're thinking they'd both be offset inwards or both outwards? I agree that wouldn't work, but that's not how it's done.
I missed half dozen posts somehow, weird. In any case, the way your shields were assembled you could certainly stack them The ones I bought were already assembled and were not offset like yours. If you don't mind the stack listing, that's a pretty good solution.
Yeah, actually it works quite well. In most cases if I go beyond a single shield I try to arrange them so one is offset right and the 2nd is offset left. That way it doesn't list so much. I have never needed more than 3 shields (touch wood), so the odd time where I'm stuck with two offset the same direction, it's only listing by .2 inches or so.
BillO:
This is an Ethernet shield with stackable headers on a Duemilanove. Horrible!
That is just either substandard manufacture, or bad design, or bad planning, or user abuse, and being stackable or offset has nothing to do with it.
@ bkenobi. I don't think the plastic block would help. They are about 2.5mm but the typical stackable headers have about 1mm extra clearance. And you can get stacking headers with even longer pins.
Nick_Pyner:
... or user abuse,...
Why thank you. I love you too.
Nick_Pyner:
That is just either substandard manufacture, or bad design, or bad planning, or user abuse, and being stackable or offset has nothing to do with it.
I think you're being too harsh. The problem as I see it is that the top of the USB socket is higher than the top of the female headers, so the shield won't seat properly if it extends as far as the USB socket. Using offset headers moves the male headers down which means the shield sits higher and if this is enough to avoid the problem then it's a good solution. It's not the offset that matters, but the extra height. Putting a spacer over the pins so that the headers don't engage so far would also be a reasonable solution. What's definitely NOT a good idea is leaving the shield resting on the USB socket and then let the far end be unsupported except for the bending stiffness of the pins. That's what is shown in the picture but I don't see that as abuse - it's a demonstration of the problem.
PeterH:
What's definitely NOT a good idea is leaving the shield resting on the USB socket and then let the far end be unsupported except for the bending stiffness of the pins. That's what is shown in the picture but I don't see that as abuse - it's a demonstration of the problem.
Exactly! I thought I was pretty clear but maybe I left too much to the imagination. Not everyone has a good one though, so here it is with a little more rigor.
The measurement from the bottom of the board to the top of the USB socket is 12.60mm
The measurement from the bottom of the boar to the top of the header is 10.16mm
The difference is 2.44mm. So if there is only 1mm of extra clearance with the typical stackable header, then the problem shown will exist. These measurements were taken from a new, out of the box today, un-abused, genuine Italian Arduino Uno R3.
BTW, also shown below is the measurement of the plastic block of a typical male header. It measures at 2.36mm which would go a long way toward eliminating the problem.
PeterH:
Nick_Pyner:
That is just either substandard manufacture, or bad design, or bad planning, or user abuse, and being stackable or offset has nothing to do with it.I think you're being too harsh. The problem as I see it is that the top of the USB socket is higher than the top of the female headers, That's what is shown in the picture but I don't see that as abuse - it's a demonstration of the problem.
Harsh? Me? Not for one moment. I merely state the bleeding obvious.
I see the height of the USB socket over the top of female headers is just trivea to be toyed with by those with limited imagination. The real problem is clearance over an RJ45 Erthernet socket.
If that photo attached is actually a case of squeezing a shield over a USB socket, then it is a fair example of bad design, bad planning, and user abuse but I utterly fail to see how this can be fixed by using offset headers. And the second “much, much nicer” picture demonstrates nothing, and absolutley nothing that cannot be done with stackable headers. Actually, what it might demonstrate is that you can fix the problem on a Duemilanove by moving the shield 0.15” sideways, but I doubt it, there is certainly no evidence to that effect, and I can’t be bothered to look any further.
Clearly, if this is to be fixed, it will be fixed by addressing the length of the pins rather than their position. So yes, a male pinrow to hand measures 11.25mm overall while the exposed length of pin on the standard stackable header measures 11.1mm, but I’m far from sure that 0.15mm is anything to write home to your mother about, and this presupposes that laboriously pulling the pins through the plastic block, to ensure that advantage taken from the extra length, is actually worthwhile. Trust me, I've done it, and it isn't.
The only thing that matters in the pictures, both, is the only thing that cannot be seen and of which there is no mention – the length of the pins. What we do see is a pathetic exercise that depends on scrounging plastic pinrow blocks, which might be better employed for their original purpose, and sloppy soldering, which may result in the pins not being properly bedded down.
Imagination has little to do with this but measurement has quite a lot to do with it. Indeed, an absurd preference for using male headers in a situation like that is more likely to be the cause of the problem than a solution. I say that simply because I am unaware of male header with extra long pins and such a predeliction may obscure the view of stackable headers that fill the bill.
The top picture below shows a storebought ethernet shield clearing the dreaded USB port with daylight to spare and with a homebrewed terminal board on top clearing the RJ45 socket. The pins simply go to the bottom of the socket, like God and Arduino intended
Those needing more rigorous demonstration may see the minimal merit of male pins over a stackable header evident in the second picture.
The third picture shows the comparative length in the only thing that counts – the amount of visible brass.
I think it fair to say that using male headers is likely to give you four things
all of which are more likely to be liabilities than assets.
So, in summary, sticking with male headers is a dumb exercise reserved only for those devoid of imagination – as I’m sure bkenobi has already worked out. There is no little irony in that the Adafruit board originally alluded to is available as an unassembled kit, thereby enabling the headers to be replaced without the angst the OP inevitably suffered. One can only conclude that the board was designed by somebody who actually knew what he was doing, sort of, only to find it assembled and marketed by morons.
I might point out, that another solution to the clearance problem is to remove the problem by using a truncated board, thereby enabling the use of standard headers. I have a few and they are quite sufficient.
Nick_Pyner:
...and I can’t be bothered to look any further.
Pfffft..!
The swan song of the permanently ignorant. Not that you are of course, but you are certainly singing the song.
To each their own.
DirtBiker:
Nick_Pyner:
...and I can’t be bothered to look any further.The swan song of the permanently ignorant.
Gee, are you telling me that, if I ride my bike down that particular road, I might find something? From casual glance, Duemilanove looks very much like most other Arduinos.
Nick_Pyner:
Harsh? Me? Not for one moment. I merely state the bleeding obvious.
I am not sure why this has you so angry. I expressed a personal preference and have not demanded that anyone else adopt it. I am entitled to prefer things a certain way, and equally entitled to express that in a public forum.
If that photo attached is actually a case of squeezing a shield over a USB socket, then it is a fair example of bad design, bad planning, and user abuse but I utterly fail to see how this can be fixed by using offset headers.
It is an all too common example of bad design. Both of those shots were commercially available Ethernet shields. One, using stackable headers (yes, they were too short), like many other offered for sale, does not work very well. The other, using offset headers, works well. That's how it can be fixed and your inability to see that is sad, but not really my concern. The "abuse" you speak of was merely for demonstration purposes. No actual shields were injured in making those pictures.
And the second “much, much nicer” picture demonstrates nothing, and absolutley nothing that cannot be done with stackable headers.
It quite obviously demonstrates the problem we saw with the first one was solved by the makers of the second board using offset headers. I did not claim it could not be fixed with longer stackable headers, did I?
Actually, what it might demonstrate is that you can fix the problem on a Duemilanove by moving the shield 0.15” sideways, but I doubt it, there is certainly no evidence to that effect, and I can’t be bothered to look any further.
Exactly the same thing is true of a Uno too.
Clearly, if this is to be fixed, it will be fixed by addressing the length of the pins rather than their position.
Clearly there is more than one way to do this. You prefer extra long, easily bent or broken stackable headers, I prefer offset headers that may indeed make the shields a little wider. Each solution has it's merits and each has it's drawbacks. Pick the one that works out best for you.
What we do see is a pathetic exercise that depends on scrounging plastic pinrow blocks, which might be better employed for their original purpose
What? Where did you see this? It was merely a suggestion by the OP. No one, as far as I know, has expended any real effort in scrounging. Certainly not enough for you to get your knickers in a knot and start insulting people you don't know.
and sloppy soldering, which may result in the pins not being properly bedded down.
Really? The Arduino folks do sloppy soldering? I actually don't see this in those pictures, and I can assure you that that particular Duemilanove has no sloppy soldering.
Indeed, an absurd preference for using male headers in a situation like that is more likely to be the cause of the problem than a solution. I say that simply because I am unaware of male header with extra long pins and such a predeliction may obscure the view of stackable headers that fill the bill.
Wow, you are really bent out of shape over this, huh? Sorry to here that, but I am entitled to my predilections and should be allowed them without being harangued over it. Your absolute hate of the concept of using offset headers seems to have consumed you and drawn you into near religious fervor over it. I think it's time to relax and leave others to their little predilections.
I think it fair to say that using male headers is likely to give you four things
- a wider board
- a substantial hit in wiring convenience
- one less row of holes
- twice as much soldering
all of which are more likely to be liabilities than assets.
Well, maybe two of the above (1 and 4), but I really do not see them as liabilities as they have never been in my experience.
So, in summary, sticking with male headers is a dumb exercise...
Well, that is your opinion and you are welcome to have it. I just don't share it and I might not be entirely alone.
I might point out, that another solution to the clearance problem is to remove the problem by using a truncated board, thereby enabling the use of standard headers. I have a few and they are quite sufficient.
There you go! Accepting alternate solutions and all. Atta boy! Maybe there is hope for you yet.
