How many spikes does it take to build a wall?

DaveC426913

Gold Member
17,935
1,567
When I first started this, I thought I'd need only one spike to stitch the 6x6s together and to fix it to the ground.

But OK, I need two after all.

Having finished two, I realized it's still not secure!

How did I end up needing three spikes to stitch one section of wall? What did I do wrong??

spikes.jpg
 
14,952
4,612
You lapped the upper and lower sections identically. If you had staggered them successively, you could have used just 2 spikes

244497


You don't want to line them up identically, because then you have a pivot point(*) where they line up. The way I've shown it, there is no single pivot point. It's not a LOT better than lining them up but it IS somewhat better.

* or what you might think of as a push point
 
Last edited:

Tom.G

Science Advisor
2,544
1,368
[size=small]added in edit.[/size]
How did I end up needing three spikes to stitch one section of wall?

Or worded differently: You needed to immobilize the ends of each timber.
 
Last edited:
14,952
4,612
Or worded differently: You need to immobilize the ends of each timber.
No, that's too vague. His technique did that, but he used 3 spikes. To really immobilize the ends of each timber, you'd need 4 spikes. Dave's goal was to reduce it to 2 spikes.
 

JBA

1,285
312
Three spikes are required if he wants to insure that the wall sections act as a continuous structure.
 
14,952
4,612
Three spikes are required if he wants to insure that the wall sections act as a continuous structure.
Seems to me it would be 4. How do you get 3?
 
32,680
8,551
The three spikes fix the relative position of all bars along the direction of the wall.

Proof that you need three as soon as you have two layers and if you cannot use the ground: Layer 1 has two elements, they both need a spike through them. To fix their relative position these two need to go through the same element in a different layer. But then the second element in that layer doesn’t have a spike going through it - we need one more spike.
 

DaveC426913

Gold Member
17,935
1,567
You lapped the upper and lower sections identically. If you had staggered them successively, you could have used just 2 spikes

View attachment 244497

You don't want to line them up identically, because then you have a pivot point(*) where they line up. The way I've shown it, there is no single pivot point. It's not a LOT better than lining them up but it IS somewhat better.

* or what you might think of as a push point
Sure, but I could have done the same two spikes with my staggered arrangement: #1 and #3 would have immobilized the ends, just like yours:
spikes2.png



Thing is: both your and my 2-spike implementation have the same flaw: the two sides are not stitched to each other - only to the ground. If you pushed inward on one side or the other, the stitch would still shear.


EDIT: OK, I see you qualified it, recognizing that there is still some give in your joint, just not as much as in mine.
 

DaveC426913

Gold Member
17,935
1,567
Seems to me it would be 4. How do you get 3?
I see 3 as well. How do you get 4?
In my initial diagram, all ends are secure.
 

DaveC426913

Gold Member
17,935
1,567
The design I used on the original wall only required one spike. I was making curved wall, so a single pivot was desirable.
spikes3.png

This time the walls are straight. I made what I thought was a minor change without fully considering the consequences.
 
14,952
4,612
I see 3 as well. How do you get 4?
In my initial diagram, all ends are secure.
Yeah, but your upper and lower on the right can pivot around the spike. You'd need another spike over there to make it such that every beam has 2 spikes, so that it cannot pivot.
 

DaveC426913

Gold Member
17,935
1,567
Yeah, but your upper and lower on the right can pivot around the spike. You'd need another spike over there to make it such that every beam has 2 spikes, so that it cannot pivot.
But they cannot pivot since they're secure at the other end.
 
14,952
4,612
But they cannot pivot since they're secure at the other end.
Ah, well in that case two spikes should be fine as in your post #8. I had not realized that you needed them stitch to each other, thus my post #2. With that caveat, your post #10 shows a good solution except for the obvious extra work of making the lap joints.
 

DaveC426913

Gold Member
17,935
1,567
I had not realized that you needed them stitch to each other
One winter of ground heave would put them out of true if they weren't secure.

your post #10 shows a good solution except for the obvious extra work of making the lap joints.
And that it's much less pretty.
 

Baluncore

Science Advisor
6,408
1,917
You appear to be making chains of links.
You seem to require the ends of all bars to be pinned.

The left pin correlates all the bars of the left link.
The right pin correlates all the bars of the right link.
The middle pin is a hinge pin that joins the interlocked fingers of two links.

Avoid the situation where finger joints are irregular and not alternately aligned.
 
331
137
Why not over lap more? Cut one of the 6x6 in half, go from there.

244590
 

Attachments

russ_watters

Mentor
18,503
4,715
A couple of years ago I built a similar wall, on a slope. The slope enables almost every length to be a single timber, with one pin at each end, progressively stepping down.

But at the bottom there is a corner, which if I remember correctly matches your 3-pin configuration.
 

Want to reply to this thread?

"How many spikes does it take to build a wall?" You must log in or register to reply here.

Physics Forums Values

We Value Quality
• Topics based on mainstream science
• Proper English grammar and spelling
We Value Civility
• Positive and compassionate attitudes
• Patience while debating
We Value Productivity
• Disciplined to remain on-topic
• Recognition of own weaknesses
• Solo and co-op problem solving

Hot Threads

Top