Securing Your Seed Phrase with Washers

[HideYourKeys]

The only point I do not understand in your Jig is how the washer is fixed in the correct position and what prevents it from moving after each letter is stamped.

From Stamping JIg Instructions Manual:

Once the washer is inserted, rotate the BitJig to the lock position, securing it firmly, note that the washer holder inside the base moves down

I assume there's some thread inside to "clamp" the washer in place, but I can't envision how that works since the washer fits through the hole of the base that moves down.

The only point I do not understand in your Jig is how the washer is fixed in the correct position and what prevents it from moving after each letter is stamped.

From Stamping JIg Instructions Manual:

Once the washer is inserted, rotate the BitJig to the lock position, securing it firmly, note that the washer holder inside the base moves down

I assume there's some thread inside to "clamp" the washer in place, but I can't envision how that works since the washer fits through the hole of the base that moves down.

Exactly.
Still, I don't understand how it is clamped, avoiding unwarranted movements.
I don't know if it is an industrial secret,  but it would be nice to know more.
I guess this is one of the most critical aspects for a good jig.

As LoyceV said, there is a threaded piece inside, which allows you to secure the washer. That piece has an inner diameter, which is slightly smaller than the washer's diameter, so it locks the washer when the jig is fully rotated to the lock position. In order to introduce the washer (which has a larger diameter as we already know), it is possible because this threaded piece which "clamps" the washer, has several lateral "gaps" which provide a slightly larger diameter than the washer. Therefore it can be introduced if the washer is rotated vertically and introduced through the gaps when the jig is at the fully "unlocked" position (pictures on pages 4-5 of the manual might help you visualize it). Tolerances are quite tight though, there were a few iterations before it was good enough...

We got a review last week from a substacker who wanted to test it, you can check the 4th image, which shows it better (it is in Spanish though, but you can translate it)

https://cryptosecurity.substack.com/p/hide-your-keys

[n0nce]

[...]
Looking forward to your thoughts and ideas!

Looks really good, you did a great job with the modeling and I love your innovation that removes the need of an anvil!

I'd like to print the STL and try it for myself, to be honest, since I made my own open-source, customizable steel washer backup jig back in 2021:
n0nce's Steel Washer Backup jig (customisable)

It is much simpler than what you made, but people can get an STL for any washer size (metric, imperial, any size, any thickness).

[HideYourKeys]

[...]
Looking forward to your thoughts and ideas!

Looks really good, you did a great job with the modeling and I love your innovation that removes the need of an anvil!

I'd like to print the STL and try it for myself, to be honest, since I made my own open-source, customizable steel washer backup jig back in 2021:
n0nce's Steel Washer Backup jig (customisable)

It is much simpler than what you made, but people can get an STL for any washer size (metric, imperial, any size, any thickness).

Thanks for your feedback! I would love to give you the .STL files so you can try it, but that would be a risk as the files might spread all over the internet. Let me explain my thoughts:

The target is to get enough visibility, scale the project and if this jig "gets known" after some time, sell the mechanized one in CNC at a reasonable cost, and then release the .STL files for the basic version (therefore the official one in metal could be purchased, and the one in plastic could be made with a 3D printer + buying a steel disc a certain diameter and thickness, which are quite affordable). This would be a similar approach with blockstream Jade for instance, which is a HW that can be purchased officially, but the firmware is OS and you can flash it on many ESP32 boards.

In the meantime... if you live in Europe I can send one to you for free, so you can test it    (you do not have to give a real name, and it could be sent to an InPost locker, so no real address is needed). At the bottom of the webpage you can see the TG, mail, etc... to contact ( https://hideyourkeys.io/ ), or send me a DM here

[HideYourKeys]

I made a video, so the curious ones can check the results, and the whole process:

https://www.youtube.com/watch?v=SWVvvpQr304

Notice that I only struck each letter once, and didn't apply much force, for other washers I have tested the letters and numbers are engraved at a deeper level (if you hit them harder, of course)

[fillippone]

I made a video, so the curious ones can check the results, and the whole process:

https://www.youtube.com/watch?v=SWVvvpQr304

Notice that I only struck each letter once, and didn't apply much force, for other washers I have tested the letters and numbers are engraved at a deeper level (if you hit them harder, of course)

Wow very nice.
That is a serious jig.
I like the mechanism that holds the washer in place. I had to apply a scotch on the bottom of the jig, but you improved this by far, as the washers apparently aren't moving.
Well done.

[apogio]

Notice that I only struck each letter once, and didn't apply much force, for other washers I have tested the letters and numbers are engraved at a deeper level (if you hit them harder, of course)

Very good result! Congrats!
I liked the fact that you didn't have to use an anvil.

[satscraper]

Still I would advocate washers holding binary code of  numerical order of relevant word in BIP39 list. Such washer would look like this.

(a washer containing the word "sausage" -  decimal  numerical order is  1534, in binary representation - 10111111110. The numbers near the edge of the washer are placed in the figure only for guidance in the position of the bits. The vertical is the marking reference.)

The benefits - such washers are easy to stamp ( as the single  center punch is needed instead of multiple bits with letters), and their purpose are not so obvious for others.

[LoyceV]

The benefits - such washers are easy to stamp ( as the single  center punch is needed instead of multiple bits with letters), and their purpose are not so obvious for others.

The drawback: you lose the "error correction" you get when using words. A binary "typo" won't be very obvious to detect.

[satscraper]

The benefits - such washers are easy to stamp ( as the single  center punch is needed instead of multiple bits with letters), and their purpose are not so obvious for others.

The drawback: you lose the "error correction" you get when using words.

Nope, the order of washers  (i.e the order of words) can be marked by daps on washers' ribs. BIP 39 "error correction" is the responsibility of  last word, thus such function is preserved as long as  the relevant washer is  on the bolt .

A binary "typo" won't be very obvious to detect.

 I have such washers, they are very easy to make without any "typo" as well as to decrypt. But sure, this method is not  for clumsy hands, i.e not for those, who can make "typo" at very simple doing.

Check twice!

[LoyceV]

Nope, the order of washers  (i.e the order of words) can be marked by daps on washers' ribs. BIP 39 "error correction" is the responsibility of  last word, thus such function is preserved as long as  the relevant washer is  on the bolt .

I meant the error correction you get with words: if you misspell abandon, it's still pretty clear you mean abandon. But if you misspell 10111111110, there's no way of knowing other than carefully checking.

this method is not  for clumsy hands, i.e not for those, who can make "typo" at very simple doing.

I've seen topics from people who lost their crypto because they made multiple mistakes writing down private keys. But you're right: it can be done. At least try to recover your seed from scratch before funding (which is good practice to do with any backup anyway).

[fillippone]

Still I would advocate washers holding binary code of  numerical order of relevant word in BIP39 list. Such washer would look like this.

(a washer containing the word "sausage" -  decimal  numerical order is  1534, in binary representation - 10111111110. The numbers near the edge of the washer are placed in the figure only for guidance in the position of the bits. The vertical is the marking reference.)

The benefits - such washers are easy to stamp ( as the single  center punch is needed instead of multiple bits with letters), and their purpose are not so obvious for others.

Don’t be your own enemy.
One key benefit of such solution is the fact they can be easily interpreted by others.
KISS principle is the reason why we are stamping washers in the first place.
Stamping binary only raises the bar for securing the generational wealth.
Just remember this is not for short term self custody, but for long term inheritance. The person receiving the washers must be able to understand what they are.

[satscraper]

Don’t be your own enemy.

I will not be. I trust my own gut.   I take to binary like a duck to water.

One key benefit of such solution is the fact they can be easily interpreted by others.
 

In my view this is disadvantage rather than benefit.  

Just remember this is not for short term self custody, but for long term inheritance. The person receiving the washers must be able to understand what they are.

Sure, I don't think that  mentally slow generation is what is going to happen next  

[fillippone]

In my view this is disadvantage rather than benefit.  

Part of why I stamped my washers is that if something happens to me, my relevant one(s) can retrieve my precious sats.
Of course, this involves some trust.
But I sleep better thinking that the precious satoshis aren't going to be gifted to all of you as added disinflation.
Also, I sleep better knowing that my trust doesn't steal the funds from me (yet, I do a yearly on-chain and off-chain check).
Sorry if I am selfish toward my loved ones.

[HideYourKeys]

I have been thinking about launching a free jig for people who have 3D printers and already have metal stamps, as an additional resource.

It won't be as complex as the Bitjig, and it won't require additional parts (the Bitjig has a steel disc, screws, etc.). For this case, an anvil would be needed, but the design seems much easier to grab than the Blockmit (I've only made the design for 24mm washers, but I want to make it for 18mm washers as well). It's a preliminary design, but maybe we could consider integrating the anvil by purchasing a steel disc. For now, this is how it stands.

I would like to know if you guys like the design, here is a render:

[HideYourKeys]

I also printed a quick prototype, to check if everything fits.

[m2017]

It's a preliminary design, but maybe we could consider integrating the anvil by purchasing a steel disc.

Ideally, the lower part should be made entirely of metal (in principle, if you have the tools and skills, not necessarily yours, but the craftsmen’s, you can carve it from an iron blank or melt a mold from fusible metals, such as tin or lead^*, if ot don't turn out to be too soft to impacts hammer), then an anvil is not necessary.

^* - can melt a mold even from non-ferrous metals (brass, bronze), but it will be more difficult due to high melting temperatures.

Have you already tested this wonderful device? Will the lower part under the puck begin to collapse from impacts? Keep in mind that stainless steel washers (as opposed to aluminum) have a fairly high hardness (approximately 353 MPa) and the applied force must be at a high level in order for the text to print well.

[HideYourKeys]

It's a preliminary design, but maybe we could consider integrating the anvil by purchasing a steel disc.

Ideally, the lower part should be made entirely of metal (in principle, if you have the tools and skills, not necessarily yours, but the craftsmen’s, you can carve it from an iron blank or melt a mold from fusible metals, such as tin or lead^*, if ot don't turn out to be too soft to impacts hammer), then an anvil is not necessary.

^* - can melt a mold even from non-ferrous metals (brass, bronze), but it will be more difficult due to high melting temperatures.

Have you already tested this wonderful device? Will the lower part under the puck begin to collapse from impacts? Keep in mind that stainless steel washers (as opposed to aluminum) have a fairly high hardness (approximately 353 MPa) and the applied force must be at a high level in order for the text to print well.

Thanks for the feedback mate, you are totally right about the hard surface at the base, in fact I had quite a few problems with this. Currently, I am working on two options.

The first one, which I called BitJig, has a steel disc glued inside, and also a nut on the base, so all contact is metal-metal-metal (washer-disc-nut). The first prototypes did break, but I made many improvements over time, and now they don't. You can check the other video I posted: https://www.youtube.com/watch?v=SWVvvpQr304

The first prototypes did break because I was placing them on a relatively "soft" surface, like a carpet, or things like that, to reduce the noise, but that should not be done at all! because the base might break. The latest units work quite well, I have reused some of them for around 10 seedphrases sets (I gift many to friends and family), and they do not show any sign of rupture. I do place them on top of a hard surface when engraving the words, such as a concrete or ceramic floor (even on top of a hard kitchen plastic board). Note that in the video, I only hit each letter once, and the jig was on top of a blue mat which is relatively soft (but it looked great on camera   ), when engraving the words on top of a hard surface, the results are better. Also hitting each letter hard twice engraves them deeper.

These would be the BitJIg and BitJig mini renders (there are some pics in the previous pages):



The second approach, which is the one I am working on, is to release a free model (the one from these recent pictures). This jig would be easier to grab than the blockmit (the original one you can find online, it is not bad, but I did not like it much when I used it, so I designed my own ones), and easier to stamp, but less complex, so anybody with a 3D-printer could try it. With regards to this one, I was thinking about gluing a thick steel disc at the base, of 5cm diameter approximately, but I am not sure if that extra complexity is worth it, the blockmit is used with an anvil so I guess it might not be necessary... This one cannot fix the washer as the BitJig does, as it would lack the "grabbing mechanism".

Also in the future I might substitute the BitJig base by a CNC'd one, I have some prototypes already, but we'll see, step by step  

[m2017]

The first one, which I called BitJig, has a steel disc glued inside, and also a nut on the base, so all contact is metal-metal-metal (washer-disc-nut). The first prototypes did break, but I made many improvements over time, and now they don't.

What if add a round piece of rubber (you will need to select the appropriate thickness and elasticity) between the metal base and the washer (between metal and metal)? To reduce the ringing of metal on metal and reduce the load on the base due to the elasticity of rubber and the ability to absorb the kinetic energy of a hammer blow.

The first prototypes did break because I was placing them on a relatively "soft" surface, like a carpet, or things like that, to reduce the noise, but that should not be done at all! because the base might break.

It is better to place it on a hard surface, such as a concrete floor or thick wooden surfaces (for example, a block), but place a thick piece of rubber (should not be soft) between the Jig and the surface, which will avoid damaging the surface and creates less noise, and also absorbs energy upon impact that would be transferred to the surface.

I do place them on top of a hard surface when engraving the words, such as a concrete or ceramic floor (even on top of a hard kitchen plastic board).

In my opinion, the best surface option for placing the ig is wood (a block - due to the structure of wood, which has its elasticity and sufficient hardness), on the bottom of which a piece of non-elastic rubber will be glued, which will allow to place it on any surface without damaging it or creating noise, as if you had a team of construction workers.

Also hitting each letter hard twice engraves them deeper.

You meant to say a little deeper, right? Because stainless steel is not easy to engrave with stamps (thin and not deep lines of letters). If turn on the paranoid , then a thin and shallow font is more easily destroyed and loses readability

The tip of the engraver/stamp has a “\/” shape (clearly visible in the figure), but due to the hardness of the washer it cannot penetrate deeper into the metal and make the text better readable. Therefore, you can use a file/grinding stone and sharpen the ends of the engravers to make them a little wider (“\_/”), which will allow you to make characters with wider lines.

Nuances:
You will need a 2nd set of stamps and use the 2nd set after applying the 1st set. First, a garver with a thin end - the 1st blow ("\/") is used, and for the second blow it is used with a wider one ("\_/").

[LoyceV]

What if add a round piece of rubber (you will need to select the appropriate thickness and elasticity) between the metal base and the washer (between metal and metal)? To reduce the ringing of metal on metal and reduce the load on the base due to the elasticity of rubber and the ability to absorb the kinetic energy of a hammer blow.

The rubber will indeed absorb part of the kinetic energy from the hammer, which means there's less energy remaining to stamp a letter into the washer. The harder the surface under the washer, the better you can stamp.

[HideYourKeys]

What if add a round piece of rubber (you will need to select the appropriate thickness and elasticity) between the metal base and the washer (between metal and metal)? To reduce the ringing of metal on metal and reduce the load on the base due to the elasticity of rubber and the ability to absorb the kinetic energy of a hammer blow.

I tried this, in fact I created some TPU discs and glued them inside the BitJig, but the results were worse, that was my first approach and I switched to the "all-hard-materials" approach

What if add a round piece of rubber (you will need to select the appropriate thickness and elasticity) between the metal base and the washer (between metal and metal)? To reduce the ringing of metal on metal and reduce the load on the base due to the elasticity of rubber and the ability to absorb the kinetic energy of a hammer blow.

The rubber will indeed absorb part of the kinetic energy from the hammer, which means there's less energy remaining to stamp a letter into the washer. The harder the surface under the washer, the better you can stamp.

It is JUST like this, the harder the surface, the better results you get

In my opinion, the best surface option for placing the ig is wood (a block - due to the structure of wood, which has its elasticity and sufficient hardness), on the bottom of which a piece of non-elastic rubber will be glued, which will allow to place it on any surface without damaging it or creating noise, as if you had a team of construction workers.

Wood is not bad, I am currently using a 4mm-thick small metal plate that I had. I place the BitJig on top of that, and then I place everything on the floor, or a plastic board (I could use the floor directly, but just to make sure that the nut does not make small marks on the floor). I have also tried without the plate, and the results are quite good as well (in fact that plate is ready to be thrown away, it's just scrap I had, I still use it so I don't mark the plastic boards).

The tip of the engraver/stamp has a “\/” shape (clearly visible in the figure), but due to the hardness of the washer it cannot penetrate deeper into the metal and make the text better readable. Therefore, you can use a file/grinding stone and sharpen the ends of the engravers to make them a little wider (“\_/”), which will allow you to make characters with wider lines.

Nuances:
You will need a 2nd set of stamps and use the 2nd set after applying the 1st set. First, a garver with a thin end - the 1st blow ("\/") is used, and for the second blow it is used with a wider one ("\_/").

I'd say the results are "good enough" with one set of stamps, the words have to be readable, as long as you can read them the do not have to be super deeply engraved (imho). Just like in the video, or maybe a bit more, but as long as they are readable I am fine with it.