Hackberry portal Forum HACKberry Q&A Socket Fit

This topic contains 27 replies, has 8 voices, and was last updated by  Hiroshi Yamaura 4 years, 2 months ago.

Viewing 28 posts - 1 through 28 (of 28 total)
  • Author
  • #287

    Brian Little


    I’m a traditional prosthetist who’s super interested in this project. I was wondering how you guys are getting the sockets to fit, what the skin interface is — and how you get the electrodes to stay in place and configure them. How do you train your users?


    Hiroshi Yamaura

    Hi Brian,
    Sorry for not contacting you.

    Previously I emailed you, we use velcro band and elbow supporter.
    Although we occasionally use 3D scanner and refer the stump 3d data in CAD, we are searching the method that doesn’t require the measurement. Using velcro and elbow supporter is a example of the method.

    On the other hand, we are examining the possibility to combine HACKberry and traditional socket.
    If we measure the shape of attachment and add the shape to HACKberry, traditional prosthesis user will be able to use HACKberry as another attachment.

    As for sensor, we don’t use electrode. We are using photo-reflector as a kind of pressure sensor and it is fixed by velcro band.
    Please check this post. This sensor is very easy to use. It doesn’t require the training.



    Hi Hiroshi ,

    Thanks for this Excellent Bionic hand, You have given Hope to many amputee,s in the world, And I am hoping my friend, who has 1 thumb and stumps of fingers left on his right hand, can benefit from your Invention.
    Im only a novice, but would just fit the above and have a longer Arm length with hole for thumb?

    Thanks for any Idea s


    Hiroshi Yamaura

    Hi Andrew,
    I recommend you to take part in the reddit prosthesis project.

    The purpose of this project is to provide the reddit user Ryan with the bionic hand like i-Limb digit by refining the HACKberry. I’m not sure but Ryan’s situation seems to be very similar to your friend.
    The project organizer Cameron is our great collaborator.
    I will also support this project as much as possible.

    Why don’t you post the photo of your friend’s right hand to the project’s community?
    After that, you can discuss how to collaborate with them.


    Hiroshi Yamaura

    I’m still wondering whether we should use velcro type socket or other type socket.
    I’d like to share you how to make socket from 3D scanned data.

    1.Download Meshmixer
    We use Autodesk Meshmixer. Please download from the following link.

    2.Scan a stamp
    Get the 3D data of the stamp using 3D scanner.
    I recommend Sense [http://www.amazon.com/dp/B00HLJFMMO/].
    It is much cheaper than other scanners but has sufficient ability.
    In this document, I use Ryan’s 3D data.
    We can download the data from Wevolver[https://www.wevolver.com/cameron.norris/ryan-cashman-partial-hand-prosthesis/medical/files/].

    3.Open the 3D scan data from Meshmixer.

    4.Reduce triangles
    The data contains too much triangles.
    -double click the surface
    -change “Pecentage” to “Triangle Budget”
    -“Tri Count” 100000

    5.Correct the scale
    The data is too small.
    It is maybe due to unit conversion.

    -“Scale X”(,Y or Z) 25.4

    6.Set the clearance
    Make the clearance between Ryan’s skin and innur surface of the socket.

    -double click the surface
    -change “Constant” to “Nomal”
    -“Offset” 2mm

    7.Convert to shell

    -drag and drop the transparent appearance

    -double click the outer surface

    -select the innur surface

    -double click the outer surface
    -“Direction” “Normal”
    -“Offset” 3mm

    8.Cut the bottom
    -“Plane Cut”
    -move the plane by dragging an arrow

    9.Export the file

    That’s all.
    You can use this data to convine the HACKberry.

    The data can be downloaded from the following link.

    Please give it a try.


    Cameron Norris

    Hiroshi, thank you for the kind words and for this helpful tutorial.
    I recently provided a company who 3D print medical-grade silicone with the socket data you created for Ryan.
    If the print and fitting is successful and affordable, it could be a great way to produce attachments for more ‘difficult’ injuries. I will also be experimenting with printing a flexible sheath for the liner from NinjaFlex. The idea is for the sheath to fit around the soft silicone liner so that the limb can be encased in a rigid shell to increase load-bearing without discomfort.

    Also, you’re absolutely right about Ryan’s situation and the description of Andrew’s friend.
    Andrew, I’ve attached a photo of Ryan’s 3D scan for you to take a look at.
    Just like Hiroshi says, you should join the project to see how we can help your friend.

    Ryan Cashman Partial Hand


    Hiroshi Yamaura

    3D printing silicon liner and a flexible sheath and combining them is a very good idea.
    I’m looking forward to hear your progress.


    Brian Little


    I have to say — that is amazing.

    Myoelectric protheses to date are plagued with activation problems. I suspect the EMG based electrode systems have been used so far so that you can call the device “mind controlled”

    This optical sensor seems to be much more robust and easier to use.

    Is the output directly proportional to the compression of the sensor or can you change the grip strength of the hand for users who have weak muscles?


    Hiroshi Yamaura

    Thanks Brian.

    Is the output directly proportional to the compression of the sensor

    The compression of the sensor is proportional to finger joint’s angular velocity.

    can you change the grip strength of the hand for users who have weak muscles?

    We can’t change the grip strength because the servo motor doesn’t have such function. It only can change the angle. If we want to change the grip strength, we have to use more expensive servo motor or DC motor with current control system.

    On the other hand, we tested this sensor system with elderly people and children(non-amputee). They could successfully control the hand. Therefore, I think this system can be applied to users who have weak muscles.


    Hiroshi Yamaura

    exiii has stilll been examining 3 methods for making socket.
    No.1 Velcro type socket.
    No.2 Using 3D scanner and making 3D printed socket.
    No.3 Combining traditional prosthesis socket and HACKberry.

    Today, I’ll introduce the example of No.2 type socket.
    I used Sense 3D scanner(http://cubify.com/products/sense) and made the socket using the method above.

    The problems were the battery and the sensor.
    In this model, I made a large hole at the medial side of socket so that the sensor can be set on the stump.
    As for battery, the battery box was made separately from the socket.
    This is more natural than current velcro type socket.

    BTW, Polish engineer Bartek reported he made a plaster cast of stump and got the 3D data using 123D catch(http://www.123dapp.com/catch).
    This is more affordable way than ours.

    Do you have 3D data of Delaney’s arm?
    If you could send me the data, I may be able to make the socket data and send it to you.


    Hiroshi, I’ll send you the data. Thank you for offering to help.
    We plan to work on the socket option as well. The velcro option was the quick solution for the newscast.
    Cameron and I have been discussing about Delaney’s socket. Cameron is concerned about Delaney’s wrist movement. Anyway, I’m looking into a socket that has some space to allow her wrist movement.


    Hiroshi Yamaura

    Hi Tima,
    I created Delaney’s socket for the reference (I could make only 1 size example. Sorry).
    The purpose of this socket is just to show how to connect 3D scan data with HACKberry’s wrist parts.
    You don’t have to use it.

    I used Meshmixer for editing STL data and Fusion 360 for making surfaces.
    If someone want to know its modeling process, check this Fusion 360’s link.

    I’m looking into a socket that has some space to allow her wrist movement.

    Yes. We should make the best of the degrees of freedom.
    Please let me know if you come up with a good idea.


    Hi Hiroshi,
    Your socket looks great!
    I’d like to print this one out and try it on her.
    We don’t know until we try and then we’ll be one step closer to knowing what we should do, right?
    The socket in post #615 looks quite loose with the big cutout on the arm whose is wearing it in the pictures. If that remains attached, maybe it’ll work with Delaney.
    Thank you very much for your help, Hiroshi! Hope you continue to help.


    Hiroshi Yamaura

    Haha, you’re right.
    Please try it and give me a feedback.
    Of course We’d be glad to continue to give you our cooperation.

    As for #615, the cutout doesn’t affect the fitting so much.
    In his case, lateral side and upper/lower sides were important to receive the load but medial side was not.
    I’m not sure but we may be able to set the sensor here.
    If someone think we should set the sensor on socket, it is not good idea.
    The load from the hand affect the sensor output and user will lose the control of the hand.
    Therefore, I think we should separate the sensor from the socket.

    BTW, Tima and I discussed about how to apply the HACKberry for user who has short stump(has couple inches below the elbow).
    I introduced this picture.
    I mean, if someone has short stump, he can’t get enough suspensibility from forearm.
    Therefore, we should consider using upper arm cover.
    Another example of upper arm cover is as follows.


    Hi Hiroshi,
    Here’s the socket you made (#619). Looks very nice. Thank you!
    We’ll try it on Delaney and let you know how it fits.

    Also, with the permission from Delaney and her parents, here’s the link to the stl files of Delaney’s arm scans if anyone would like an access to them.


    By the way, I didn’t think the sensor on the socket is a good idea either. Seeing how the weight of the hand can cause the muscles to flex in the way that interferes with the sensor when I didn’t get the velcro attachment on quite right. A lot of people try to talk me into it. I’m glad you confirmed that.


    Hi Hiroshi,

    The socket looks a little loose and won’t stay on Delaney’s arm, although not as much as the socket in post #615. Sorry I don’t have the picture showing the socket on Delaney’s arm yet (still waiting for the pictures from her parents).
    Are we supposed to have some kind of sleeve on Delaney’s arm before sliding the socket over it? Is that sleeve supposed to help keeping the socket stay attached?

    I have only seen a conventional prosthetic socket that is very tight on the skin to stay attached without having anything in between. I guess that might prevent me from understanding this arrangement.


    Hiroshi Yamaura

    Hi all,
    Tima and I were discussed about Delaney.
    We’ve come to the conclusion that the topic about her arm is too intimate to share here since she is still young.
    Therefore, let’s stop to talk about this project in this forum.
    Of course the project will continue and exiii will give our coopetation.
    The result will be shared from Pikes Peek Makerspace in the future.
    Let’s wait until the time comes.

    BTW, I made the movie that explain how to make a socket for HACKberry from 3D scan data.
    The required softwares are Meshmixer and Fusion 360.
    It is a tentative method. If someone finds a better way, please let us know.


    Hiroshi Yamaura

    私たちの協力者の小林さんが3DスキャナとFusion 360でオリジナルのソケットを作り、作業の様子を共有して下さいました。


    One of our collaborator Yoshiaki Kobayashi created original socket using 3D scanner data and Fusion 360.
    He is prosthetist on active duty.
    He combines his knowledge as prosthetist with digital fabrication technique.
    I believe this process will definitely help others who want to make 3D print socket.

    His advices are as below,
    -Place the scan data to appropriate point. He set a center of elbow joint as origin.
    -Be careful about angle of wrist.
    -Make the edge flared to avoid digging into skin.


    Hiroshi and Yoshiaki,

    Delaney’s prosthetist wants to make a conventional socket for her to try and asked me to provided the below parts to him as well.


    I just had a chance to watch Yoshiaki’s video. That looks great! Will try it.


    Hiroshi Yamaura

    Hi Tima,
    Yes, the video is very interesting for me too.
    He will make next version of socket by 27th May.
    I will share his update 😉


    Justin Brawley

    Hiroshi Yamaura

    would you consider sharing your files for 3d printed socket that you have pictured above?


    Yoshiaki Kobayashi


    Nice to meet you. I’m prosthetist Yoshiaki Kobayashi. I designed prosthetic hand (wrist parts are HACKberry’s) that corresponds to the short stump of the forearm, including the socket-cuff in a portion of the proximal side from the wrist.First of all in the photograph let me introduce, if necessary data and how to make will be to introduce therefore please.


    Hiroshi Yamaura



    View post on imgur.com


    View post on imgur.com


    View post on imgur.com


    View post on imgur.com





    ・?dl=0 を削除
    ・www.dropbox.com を dl.dropboxusercontent.com へ変更




    Hiroshi Yamaura

    Nico made the socket for HACKberry from his 3D scan data.
    He also upload good tutorial movie.
    It is very nice! Thans Nico.

    Please check this movie if you want to make the socket data.


    Hyo Sang Yoon

    Hello, Hiroshi
    Can I share the socket file attached to the

    STL file


    Hiroshi Yamaura

    This is not exactly the same one but may help you.
    It is the latest socket for our collaborator Morikawa-san.

    View post on imgur.com

Viewing 28 posts - 1 through 28 (of 28 total)

You must be logged in to reply to this topic.