Custom Sample Holder (v3) - design and automation #124
Comments
Sample Holder v1The 1st version of the sample was designed from two components:
Both components are assembled with snap fit joints that are part of the closing cap.
Joining both components would create a seal and electrical contact between the sample and the conductor at the point of surface. The design is 3d printed with resin v4. This design had multiple flaws:
Experiment not conducted Reference pictures: Sample Holder v2It was designed from two components as the first one.
The snap fit joint were removed and replaced with "twist and lock" mechanism. Rubber washer and O-ring was added in the design for seal and compression of the contact point for better electrical contact. Similarly, the route angles of the main body's "tunnel" was improved. The design was successfully 3d printed with resin v4. Compared to v1 this one had:
However; It was longer than the v1 so it hit the bottom of the cell. One of the locking mechanism had to be removed for the holder to be inserted in the cell. Prototype experiment was conducted on November 5th, 2024 Pictures of the 2nd version of the design:
|
|
Amazing, thank you for the clear summary! Slight clarification:
I think this one is not as critical. For example, I wouldn't mind if the conductor were permanently fixed into the sample holder with additional resin that gets cured after the conductor has been inserted. In the previous discussion, we've been leaning towards some kind of threaded mechanism to allow the user to tighten it to a particular pressure. We've also verified that we can do threaded connections with clear v4 resin. It might be difficult to remove the sample after a long test in 80 deg. C KOH, but I think it would be OK to leave that for later testing, and I think it could be solved with some Teflon tape around the threads. It may be worth again considering two long, thin, flat strips of a SS sheet going into the holder with the SS slightly raised where the electrical connection with the sample is meant to take place. I think this would provide a larger surface area of contact without leading to damage of the sample (i.e., larger surface area / more uniform pressure) compared with a round wire. I think that the threaded mechanism would provide a reasonable amount of pressure. I've worked with some really thin SS sheets before (can be cut with regular scissors, rather flexible). I'm not sure what sizes are standard for SS strips / "flat wire" (I had some McMaster links before), but I don't think it would be a big deal to assume cutting a strip from a larger sheet manually. This would also likely require a redesign such that both SS strips are inserted straight into the holder vertically, with only that slight bend near the point of contact. A connection would need to be made between the two sheets (either by using a larger sheet and cutting out a smaller rectangle to make a sort of upside down "U" shape or making the connection between the two with some kind of clamp. If the SS is thin enough, we could also do a bit of origami with an extra long strip of SS Lmk and I can draw some diagrams. |
|
Thanks for the great summary of our work so far @MiceeNS and the cool ideas for the next design @sgbaird. I agree, that the removability of the electrical conductor is not a high prio. It would be a nice addon for when we start using Pt-wires, as these are a bit costly but not too important for now. I like the idea with the flat stainless steel strips, as we would natively get a broader contact area this way. What's your opinion on the closing mechanism @MiceeNS? Would you also give threads a try or do you prefer improving the current design? |
|
Yes I will do now a design with threads(because I do not want to waste additional time figuring out if something works or not), and I am thinking of a design with the stainless strips. However, after we print the design with the threads and hopefully everything works and there is good electrical contact I can think of some additional mechanism. Also will update this issue with the v2 today
I believe I understand you, no need of drawing at this point. |
|
Check this design out and let me know, next design I will do with flat wire. Threads are "custom". Also, I just noticed some important fillets are missing.
|
|
Oh, interesting! I was imagining the whole mating piece being the threaded piece (rotating the main piece is the locking mechanism), rather than having multiple screws on the outside. @programlich what are your thoughts? What are you picturing when you're thinking of using a threaded design? In the current design, do you think we'd need longer screws and a custom nut, too? (The nut just for an extra bit of security). If we take this approach, maybe worth considering using Teflon screws and nuts: https://www.mcmaster.com/products/screws/material~ptfe/ (e.g., https://www.mcmaster.com/products/screws/rounded-head-screws~/material~ptfe/) I think this design could work well without a lot of iterations required. Maybe worth having four screws in a square pattern to give a more even pressure distribution of the sample contacting the SS. |
I thought of that but you might not always get the sample straight or it might fall off.
Good point, can add tomorrow.
I think we wont need teflon unless tolerances are way off. I think teflon is more for a leaking problem in threaded connections.
Totally agree. EDIT: I would like to buy the plastic screw and test the threading before printing anything. Standard threading has way smaller pitch and height especially with small screws like in our case is real pain. The standard threads I tested were not working even tho I printed 3-4 designs. But I think the problem with those threads was in the beginning of the thread it was not designed properly. It doesn't have to be mcmaster they are expensive. Will try to order from somewhere tomorrow And by the way Fusion 360 does not have NPT thread feature. |
Great point. Now that you mention it, that sounds like a headache trying to not rotate the sample during it. (EDIT: unless there's an inset for the sample, but I still think it's worth pursuing the Teflon screws like you mentioned)
I was thinking more so in terms of standardization. I think Teflon would be less likely to get "stuck" or have some kind of unintentional bonding compared with the clear v4 resin (while also having the best chemical resistance), but we don't know if that's a real problem for the resin. Aside: I found a potential candidate for the SS 316, which I think we would just cut to shape with scissors or similar. I don't think there's a product listed as "flat wire" that is thin enough and wide enough. See https://www.mcmaster.com/product/3774T55 The next closest thing I found was a strip ~4x thicker, and 1" wide (too wide): https://www.mcmaster.com/product/9090K3-9090K53. |
Great work! One question about this design: If we removed the threads from the part, which actually holds the sample, would we then get a system, in which both parts are nicely pressed together when screwed? I imagine it as the large part of the sample holder acting as nuts. Could I express the idea?
I thought about it like you @sgbaird but also totally forgot about the point of getting the sample straight which @MiceeNS mentioned. So let´s go with this design for now.
If it´s possible to get a secure locking mechanism without external nuts, I´d prefer this. Handling the sample holder, screws and nuts is just more complicated than only sample holder and nuts. Time is sometimes critical when the sample is very reactive with air.
Nice one. Should be absolutely doable to cut a "wire" from it. As a sidenote, we might think about trying out Nickel in the future, too, and compare it to stainless steel. Will perhaps show better resistance. |
|
We can test print one spring, in the name of science 😆
|
Is this the printed spring? I also noticed several other printed parts, such as nuts and bolts.
This print has supports that are very challenging to remove. Update: IMG_0973.MOVand the nut and bolt fit well |
|
Yes those are the parts. Test the bolt to the nuts and try what you can do
with the spring.
Let me know how it goes
…On Thu, 5 Dec 2024 at 09:16, Yanghuang Liu ***@***.***> wrote:
image.png (view on web)
<https://github.com/user-attachments/assets/31500d53-f6eb-4be3-bb8e-c09adead7ebf>
Is this the printed spring? I also noticed several other printed parts,
such as nuts and bolts.
This print has supports that are very challenging to remove.
—
Reply to this email directly, view it on GitHub
<#124 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ASKERIFOKK2ME6LB2BWZHC32EBN5BAVCNFSM6AAAAABSMLULXWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKMRQGQ2DMNZTGM>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
|
@Neil-YL thanks for your help! @MiceeNS it looks like you sent the print for the springs and the bolts to the Bambu Lab printer, not the Formlabs printer. The former (PLA) isn't chemically compatible, and the clear v4 resin is much more rigid. @MiceeNS could you identify a suitable combination of "Extreme-Temperature Chemical-Resistant PTFE Pan Head Slotted Screws" and PTFE Hex Nuts based on the model, i.e., whatever is close in size and length to what you're hoping for with. 
If you could add it to the 3D model (and include some snapshots) that would be great so you make sure the size and length look OK. I suggest having the through-hole in the sample holder straight without threads so we can avoid using a tap and just use the nut. Could you also identify a corresponding taper chamfer tap that would allow us to create or clean up threads? Then, print a couple different sizes of pucks with holes (with and/or without threads) to help with seeing what would work best with the tap and the teflon screws. This will provide some flexibility for us to pivot to a nut-free design if we wish. |
|
For context, I'd like to get a set of five screws, five nuts (1 extra in case of breakage), and a tap ordered sometime soon. I'm also planning to get the 316 SS linked above. |
Yes, I send them on the bambulab on purpose to see if the prints would actually work and the thread could fit. Regarding the nuts such a good idea! Will do and send as soon as possible, buy and ship them to you. |
|
Update regarding the last changes. Bolts nuts and tap is ordered, lets figure out the design for flat wire then I will order the sheet metal also. |
|
Some renders..
Working_Sample_Holder_Version_3.1_2024-Dec-09_03-39-06AM-000_CustomizedView4687765946_mp4.mp4
|
|
For testing the Tap with modified threads sent for print on Bambulab. If you received the components from McMaster Carr this can be tested with the bolts.
|
|
short question regarding the electrical contact on top of the sample holder: Do you have a link to the part, which is meant to connect the wire to the contact? Also, did you get a chance to brainstorm the counterpart on the tap of the cell? I´d really like to see, what kind of electrical connection this is |
The part that connects the wire to the "connector" is just a bolt something similar like in the picture below, you insert the wires inside the compress them with the bolt The "connector" is like small pin with curved face on the counter part I intend to put compression spring so when you insert the sample holder that counter part can retract a bit and have a constant force for a better electrical contact. Picture of the connector:
Other idea that I had is with socket connectors, something like in the picture. The female socket on the tap itself And the male part on the sample holder If we go with flat wire, we can use this type of socket to connect both wire in the sample holder |
|
I lean heavily towards using a molex connector or a banana plug. See #104 (comment) One advantage of a Molex connector is that they usually involve crimping rather than soldering, as opposed to banana plugs which typically require soldering (and then we need to figure out a good way to get SS soldered to the banana plug in a robust way).
There would also likely be a fake tip slightly offset from the center of the sample holder, to allow for picking up and moving sample holders around. Off-center for the sample holder fake tip because I think it would be easiest to get the desired electrical connection by having the molex connector or banana plug fully centered on the sample. Aside: flat wire = thin SS sheet that's been cut (regular scissors are fine in my experience) such that it's like an upside down U (imagine it's unrounded) with a small "line thickness". This also assumes there are two insertion points from the top, which allows for having a direct electrical connection on both sides of the sample in a straightforward way. Lmk if a diagram would help. |
|
I´m not sure if I get your idea completely, @sgbaird. Would your design idea be an electrical connection potentioastat - tap - sample holder The advantage of the first one is, that we would only have to move the sample holder around and didn´t have to worry about (dis)connecting the electrical connection from the sample holder manually with every sample exchange. Besides this, I agree with you that banana plug/molex has quite an advantage because once the connection is established, it is always the same, so we could be sure, to have a constant electrical resistance regardless of slight differences in the positioning of the sample holder. |
This looks completely fine to me. Easy and functional. |
|
Thanks for the great comments! The desire for being able to move electrodes around is something more specific to my goals/vision. I want to make it easier to reuse the same potentiostat for entirely separate workflows in the same fumehood/lab, without having to manually go in and swap electrode connections each time. The current focus is on the working electrode, but the same principle applies to the counter and reference electrodes. Also, I remembered that I'm leaning away from a "fake tip" with an OT-2 pipette towards instead using a low-cost robotic arm (specifically MyCobot 280 Pi is the one I would first try with, then move to something a bit more expensive/higher performance if needed). Instead of a fake tip, it would be something easy to grip onto with the corresponding gripper. This could be using a parallel gripper, adaptive gripper, or suction gripper with the MyCobot. @MiceeNS, thoughts? Looping @gursi26 and @SissiFeng to weigh in, since both of you have had experience with this arm and with robotic arms in general. See https://accelerated-discovery.org/t/455 for context.
It would be potentiostat -- "molex holder" -- sample holder. The workflow would be:
Step 5 will likely require some trial-and-error. For example, will the mechanical connection of the molex connector be so strong that it picks up the sample holder along with it? (seems unlikely to me based on weight of the sample holder and my experience with molex connections -- banana plugs would be a different story, I could imagine a banana plug picking up the sample holder, in which case we'd need a workaround to separate the two, which has a precedent in the N9 platform (figure 1.D) and digital pipette v2 (not yet released) with a syringe and a tip)
True, as long as we can get the connection/disconnection to be fairly reliable. There's an equivalent option, which is similar to what @MiceeNS has brought up. The "molex holder" would be embedded into the tap (with the connector pointing up instead of down) and the sample holder molex connector pointing down instead of up.
👍 |
|
I have some ideas. It looks like there are two items that need to be grabbed. Swivel adapter for switching between two gripping heads: |
I have picked up this printed part, are there any tests I need to carry out with it? |
The threads are for testing standardized NPT ( 4-40 screw at the printed part) and nothing I could change there. We need to see if the 4-40 can fit in the thread. the holes without thread should be threaded with the TAP and see if the bolts can be screwed. If the bolts do not fit in the printed threads, try to thread those also with the TAP. Basically what we want to test if its possible to print thread that we can get rid of the nuts in the design. Bolts tap and the nuts should be -> This can go into the training lab space on the 2nd floor on 2620 Speakman Drive, Mississauga Thank you, |
My previous experience for small threads on 3D printed part is using embedment nuts or thread-head inserting into the 3D-printed holes, may need to apply some glue for critical connections. 
|
We can not use because they are not chemically compatible and metal can interfere with the electro magnetic fields. The real saga |
|
We don't have anyone regularly going to NRC from the AC, so it may take some time before we get the bolts/nuts/tap back to UoT for testing. |
|
Thank you all for the great ideas. A lot is going on here! About the electrical contact:
Now I understand you better, @sgbaird. I think this is a very valid point, which we need to address even though counter and reference electrode have a lower priority than the working electrode, as they can be used for at least a week. However, I agree, that we should standardize the electrical connection potentiostad - any electrode to simplify the future exchange process. On the weekend, I discussed this topic with an electrical engineer (thanks @ggramlich26 ) and he suggested to use spring loaded pogo pins for the electrical contact. The spring allows a reliable contact and these parts are commercially available with cold coating, which would be great in terms of corrosion resistance. The pogo pins might be loaded on a little PCB, which also connects to a banana plug/whatsoever to the potentiostat. Recording.2024-12-15.203808.mp4
For the future, we might think about building small adapters for the reference and working electrode, so they can also be connected via the pogo pins. What do you think, @sgbaird? Would this be an option? |
|
@MiceeNS, could you please post the specs of the PTFE bolts and nuts for this sample holder design? I might have the chance to print a sample holder with a formlabs printer here in Germany in the near future and would like to order the bolts & nuts |
|
Line Product Ordered Shipped
|
|
The stainless steel wire by now shows quite some signs of corrosion after using it in hot KOH at different potentials.
For the sake of cost reduction the nicest thing would be to have a stainless steel or copper wire where only the end is coated/massive platinum. That way only Platinum would be exposed to KOH while the wire which is inside the sample holder and thereby not exposed could be of any cheap material. |
|
Looks like the image upload might not have processed all the way. Could you edit the post and reattach? |
|
AS a sidenote @sgbaird : would you mind creating a github project for us, so we can keep track of all the minor things we need to address? |
Is this the mechanism I should be picturing? https://youtu.be/-oxunE-RMjA. From the video, it seems like it depends on mechanical pressure being constantly applied to maintain the connection. Is that correct? Not sure how that would affect the results (similar to what was mentioned about the sample to wire connection not being robust enough). |
Added you and @MiceeNS as admins, and made the project publicly visible. I didn't add much to it - feel free to customize as needed: https://github.com/orgs/AccelerationConsortium/projects/5 |
Yes those are the pins. This link is to the same ones I used in the design https://www.mill-max.com/products/spring-loaded/pogo-pin-header-strip/827/10-002101 In the description of the pins it says: So I think that will (EDIT: not) be a problem Also useful link: https://www.ptr-hartmann.com/know-how/spring-loaded-connectors/ |
You mean that's a problem because of too much force pressing the sample holder up? Perhaps we can find Pogo pins with less spring force? I think we could solve this by either not pressin the pogo pins very far to be in the range of very low forces (30g would be totally fine) or add some weight to the sample holder by simply placing a steel ring or something similar on its top. |
Done. Please feel free to add and modify anything. |
My bad I miss typed, I meant that will not be a problem 60-100grams. |
|
Update: Do you think the banana jack should be from the top or from the side? Like this the banana jack would create a box that would be 40mm height. Another option is to find smaller banana jack or place it sideways or remove it from the PCB. The pins by itself are small but placing a banana jack next to them for the potentiometer is a problem. |
|
Is it banana jack and pogo pins? I was under the impression that the pogo pin suggestion was in place of the banana jack. Also, should be from top. Also, I'm a bit confused why there's a potentiometer. Is what you showed above an example of three possible designs? |
|
I think it depends if we want to have the banana jacks removable with a robot arm in the future. If that's the plan, we should have it placed from the top but if we plan to leave it more permanently connected to the cell, it should be from the side to leave more room. Also, please try to not place the PCB into the tap, but rather on top of it. I am worried about taking material away from a component which will be attacked by hot KOH vapor. @sgbaird, I think Bojan refers to the Potentiostat as the connection with the current design would be sample holder -> pogo pins -> PCB -> banana jack. I think it would be best to setup a meeting with the three of us because here's a rather important decision to be taken which is best discussed in person. |
|
@MiceeNS, one question regarding the ptfe screws for the current design of the sample holder. Would metric M3 screws with 8mm length fit through the designed holes? I am trying to order the screws and bolts but it is challenging to get ptfe screws in imperial dimensions here. |
|
Printed spring with Formlabs high temp resin: signal-2024-12-19-12-13-10-751.mp4Looks pretty promising to me. Thanks @ggramlich26 |
Because the potentiostat has banana pin.
The screw clearance is okey, however from wherever you order them check the nut size. On Mcmaster the metric is bit smaller and the cutout is a bit big for that nut size. I can adjust that quick. Let me know if you want it changed before you print.
|
|
If it's something that can be done quickly, it would be nice, if you could adjust the but cutout. No pressure though, as I'm in holidays now :) |
|
@programlich |
Comment Link ; #104 (comment)
Shared file links of previous design versions
Sample Holder v1 (without lid attaching feature) shared file link: https://a360.co/3CUgaXG
Sample Holder v2 (without lid attaching feature) shared file link: https://a360.co/3OpeSGK
Design requirements:
Let me know if I missed anything from the requirement so I can update this list
@sgbaird @programlich
The text was updated successfully, but these errors were encountered: