K
Welcome @jo.foe, @sepi, Jens (@Salzmann0704), and Timothy (@Vorg) ! It's great to have you all join the forum, many thanks for your interest in our collective project.
No worries about your not-perfect English @jo.foe, quite a few of us are not native English speakers ourselves.
Glad to have you all here.
@sepi said in Sourcing parts:
@Vorg I did my homework and research the material but sourcing mostly industrial use material can be challenging as an amateur. Especially since the manufacturer in this case is from the USA and I'm in Europe.
We strongly sympathize with you on this front! We've tried to make it as sourcable and low-cost as possible, but this has it limits.
@quinnale this look great!!
@quinnale said in Towards a working system:
In addition, by printing the gasket template for a guide, and then cutting the holes and cavity with some cheap circular punches and an x-acto knife, I got some jenky gaskets.
I've done similar, it's definitely possible as you've shown!
@quinnale said in Towards a working system:
Also, should I have made this a blog post? Happy to delete and move if that's better.
Oh no worries there, I made that category in part as a container for Daniel's blog - WordPress and our forum software can interact (via ActivityPub), so that we can reference and comment on his (and other people's) longform WordPress posts directly here in the forum, like what's happening here.
@quinnale said in Towards a working system:
Also, here's an update on the 3D-printed pump! Been running it for about an hour, no issues. Unlisted Youtube video . If anyone is curious I can provide more details. Gonna try run it to death, see if the plastic or the tubing fails first.
That's awesome! Having another pump option besides only Kamoer would be great (esp. an open-source one!). It's this pump you've used?
@sepi said in Towards a working system:
eing a complete newb, I slowly gather an understanding of the individual components of the cell.
Glad to hear it, that's one of the whole points of this forum!
Hi @sepi and @jo.foe, apologies for the delayed response, I've been offline for a few days. Perhaps this might help: https://fbrc.dev/faq.html#what-is-a-flow-battery I've tried to write up a straightforward explanation but let me know if it's not clear or if it could benefit from adding something.
@jo.foe said in Whole system concept for more security and efficiency:
Is it pumped everytime from and back into the same reservoir?
The two pumps are (in normal operation) continuously circulating electrolyte from their respective reservoirs to the cell, where the electrolyte goes through either the positive or the negative flow path (manifold + flow frames), and then back to the reservoir.
@sepi said in Designing the large-format cell:
Any particular reason why you don't consider milling or laser cutting? FDM printing large surfaces that need to stay flat seems to be a challenge on general. Have you envisioned printing the structure (manifold?) on a prefabricated sheed of plastic?
I have considered this, and I'm not against it, I am just pursuing FDM first until I can prove that we need to try something else. With our current approach it's desirable for sealing purposes to have a flat, sealed top surface on the flow frame, ie overhanging geometry, so laser/milling are out, unless you try to bond a layer on top, and it starts to get complicated. Milling plastic, especially fine details, is also hard to do well for amateur machinists - it tends to just melt unless you have your speeds and feeds totally dialed + flood coolant. We hope to have an FDM printable design to allow as many others to get in on this, since FDM printers are much more abundant and easy to use than capable plastic CNC mill setups.
Printing on plastic sounds kind of hard, especially for PP filaments, and it doesn't get rid of the overhanging geometry question.
@sepi said in Designing the large-format cell:
Also did you ever consider TPU for the parts in contact with the electrodes? It's pretty resistant to chemicals and ribbery and could thus male seals.
We haven't tried TPU yet, if I recall @danielfp248 saw somewhere that it wouldn't be compatible with the zinc-iodide chemistry that has been our main workhorse - specifically the charged triiodide species, which is really a pain in terms of chemical resistance. We'd need to confirm that it's resistant to triiodide. You're right that it's flexibility could help make it a good seal though.
@sepi said in Designing the large-format cell:
I guess that both injection molding and casting don't allow for fancy internal geomtries needed for the manifolds.
Both injection molding and casting can do internal geometries, but as @Vorg pointed out it's cost prohibitive on small-scales.
@Vorg said in Designing the large-format cell:
I've never done any 3d printing, but from what I read, they don't do overhangs well because the hot plastic would not have any support under it and adding supports in the channels would restrict flow.
FDM printing can "bridge" to some extent, which we are exploiting to print our prototype flow frames here. @danielfp248 made a great post showing some of his efforts to print it here, I highly recommend you check that post out - we're definitely making progress with this approach! Eventually, if we got a really great flow frame design that was locked-in, it may make sense to make a lot of them with injection molding.
@wordmark@mas.to said in Designing the large-format cell:
Previous thought: so u say this cell makes O2 and H2 going out from separate outlets?
No, that is not at all the intended operation, but it is a possible unintended side reaction if something goes wrong or is poorly designed/outside the operating mode.
@wordmark@mas.to said in Designing the large-format cell:
What volts amps will your stack require? Production costs per stack?
@wordmark@mas.to said in Designing the large-format cell:
what are the properties for such a battery? (How does it compare to a 48V 14kWh #LiFePo4? (Which can burn down houses if cells are not high quality)
We are very far from having a datasheet on such an RFB, as we are still very much in the R&D phase, but there are some datasheets out there from commercial companies which could give you an idea, but generally RFBs don't make sense below 10 kWh in size because of their size and requirement of centrifugal pumps.
We have a FAQ that might answer some of your questions here.
A snapshot of some messing around with OpenFOAM. Initial conditions aren't correct yet and no porous zone is present. But it runs on my laptop!
Daniel has been prototyping this design, he made a post about it here: https://fbrc.nodebb.com/topic/9baae384-75b5-40bc-ae3c-de14abd8f8e4/working-on-a-large-scale-open-source-flow-battery-design-and-kit
@wordmark@mas.to do you mean in general or in the context of this RFB cell? Ideally, we won't be generating any detectable H2 or O2, in practice, the flow paths in some cells are arranged so that the cell can be placed in a way where the fluid flows through the porous electrode upward, against gravity. If side reactions that generate H2 or O2 occur, and you make bubbles, those will ideally clear through the cell upward in the direction of flow.
I've just finished the CAD assembly of the first iteration of the large-format cell, which we will manufacture and test for leaks and other issues.
The repository is here.
Active area is currently 175 cm², and flow frames are hopefully printable on a 200 mm x 200 mm FDM printer. The flow frame is sized to fit on the printbed and still leave room for a healthy brim.
Manufacturing files in CAD/exports.
| Filename | Process/Material | Quantity |
|---|---|---|
endplate-EndplateBolt Holes.step |
12 mm (1/2 in) plywood/MDF | 2 |
outer-current-collector-Outer Current Collector.step |
1 mm brass | 2 |
flow-frame-Flow Frame - Barbed, For Single Cell, Northeast.stl |
FDM / PP or ABS | 1 |
flow-frame-back-Flow Frame (Back).stl |
FDM / PP or ABS | 1 |
Gasket.svg |
TBD - thin <0.5 mm gasket material | 3 |
Outer Gasket.svg |
TBD - thin <0.5 mm gasket material | 1 |
Inner Current Collector.svg |
1 mm graphite foil | 2 |
| M10x50 bolts/nuts/washers | 12 |
All flow connections are on one flow frame, which passes the other electrolyte through to the next half cell through internal manifolds.
There are some shunt current flow paths and flow diffuser geometry in place, but these are not optimized and mostly representative. We will see if this type of geometry is even printable/sealable before spending much effort on CFD/FEM/shunt current modeling. This general approach is, however, roughly consistent with what I've seen in applied research and industry.
The "back" flow frame doesn't have barbs, and connects through the barbed flow frames internal manifolds to receive/output flow. It's flow fields are mirrored versions of the other flow frame, and the backside of this flow frame is completely blocked.
For endplates, the design is a simple flat sheet. We will see how 12mm plywood holds up with a 12-bolt pattern. Eight of these bolts straddle evenly the four barbed flow ports. There is a tab cutout for the current collector connection. Instead of making the current collector tab longer, I decided to cut out the endplate slightly like this, so that the current collector is cheaper to make on services like SendCutSend (which calculates largely on rectangular part area, and brass >> plywood for these thicknesses). Technically, only need the large 20 mm port holes on one side of this cell, since all the barbs are on the same side, but to keep unique part count/costs down when ordering and simplicity for now, I haven't created separate port-hole-less parts yet for the endplate and inner/outer current collectors.
Gaskets and inner current collectors all largely follow the shape of the flow frames, with the Outer Gasket.svg having larger port holes to allow the barbs through.
I've also modeled the pump we're using:
The code to control the pumps using an Arduino Uno and thyristors is here, and a corresponding GUI based on the MYSTAT software is here.
I haven't written any documentation yet because things are still changing quite a lot right now - I'll keep posting updates in this thread as we progress!
@lucas welcome! We will be doing more power electronics as the project continues, see some of the discussions and diagrams elsewhere in the forum here for example. We plan to experiment with some of the OSHW power electronics options from OwnTech for our large-format cell/stack: https://www.owntech.io/
Here is an updated version! https://spectra.video/w/nJ8XNYu1MXNPSDLKV3KVTh
Also, apologies for the wait, but here are some more detailed instructions on how to build and wire everything (though you seem to have figured it out)
A video to accompany the written documentation: https://spectra.video/w/nJ8XNYu1MXNPSDLKV3KVTh
Improved documentation page on the electronics: https://fbrc.codeberg.page/rfb-dev-kit/electronics.html
@gus said in Following your documentation – feedback & questions:
I have send you an email,
Received, thank you!
@gus said in Following your documentation – feedback & questions:
Thank you. I am planning to use a Raspberry Pi for long-term tests. Are you using a Raspberry Pi too, or a PC?
I am using an old laptop that was in our lab, but a Raspberry Pi would work fine. I usually install ZeroTier or similar and then VNC in for remote access.
@gus said in Following your documentation – feedback & questions:
And also, I wanted to ask you about the current collector's material – is brass a better option than pure copper? What was the criterion for the material type and thickness selection?
Brass is somewhat more corrosion-resistant than copper, which is why we specified it. 1 mm is an arbitrary first guess for a reasonable thickness that's stiff and conductive enough, we haven't done any calculations for this however.
@kirk said in Following your documentation – feedback & questions:
In the CAD files we have specified a 2D endplate (https://codeberg.org/FBRC/RFB-dev-kit/src/branch/main/CAD/exports/Metal Endplate.step), that could be laser cut or milled from aluminum. @danielfp@chemisting.com has received endplate versions of this in aluminum, the only catch is then you need insulating washers so as to not short them. Another rigid polymer could work too.
I found a pic of this setup: 
@gus said in Following your documentation – feedback & questions:
The cell itself doesn’t leak. I used a torque of 3 Nm for now, but it already looks slightly distorted. Have you considered using more bolts?
In our meeting today, @quinnale mentioned that in their lab they use 1.5-2 Nm max to tighten a similar size of flow cell. @danielfp@chemisting.com also builds his FBRC cells without a torque wrench, by tightening them only as much as he can holding the short end of the allen key.
@gus, awesome work! This is so exciting to see you've independently replicated the setup. Apologies on the delay for the pump wiring, I have recorded the raw video of me setting it all up as well as the pin numbers and I am in the process of editing it and updating the docs to match.
@gus said in Following your documentation – feedback & questions:
The new reservoirs seem fine now.
Did you encounter this problem too? How did you solve it?
Our best strategy against leaking in PP prints has been 100% infill, 5 perimeters. I know some have success with slight overextrusion / flowrate multiplier of up to, say, 1.02 times. We have had reservoirs leak there before but tuning print settings usually leads to a tight reservoir. Could you share the file you modified? It looks like you added a big cylinder.
@gus said in Following your documentation – feedback & questions:
The cell itself doesn’t leak. I used a torque of 3 Nm for now, but it already looks slightly distorted. Have you considered using more bolts?
I think we need a stiffer endplate - similar hole pattern geometry, just thicker. This is an area to improve for sure, it's not ideal as-is. In the CAD files we have specified a 2D endplate (https://codeberg.org/FBRC/RFB-dev-kit/src/branch/main/CAD/exports/Metal Endplate.step), that could be laser cut or milled from aluminum. @danielfp@chemisting.com has received endplate versions of this in aluminum, the only catch is then you need insulating washers so as to not short them. Another rigid polymer could work too.
@gus said in Following your documentation – feedback & questions:
I also wanted to ask: is silicone resistant to this type of chemistry? Since it’s used for gaskets, maybe silicone tubing could be used as well?
We've tested and silicone tubing doesn't work for zinc-iodide, unfortunately. The gasket application of silicone is forgiving enough that it seems to work for this purpose, however, but we may be pushing our luck.
@gus said in Following your documentation – feedback & questions:
Regarding the Arduino program: it seems there are four defined pins — In1 (pin 9), In2 (pin 8), In3 (pin 7), and In4 (pin 6) — but they don't appear to serve any purpose.
These are legacy pins from when we used to use an H-bridge driver for different motors, before we got these stepper motors with built-in drivers.
This code is outdated, I have the new code but it's on my lab PC which is offline right now 
I am going to upload it soon.
@gus said in Following your documentation – feedback & questions:
As for the main Python program — should it work with no issues once the Arduino and MYSTAT are connected to the computer?
The program here * should * work on linux when run as root (in order to have proper USB access). You can control the pump speeds via the Arduino with the software, even without the MYSTAT connected (under the "charge/discharge" tab) (this will all hopefully be in a video soon!)
I'll have some more info soon (tomorrow hopefully?)
@kirk said in Following your documentation – feedback & questions:
We have used the Membrane Frame on-and-off through some of our testing as we tested different separator materials and thicknesses. I just went back and checked some old notes, we did a lot of testing with three layers of photopaper in the end, so go ahead and just use three layers for now! I see one section of the docs said four layers and another said three, as well as the old photo.
*use three-layers of photopaper and the membrane frame, I meant to say!
