It is mostly 3D printable and there is one brass/copper plate that needs to be machined in a simple way, definitely no problem for any university machine shop. We also now have identified pumps/tubing that work much better.

@a.rahimzadegan, right now the fastest way to get the latest version of the kit is to build it following the instructions at https://fbrc.codeberg.page/rfb-dev-kit/. We would like to sell them eventually but we're not there yet.

Hi Ash, and welcome!

Apologies for the delay, I just finished moving apartments and was unavailable for several days.

Can you tell me or show me pictures of what equipment you have exactly? Do you have a version of the FBRC cell? The MYSTAT software can only control the MYSTAT potentiostat and an Arduino UNO R3 (used for controlling pump speeds). It can't be used with another brand of potentiostat at this time.

Hi Otmar, welcome to FBRC and the project, looking forward to working with you!

@kirk said in How should we control the centrifugal pumps? TRIAC/thyristor etc? Need help from controls/electrical people:

It seems to work! At least enough for testing purposes. Here is a video: https://spectra.video/w/8xipM8aXnBkDXnu4kkRpqT

Here is the code for this test: https://codeberg.org/FBRC/RFB-test-cell/src/commit/d10834bc7dd67736e708c9a33832a5602ab3ca28/firmware/FlowrateRampTest.ino

@methylzero@mast.hpc.social said in How should we control the centrifugal pumps? TRIAC/thyristor etc? Need help from controls/electrical people:

Nice! Better than I expected honestly. At the low end of the speed range it sounds a bit unhappy. The thermal protection is only TP111 so it may not be fast enough to save the motor if it is stalled.
If this motor does work out, the manufacturer can make bigger ones and apparently you can choose the wet-side material. http://www.china-haiyi.com/product-48054-173640.html

Thank! I read some stuff that TRIACs can work for very small motors, and indeed this is only around 6 W, so I figured why not just try it. Yes, at the low end it made some funny sounds, nothing horrible, but this is not definitely not the optimal control strategy. It should hopefully allow us to do single-cell flow testing at close-to-appropriate flowrates, without having crazy fast flow or having to add a bunch of (chemically resistant) plumbing like a bypass/pump-around. For wet-side I think they have two standard options of PP and PVDF for the housing/impeller. Also, for some bigger pumps, they offer BLDC motors stock, apparently, which should be easier to slow down efficiently. Didn't know that TP111 designation either - sounds like it should auto-shut off if it gets too hot at steady state, but won't protect from a stall.

Also, just FYI, R.Flo, a Ukrainian all-iron RFB startup, posted a pic on LinkedIn with these pumps:

Which look to be the same version, but larger.

@methylzero@mast.hpc.social said in How should we control the centrifugal pumps? TRIAC/thyristor etc? Need help from controls/electrical people:

But one thing not good about these pumps is that they might not work great with really dense solutions, max. density is 1.1-1.3 depending on the model, which is .... not much.

Yeah I saw this and... we will cross that bridge if/when we get to it at this low of a price point I read the datasheets with a shaker of salt.

From CRC handbook for potassium iodide (just as a reference point for a salt we currently test with):

In real electrolytes we'll have other salts present at the same time, but even with a 1.3 SG max we may get to reasonable concentrations---as in, high enough to allow us to build out the rest of the system for a prototype. I figure over 1.3 SG the pump either fails faster or has otherwise reduced performance, but maybe we'll find out the hard way

@methylzero@mast.hpc.social @BillySmith @H4K1 @slash909uk@mastodon.me.uk

It seems to work! At least enough for testing purposes. Here is a video: https://spectra.video/w/8xipM8aXnBkDXnu4kkRpqT

Tested the pumps today, they work just fine switched on 110 V AC, will try the triac, if that doesn't work/it fries the motor, will disassemble and try to get a different motor on there.

Video of pumps running: https://spectra.video/w/9VddoPTvMvDCJ121B4fabf

Nice work, Daniel. I am thinking of plumbing solutions to the imbalance issue:

From A review of all-vanadium redox flow battery durability:

After studying the capacity fade for mixed acid electrolyte, UET [154] found that, during long‐term operation, the ratio of catholyte and anolyte concentration remained constant: 1.3:1. Based on this finding, they designed an overflow system with different volume (volume ratio: 1.3:1) anolyte and catholyte tanks, in which the volume ratio and total vanadium were kept constant. With the new design, the VRFB achieved long term capacity and efficiency stability. However, this design is only valid for the mixed acid electrolyte system. Recently, Wang et al [152] developed an electrolyte reflow method to solve the electrolyte imbalance issue for the sulfuric acidvanadium electrolyte system. Figure 10 shows the schematic of their method; without reflow, eventually all of the anolyte will move to the catholyte tank, while with reflow, the anolyte tank will always contain some electrolyte. Similar to the UET method, the volume ratio of catholyte to anolyte is a key parameter affecting the capacity stability and is highly dependent on the operating current density. Cycle life and total capacity were all improved with the reflow method.

There is also Capacity balancing for vanadium redox flow batteries through electrolyte overflow but it was retracted - they think they accidentally had a pinhole in their membrane for the test. But they did build a real overflow system:

That's great! Do you have a link to one or know where we could procure one to test?

I agree about loading up on sensors and then cutting them to the bare minimum as things mature---though some process sensors for chemicals can be pricey!

Are those food-safe pumps centrifugal? We need those to be power efficient---peristaltic won't do at scale.

Edit: saw your other comment, if these mag-drive centrifugal pumps are common in that industry that's great! Do you know what materials are commonly used and come into liquid contact? And what range of flowrates they can provide?

Stephen Hawes of Opulo has compiled some of their tools here:

https://midscale.io/docs

The AutoBOM one seems to be based on those workflows above and looks pretty interesting

Probably too far involved but this sort of approach could be cool, if it was constrained to manufacturable shapes:

We introduce Fireshape, an open-source and automated shape optimization toolbox for the finite element software Firedrake. Fireshape is based on the moving mesh method and allows users with minimal shape optimization knowledge to tackle with ease challenging shape optimization problems constrained to partial differential equations (PDEs).

https://link.springer.com/article/10.1007/s00158-020-02813-y?error=cookies_not_supported&code=2a7f63ae-6652-4af3-9b50-1ffb6e6985da

Linking to previous introductions:

https://fbrc.nodebb.com/topic/7/welcome

https://fbrc.nodebb.com/topic/12/hi-d

Hi all, we ask new members to please introduce yourselves here so we can get to know you and your interests!

It would be nice for us to also know:

• How did you find FBRC and this forum?
• What is your interest in flow batteries?
• How do you see yourself getting involved?

Welcome, and thanks for being here!

Hey @BillySmith welcome!

That's great that you and a few fellow hackers are interested in building a dev kit - let us know how we can help! I am still scheming up the large-format cell. I would really like it if we could FDM print it in polypropylene, but not sure if that's feasible yet. I have some initial thoughts and pictures related to the large-format cell here: https://fbrc.nodebb.com/topic/11/designing-the-flow-frames-for-the-large-format-cell

In the dev kit, we were using peristaltic pumps which failed due to incorrect tubing - those pump at about 60 mL/min. In the larger cell, we will have to use centrifugal ones like the ones you linked to, but we can't have a rotating seal like in that link shows - we need pumps driven with a magentic coupling (also called mag drive, hermetic seal, etc). Here is another forum thread on it with an exploded view: https://fbrc.nodebb.com/topic/8/how-should-we-control-the-centrifugal-pumps-triac-thyristor-etc-need-help-from-controls-electrical-people/12?_=1739427920720 . That pump is the smallest mag drive I could find, suitable for "chemicals" (so made from polypropylene or PVDF) - and it's around 6 L/min flow, so two orders of magnitude higher!

They are made from polypropylene, which is ideal for our case. Right now we are trying to figure out how to control/modulate their speed/flow, since they came with cheap AC motors.

---

Again, warm welcome to the project and please let us know how you get on with the dev kit! We are still finalizing a suitable first chemistry for stable cycling for the kit.

We would like to design our flow frame using an open-source toolchain like we have the kit. I can make a rough design of the flow frame in FreeCAD.

We will need to do some FEM and fluid mechanics simulations on this design, in addition to shunt current calcultations.

For fluid FEM and pressure drop/flow distribution, we could probably use OpenFOAM. There is a plug-in workbench to run this directly in FreeCAD, I have messed around with it and gotten it working on my laptop. I can run simulations in less than a minute for the small cell barbed flow frame:

FreeCAD + OpenFOAM (CfdOF)

This is something I did up quickly for the dev kit flow frame (2 sq cm).

Minimum working example for CFD that produced the above simulation results: https://codeberg.org/FBRC/RFB-dev-kit/media/commit/215285a9ef93c7eaaf68583418111a83b9d7b0e7/CAD/CFD.FCStd

Mesh:

Flow distribution:

Pressure drop:

For shunt currents (and pressure drop too), we are probably best off manually calculating them first. They will depend on:

1. Electrolyte conductivity
2. Electrolyte viscosity
3. Flow rate
4. Manifold cross-section area
5. Flow frame geometry
6. and more...

For the "large-format" cell, we'd like to target as large of a geometric area as possible, and ideally have the flow frame be useable both for single-cell and stack testing. This means it needs to possess adequate internal fluid manifolds and flow diffuser/spreading geometry. We must also consider shunt currents once we progress to stack testing, so we'd like to design the cell with that issue in mind upfront.

We plan to start with a flow-through design, as it is much, much simpler to design and manufacture than flow field-based approaches.

This thesis has some helpful figures - I haven't read it yet myself, but it looks quite useful. The author is now a professor at University of Padua.

https://global.discourse-cdn.com/free1/uploads/fbrc/original/1X/0f77c4b4b752d2c44bc56bc870b2ac407eb9f7ca.jpeg

Basically we want to make our own version of this. Ideally we could prototype it with polypropylene FDM printing... but in any real application it would be injection molded.

A good image showing the path of a shunt current, which leads to a drop in energy efficiency as well as uneven current distribution (and possibly plating, for hybrid RFBs):

Image from University of Padua researchers: https://iopscience.iop.org/article/10.1149/MA2024-0217mtgabs

---

Right now, we need a flow frame that:

1. Has the correct geometry for flow-through graphite felt electrodes (and possibly polymer felts/spacers, for hybrid chemistries)
2. Doesn't leak
3. Distributes flow evenly through the felts
4. Offers adequate pressure drop and shunt currents when implemented in a stack

References

PhD thesis: https://www.research.unipd.it/handle/11577/3422708

Optimization paper: https://doi.org/10.1016/j.electacta.2021.139667

Flow field optimization paper (we don't have flow fields, but the simulation framework/procedure is interesting): https://linkinghub.elsevier.com/retrieve/pii/S0378775321009563

Importing a previous discussion from our old forum:

---

kirk 1 September 23, 2024, 2:05pm

For example, for residential storage applications:
• 1 kW / 10 kWh
• 70% system roundtrip efficiency including balance-of-plant
• Chemical safety risk no greater than an equivalently-sized lead-acid battery bank
Determining this ahead of time will help us chart an efficient path to get there, and
hearing from real-world potential users will inform our R&D plans!
What applications would you use a flow battery for, and what performance metrics
would you desire?

---

Sanli 2 September 25, 2024, 6:56am

I think 70% is a good target, but for the mid-term. The existing commercial RFBs also
don’t provide 70% and the low hanging improvements are in the system level (tanks,
pumps, etc) before dealing with stack and electrolyte.
As for the safety, it would be good to look at the existing standards for volume of acid
beyond with a separate compartment is needed, I think it is about 250 litres, but I
might be mistaken.

---

kirk 3 September 25, 2024, 7:28am

Good point about the secondary containment requirements for the liquid. This
regulation probably differs around the world but would be good to look into. There
are double-walled plastic tanks that count as secondary containment, but these are
probably harder to get. For our first iteration it would make sense to pick a volume
that is more manageable. Putting a big drip tray under the whole setup would be
sensible too.

---

Dogpoo 4 October 7, 2024, 2:20am

Maybe want to be thinking about insulation, or even the option of a heating element
if potentially users are going to stores the system somewhere like a utility room,
cellar or garage where temperature can vary.

---

kirk 5 October 13, 2024, 9:02am

I’m not sure at which temperature you’d need insulation. The battery will self-heat to
some degree during operation.
Also, we will likely have an ambient temperature sensor and/or electrolyte temp
sensor in a real system. The concentrated electrolyte should lower the freezing point
of the electrolyte below that of water, but we’d have to do real tests with an
environmental chamber or similar to really understand the viable temperature
window.

---

Dogpoo 6 October 13, 2024, 9:26am

Good. Etc etc etc and so on and so forth. 20 characters.

---

julianstirling 7 November 11, 2024, 2:45pm

Secondary containment will be a “fun” thing to try to enforce for the open project as
others start to experiment. I have found in the past that seemingly unneeded things
often get ignored. Lots of caution messages explaining the need for secondary
containment probably go a long way towards this.

---

kirk 8 November 12, 2024, 8:36am

Yes, good point, we will need to have lots of caution messages all around the
documentation. For R&D purposes, we want people to be able to conduct tests using a minimum amount of materials, but there are always chemical risks no matter the
quantity. For our benchtop system, the volumes are so low (around 10 mL total) that
it shouldn’t be an issue, but for stack testing, we’ll have to spec an option, the more
affordable the more likely people are to use it. And add some images and warnings of
examples of chemical accidents where lack of secondary containment caused issues.

---

julianstirling 9 November 12, 2024, 9:14am

Yeah. I have taken a quick look through he docs I see there are quite a few.
While it wasn’t safety related, we used to find for OpenFlexure everyone ignored on
the printing page that the optics module should be printed in black. This changed
when we started both explaining why and adding it to the checks:

then when you come to assemble it:

It seems that the short bullet point sentences really helped people not miss what used
to be in longer form text. The information symbol link to more detail.
Dozuki had a really nice presentation about how to do documentation that really
helped me. I’ll see if I can dig it out. Or if not remember the key messages.

---

pinecone 10 January 27, 2025, 8:15pm

What applications would you use a flow battery for, and what performance metrics would you desire?

I’m thinking about intra-day arbitrage of market-priced electricity. The ratio of
average consumed price to lowest daily price (night) is consistently about 10x or
more here, so there is some potential to save money by time-shifting consumption.
Do you have a cost estimate for a 1 kW / 10 kWh system (like above)? This would be
more than enough to shave off the price peaks for a single household.

---

kirk 11 January 29, 2025, 11:44pm

Welcome to FBRC, @pinecone !
I don’t have a straight-up answer for you right now. We don’t have a cost model or
estimate yet but we would like to and will have to build one in time. I did some basic
cost modeling in the past but for different chemistries/systems. It wouldn’t be too
hard to have a simple spreadsheet for back-of-the-envelope style calculations.
Daniel posted on another forum some basic calcs for a larger system that someone
had asked about: https://diysolarforum.com/threads/my-adventures-building-a-diy-zn-i-flow-battery.69145/post-873727

My adventures building a DIY Zn/I flow battery | DIY Solar
Power Forum
Quoting him here:
DIYrich said:
What is the usable energy of 30,000 litres?
What is the cost of 30,000 litres?
I’m wondering if it can be used for shifting summer production to winter
usage.
15,000 catholye + 15,000 anolyte at 35Ah/L would give you 525kAh which at a
mean discharge voltage of 1.23V would give you 645 kWh, this is 0.645MWh, so
very massive system. At 1mL per cm2 of electrode area you would also need to
have 1500 m2 of electrode area, which at a standard 25cmx25cm per cell would
imply having at least 24,000 cells. This is a massive system. Probably a couple of
containers filled with stacks of cells to process what is literally a pool of
electrolyte. Since the energy efficiency is 70-75%, you will need to put at least
0.86MWh in to get that 0.645MWh out.
At bulk prices of:
ZnCl2 - 1700 USD/ton
4 of 6
KI - 2900 USD/ton
NH4Cl - 450 USD/ton
For 30,000L you would need 8.17 tons of ZnCl2, 3.20 tons of NH4Cl and 19.9 tons
of KI. The total cost of the salts would be 32.1K USD.
The above doesn’t include pumps, tank costs or cell costs. Note that since no ion
selective membranes are used, this is going to be significantly lower cost
compared with a Vanadium based system. Big systems have significant additional
issues - for example pumping efficiency becomes a huge concern - so I’ll have
clearer costs for you once we implement the first 25x25cm cells.
We are however FAR from anything at this scale. Right now we are focusing on
the small scale. Once everything is optimized the costs for larger scales might also
drop further. Hopefully significant improvements in the energy density are still
possible since the solubility does allow for much higher densities.
Zinc-iodide isn’t the cheapest possible chemistry, but it’s working decently as a
starting point.

Daniel estimated 32.1K USD for 645 kWh of usable energy, so scaling that to 10 kWh
is about $498 in chemical cost. The cost related to the 1 kW power component, the
stack, requires more involved calculations, but there’s nothing particularly expensive
component-wise in the stack (like platinum or gold…)—if you’re not using an
expensive membrane. It’s mostly plastic and graphite (in various forms) with two
copper plates and some tie rods, but the design and control of it is very important,
which is what we’re focusing on now.
Peak-shaving and intra-day arbitrage seem like great opportunities for RFBs though,
that is definitely something we’d like to eventually see happen!

---

pinecone 12 January 30, 2025, 9:08am

Thanks for the info. The chemicals are surprisingly expensive.
My very rough estimate of the break-even cost for a 10 kWh peak shaving system is
about 1000 EUR, which does not leave much for the rest of the hardware after
chemicals, even when allowing for DIY construction, 3D printed parts etc.
This is a cool project though, best of luck!

---

kirk 13 January 30, 2025, 10:51pm

Thank you! And yeah, we are trying to develop a functional system, but it won’t be
economically competitive as a DIY build for quantity=1—there would have to be a
group buy or a business set up to buy chemicals in bulk, flow frames injection
molded, etc. This project is for the R&D to get to a viable system, if we get to a
functional system the idea is the project output’s are licensed for commercial use,
and a real business could make it more affordable at scale.
A kit build may be possible if there was a supplier for some of the specialty
components or similar.

Very relevant work by Trovò: https://www.sciencedirect.com/science/article/abs/pii/S0306261920300441

Seems they basically used a lookup table based on flowrate and SOC.