Light assist Season 3

tkfoil · June 18, 2026, 6:34am

I would keep those cells in case of 12s1p in a single row for two reasons:

you are adding drag in the water with your configuration
the batteries help with counter balancing the weight of the housing and motor behind the mast. The shorter the battery the less counter balance you have. I have found that 12 cells are not able to completely counter balance the weight of the motor. Maybe a 12s2p config does and that’s a config where a second row makes totally sense.

Your motor cable guide is also adding a lot of drag. If you want to attach the motor on the mast with cable ties (kind of a permanent attachment) you can either:

tape the motor cable to the mast (would choose this option for the beginning and optimize later)
guide the motor cable internally through the mast (best option regarding drag, but you need to modify the mast)
create a smaller cable guide along the mast (there are many examples of it)

mateomodar · June 18, 2026, 9:12pm

Yes, this makes all total sense to me! My idea was that it would be easier to asemble and disassemble. To have it in one part which i can just get off when the battery is dead to continue pumping without. But i realize that drag is a huge factor so ill try to come up with a more suitable design. Id love to realize this somewhat modular though to test different setups with the same motor. Also the battery layout ill keep simple due to the drag. Guess i was a bit overmotivated to design some new features, lets keep it simple

Fredo · June 23, 2026, 12:50am

Hey mate, could you share the STL of the prop in TI, or is it the design in GitHUB? Looking for good design to pair with my 5085 motor, thanks

Bzhwindtalker · June 24, 2026, 7:02am

I can not share it directly but file should be available online somewhere by now.

System is smooth, Ti blades works , did not test the spare alu prop to compare. I printed a second battery box and will assemble a pack with bak 50D or lishen 50pt cells.

I had a charger failure while on holidays, so I bought a new power supply that does 12A, for 20-30 min charges

tkfoil · June 24, 2026, 7:54am

I also had problems with my power supplies. I found a 16.5A power supply for gaming laptops with an xt60 port, so it connects without an adapter cable to the iCharger

Very with happy with that one so far!

Do you go for 12s2p this time?

tkfoil · June 25, 2026, 7:46pm

After many session on my DW Board I tried the 4’2" prone board with the Veloce H XXL today again. This time with the FD prop. I was doing flat water starts and had absolutely no problems getting up on to the board.

With that prone board throttle and pitch management become way more important. On touch downs the board more likely sinks and with that high positioned prop those touch downs happen a lot to me at the moment.

Also it seems like the battery got to its limits. I took a look at the logs and I was not dipping into the voltage cut off (start @35V, end @32V). At one point it only delivered max 20A while being at 44.4V resting voltage. This session I was feeling some heat when touching the pack. Probably the battery overheated. So I decided to end the session.
With the DW Board I did not experience such heat and power losses.

Battery Amps:

Battery Voltage:

Interestingly the battery amps peak way over the 45A limit I have set.

Edit:
Thought a while about it and do not think anymore it is a battery issue. The voltage drops are also significantly reduced. So I think the VESC did not request more amps even though I was on full throttle. Also that it always cuts at 20 amps is an evidence for a VESC issue. It might be a 75100 VESC issue related to the problems observed in this thread:

Bzhwindtalker · June 26, 2026, 5:43am

Yes, I am with you and dont belive its a battery issue, it is too repeatable. Maybe the lvc got re-programmed?

tkfoil · June 26, 2026, 7:12am

The lvc has the correct value

I think the time has come to replace this VESC. The Harry M VESC has already arrived and I am waiting for cable lugs.

It just needs to hold up until tomorrow as I will be wake thieving with friends
I will go on the water today and check if there are any fault codes. Unfortunately those do not get logged and there does not seem to be a setting to send them to the VESC Express for logging.

Jan12 · June 26, 2026, 11:38am

If you connect with VESC tool after the session without turning it off, you can read out the error codes without any extra hardware.

tkfoil · June 26, 2026, 12:37pm

Yes, it is just annoying. Should end up in the logs by default.

Bzhwindtalker · June 26, 2026, 1:06pm

I have a vesc express and SDCard in my build, but did not manage to retreive data by wifi/bt so far and make a correct setup for auto logging. Need to look into it

tkfoil · June 26, 2026, 1:30pm

It is not very easy. Some SD cards do not work. Then they have to be formatted correctly. You need to install a plugin on the vesc express where you need to setup correct IDs. On the vesc you then need to select the data points you want to pass to the express, but it is not named correctly, so you would never know. Then when everything is setup correctly it might still fail. So restart the VESC and voila, it works

Took me an hour or longer to figure it out with Claude haha

I can send some screenshots later of the setup in the VESC. Need to go on the water now

Ah aaand taking a look at logs of a real session does not work either via Bluetooth or Wifi because the download of the logs takes a couple of minutes even via USBC connection. Bluetooth and Wifi are so slow and unstable on the VESC Express that it just freezes while downloading. So always a mess with tearing up the Magic Gel and repairing it again to only connect via USB.

tkfoil · June 26, 2026, 8:42pm

Today I was testing with the dw board and I did not have any outages. So let’s see when it happens again.

@Bzhwindtalker
I was looking into my logging setup and did a bit more research this time and found a manual from MKS for VESC Express: https://vesc.com.ua/MKS%20Express%20Manual%20(eng).pdf

Found out that I did it wrong. In the manual they emphasize to connect via USB on the VESC when setting up logging. I was connected via WIFI which led to a couple of issues and ultimately to get it running I have installed the LogUI Plugin on the VESC Express instead on the VESC itself. It works this way but it logs only values that are available via CAN. When installing it on the VESC the local values get logged including “faults”. So I was complaining unnecessarily about this.

Anyway, I repeated the setup documented in the manual and it works now. But, there are some Bugs in the UI which make you believe that the chosen logging settings do not get saved to the config. I mean the settings from here:

When reconnecting or switching between VESC and VESC Express the form does not reflect the stored config values. In my example “CAN Values” and “Start logging at boot” become unchecked.
Luckily you can verify the values somewhere else. Go to “LispBM Scripting” and then at the bottom in “Console/REPL” you can print the values:

Commands:

EEPROM read-out (LogUI persistent settings)

Run these in the Lisp REPL (VESC Dev Tools -> Lisp tab):

(eeprom-read-i 0)   ; ver-code
(eeprom-read-i 1)   ; can-id
(eeprom-read-i 2)   ; log-at-boot
(eeprom-read-i 3)   ; append-gnss
(eeprom-read-f 4)   ; log-rate
(eeprom-read-i 5)   ; log-local
(eeprom-read-i 6)   ; log-can
(eeprom-read-i 7)   ; log-bms

| Addr | Setting     | Type | Expected            |
|------|-------------|------|---------------------|
| 0    | ver-code    | i    | 239                 |
| 1    | can-id      | i    | logger CAN ID       |
| 2    | log-at-boot | b    | 0 / 1               |
| 3    | append-gnss | b    | 0 / 1               |
| 4    | log-rate    | f    | e.g. 10.0           |
| 5    | log-local   | b    | 0 / 1               |
| 6    | log-can     | b    | 0 / 1               |
| 7    | log-bms     | b    | 0 / 1               |

Note: only these 3 group flags (log-local / log-can / log-bms) are
persisted - there is NO per-field persistence. If ver-code != 239 at
boot, main() calls restore-settings and wipes everything back to
defaults (can-id -2 on Express, 2 on ESC).

The printed values have the type appended (for anyone getting confused by the values):

integer values like 0 or 1 get printed as 0i32 and 1i32
float values like 10.000000 get printed as 10.000000f32

Bzhwindtalker · June 27, 2026, 7:58am

Great that you got it going on the small board with just some setup tuning and skill building

This also emphasis some expectation management for new builders / riders, where one can expect to go direct to the “pro” setup, small board, small foil, and have a bad time due to settings/build/skills not being ready. So first start reasonable with a big board and foil to get it all sorted out is not a bad advise to give

Thanks for the feedback on VESC express logger, I will look into it tomorow.

Bzhwindtalker · June 27, 2026, 3:44pm

Printing that for my wife to try efoiling again, 200mm extension with brace to mast, will use a flex-starp nano to hold it.

With my printer and spool changer living outside, at least I know PETG is fully dry by now in france

tkfoil · June 28, 2026, 8:12am

Are you sure it will hold up? That is a pretty big lever on the connection at the housing, even with the attachment on the mast.

Anyways it is a good idea to get the wife excited for this sport I am still out of luck with that

Bzhwindtalker · June 28, 2026, 3:14pm

It worked, but the prop failed the minute my wife was ready to test it in the water . I was able to find the missing piece, very clear water.

Jan12 · June 28, 2026, 3:59pm

Is that your own hub design?

I wanted to fix this a long time ago.

Design it so the 3 parts that are now separated print separate in correct layer direction, extend them a bit so they intersect with the base and glue them in. Alternatively also put a small screw through the base into that part.

That should eliminate all issues with hubs like that.
I just do not have a good quality base to start off.

Bzhwindtalker · June 28, 2026, 4:03pm

Hub from @foilstate, printed in ppa-cf, 5 sessions. Very lucky that it failed in shallow water.

Will get pa12 version to be safe

tkfoil · June 28, 2026, 6:36pm

lol

Maybe just make 2 or 3 holes through the hub and glue some screws in as a quick fix until you receive pa12 hubs. This should be enough to withstand sheering forces regardless the print orientation.