Getting little torque and speed from a big batter and big motors

I recently assembled what I thought would be the ultimate “urban commuter board”

I used 12S4P batteries with a 200a 6.6 Filipsky VESC and two 8072, 175KV leopard motors. Powering a 15T / 68T (5M) belt and pulley drive train with 250mm pneumatic tires.

When I initialized the VESC it read 80amps coming into both the left and right wheel and the wheels spin at 30+ mph

Yet, when I ride it, it can barely start my from a standing start and fails on even relatively modest hills, It simply has no torque and on a flat, paved surface it runs comfortably at ~12mph, The battery is fully charged, Has anyone else experienced something like this before? When I take this same motor and ESC and attach it to some 125mm gummies, it generates so much torque that it lays rubber. I just can;t figure out why I can;t get a stronger reaction when I use the bigger wheels.

This is about my 5th build, but my first with a 12s battery and wth the larger 8072 motors

Any help would be greatly appreciated. Thanka you - I;'ve attached a picture so you get a sense of the size Link to Picture of Board Other Pix of Board

There are some things we need to know. First the picture didn’t upload. Second, what cells do you use? 4p of 18650s isn’t a big pack and will cause a huge voltage sag with proper current settings for those motors. Third, which kV do your motors have.

One problem could be caused by not having sensors on the motors. HFI could help here. Not sure how good it will work with 80xx Moria thou.

1 Like

@Andy87 THANK YOU! The motors don’t have sensors. I’ve attached 3 pix and a video. This is my first build with the Leopart 8072, 175KV motors. They are sensorless. The battery (Samsung 18650) and ESC, I’ve used before, in fact the build right next to it has the same 12s4p 18650 samsung batter. Here is the LINK to the Leopard Motor Description. LEOPARD MOTOR LINK

In a post below the 3 pictures, I’ve also posted a video of the motors spinning unwrighted. With no weight on it, the wheels spin extremely fast and demonstrate the acceleration I would expect. I’ve analyzed the motors, pulleys and the wheels mechanically to ensure their is no rubbing or friction, All seem to spin freely… They each are drawing ~80 amps (160a total) VESC is rated for 200a. What do you thnk? THANK YOU!!

LINK TO PHOTOS

1 Like

Please find the attached video for your consideration. As you can see, the unweighted motors generate a lot of speed. When I formatted the VESC, both motors were each showing 80 - 81 amp draws from the battery (160a total) The VESC has a 200a continuous rating

I know the leopard motors. They are good motors. You can do a small test by your own. If you in stand still, try to give your board a bit of a movement when you give throttle. As soon as you roll like 1m the motors should take off and give you torque. If it is the case then it’s definitely because you do not have hall sensors on your motors. You have 3 options to make that better.

  1. you always kick push from stand still. That’s difficult if you use bindings thou.

  2. you install AS5047 encoder

  3. you will give HFI a try. Here some tutorials on how to set up HFI.

1 Like

So, without question, the motors start with an easy push. However, they don’t really engage any differently after that. The max speed I’ll acheive on a flat surface with be 10 - 12 mph (and it takes me a long time to get there ~100 yds) I even had someone run me along “bobsled style” to get me to ~8mph. Again, the same result, they give me a nice pleasant 12mph, then fail after even a short 10 degree incline.

The one thing that I notice is the batter is showing deep fluctuations. After the hill, the batter will give me readings of 25% and then quickly recover to 90%+ Does that indicate anything to you?

AND THANK YOU FOR YOUR THOUGHTS. I’ve been banging my head against the wall with this one for the last week!

Great videos! I’m going to have to watch it again slowly (I don’t have an electronics background, so I generally have to watch these things with another window open where I’m constantly pausing to ask "What is FOC and what does that control?)

Still - I definitely subscribed to your channel, Although, I think that this issue is secondary to the bigger issue of not getting nearly enough speed or torque while I’m already moving. Still, the HFI tutorial was great and is relevant to a problem I’m having on another skateboard I built. (the FOCBox piece is likely relevant as well. I’ve got one board where I’m using that thing and having several issues (namely around the regenerative braking) Thanks Andy

That’s called voltage sag :sweat_smile: But shouldn’t be that much if you chose the right pack and right settings. How old is this battery pack? Could be old cells, could be faulty cells in the pack or could be that you just asked too much current from your battery pack.with this motors I personally wouldn’t start below 8p of high discharge 18650s like 30q or vtc6 cells.

Yeah… I did some research, the other builds that used the 8072 motors all worked from 10s9p or 12s8p. I’m mad at myself for not paying closer attention. I’m gonna repurpose this set-up to some 6384 motors and see if I can’t put together a 12s8p battery. I’ve been working to build this one for almost a year now, so I’m frustrated that I missed something so obvious.

Separately, I’d really like to get a better framework for understanding the VESC. I was wondering if you might have any suggestions. I’ve watched all of the online classes sponsored by VESC and they do a good job of explaining the “what” I should do. But, I’d like to find a way to get beyond that, to understand the “why” it’s designed that way. I’ve been building boards for a couple years now, and I find I can stumble my way through a solution using YouTube and some of the blogs. But, I really need a better fundamental understanding of the way these things are engineered.

From your videos, you seem to have a really good knowledge of both “what” your doing but also, “why” your doing it. I was hoping you might direct me to a resource as I think it could really help me build some better skateboards and trikes. Please let me know any thoughts. THANK YOU AGAIN, YOU’VE BEEN A BIG HELP!!

For vesc related questions you might be right in the vesc project forum.

You’re staring down a deep dark rabbit hole my friend!..

Before you begin your learning journey, ask yourself two questions, then choose a path.

  1. Do I really need a deep knowledge of embedded systems, motor control physics, etc
  2. OR, Do I just need to know how to build great electric vehicles, using existing systems?

If your answer is #1

You can do formal education in various fields to learn about the topics. Remember, VESC is just a brand name for a product, so there won’t be any course material dedicated to VESC.

You would need to start by learning about core principles, maths, physics, electrical engineering, programming etc. After that, you may have a better understanding of motor control theory, and hopefully, start to make sense of the hundreds (probably thousands) of variables that must mesh together to make a useful motor control algorithm.

That’s all theory/physics stuff, Now you need to embed the algorithms into hardware in the form of firmware, Made up of thousands of lines of code that’s must work within the constraints of the hardware. (This is why VESC is so great, it works well across a wide range of hardware)

Finally, the user interface software called VESC tool to allow it all to be programmable on your phone or computer all whilst trying to make the extremely complex functionality all seem somewhat “user friendly”

Truth be told, 98% of the people on this forum have no idea how any of it actually “works” - Most people will never use the VESC tool and the hundreds of user-programmable parameters it has on offer.

The reality is, you don’t really need to know anything about the complexity & engineering to use it & get good results.

That’s really the job of the engineers, Make complex things simple for the masses, Agree?

:information_source: If you want to read more about the engineer, Ben Vedder, who created VESC you can read how it all happened at his blog: April | 2012 | Benjamin's robotics


This leads us to question / answer #2

How to build really good electric vehicles?

A good electric vehicle can be described in many ways, everyone has different requirements etc, I don’t have time to go down that rabbit hole in this post.

You often hear people talk about form & function, I think these are two good things to think about before your next build.

In order to wind my already lengthy reply up, I’m going to ask questions about function only!

  • Do you want your board to go up a hill? [Mechanical Torque, Battery & Motor Current]
  • Do you want the battery to last for 20, 30, 40 miles etc? [Battery Design, Cell Selection]
  • Do you want ball tearing acceleration & top speed? [Battery Voltage & Current, Mech/ Drive Train & Gearing]
  • Do you want a lightweight board you can carry around in a shop? [Design constraints; Single vs Dual Motor, Motor Size, Battery Size, Wheel Size]

I personally believe learning how to build your board, confined by the constraints of the hardware you select, to achieve the FUNCTIONS that are important to you is way more important than spending time learning all the complex theories behind how the VESC actually works. Most of that knowledge is not actually useful. However, you might be into that so don’t let me persuade you from learning, go nuts.

In the VESC tool the most important things to learn are;

  1. How to connect the motor & do motor detection. Specifically, does the motor have shall sensors, if so, what pins are what.
  2. How to setup battery current limits. “Max Battery Current” MUST be matched to the batteries capabilities, set this too high & you will harm the battery.
  3. How to setup Max Motor Current limits. Small motors will burn up with too much current.
  4. How to setup the remote controller & brake so it is safe for your weight, riding skills etc.

Fortunately, there is a motor setup wizard in the VEC tool, it basically does 1-3 for you. However, it’s not perfect, so you really should learn about that stuff. Trial & error is a great learning tool, you can tweak these settings by a few points each time and do real-world tests to better understand the impact on performance.

:bangbang: WARNING :bangbang: When “testing” Never exceed the safe limits of the battery, esc & motor. If you don’t know the limits ask the manufacturer.

All the other stuff in the VESC tool can basically be ignored, it isn’t going to make your esk8 perform better. Because once you get the setup dialed in, with safe parameters, you won’t be using the VESC tool much at all. When I say “Dialed In” I literally mean the correct settings that are determined by the specifications of your hardware. There are no secrets VESC settings that can alter laws of physics.

On a side note; I think the VESC tool desperately needs to be reworked by UX engineer, or simply a button to hide all the menus that aren’t commonly used. The tool is extremely confusing for beginners primarily due to the sheer amount of parameters, total information overload, that hits you in the face when it first opens up.

TLDR: Don’t think to yourself that learning about all that extra vesc “stuff” is going to make you better at building esk8, it probably won’t

Anyway, feel free to keep asking specific questions, about anything, bit by bit you will start to understand the VESC Project and why it has been an important innovation Esk8’s

3 Likes

That’s a very ignorant view on things. Why to stop people from learning new stuff if they are interested in something. Even if your TLDR would be right, it’s never bad to know how things are working.

1 Like

I would nver stop amyone from learning, but i am also a realist!

I know when i started in esk8 my thirst for knowledge sent me down many rabbit holes, wasted huge amounts of time on stuff that wasn’t really important.

I think it’s a good idea to warn people that they might waste their time. Time is precious!

But i would never stop anyone, that’s why i wrote this.

2 Likes

dog this build is SICK wish I has went for something like this at first…

is the cable routing no issue? I feel like they could be routed for less stress.

For what it’s worth, speaking as someone without an engineering background - my interest in electronics is really limited to making better skateboards. I wish I would have started with a “cheat sheet” - as I’ve wasted a lot of time and money stumbling into traps I should have been able to avoid. There’s so much good information out there from well meaning contributors. My challenge is the sheer breadth of that information can be distracting rather than empowering. After 2 years and a couple builds, I wish I’d had started with a simple set of “do this” (for example telling me that; a 36V battery can actually be MORE powerful than a 44V battery, in these cases or why my 2100W Meepo board can absolutely destroy my 3300W board - even though everything else on the two boards is the same)

I definitely appreciate what you’re saying, though. I wish I had the fundamental understanding of the physics behind these systems. But, in this case, I want to build a big, heavy skateboard with 10" wheels and a lot of torque in the next 4 weeks and the simple answers have been great. Hope you have a happy holiday season!

1 Like

Thanks for your kind words! It’s been over a year just to find and assemble the random parts. It’s been a ridiculous amount of dumb mistakes for me (If you want, I’ll give you a list of exactly what I got, where I bought it and what I had to get machined at a local auto shop.) The funny thing is the hardest parts to find were the tires.

Unfortunately, getting another 50 battery cells is gonna ramp up the cost on this thing. But, I gotta finish it. (I’m pretty sure Colonel Custer said the same thing when he saw the Little Bighorn village…)

Thanks again for your comment and hope you have a happy holiday season!

1 Like

Also, how could I change the cable routing to get more power?

Thanks again!

1 Like

oh haaha more power, it’s funny I was writing a comment about how your 4P battery is the weakest link in your system at the moment. I would recommend getting a EV module from batteryhookup.com to slap on top when you want a ton of range. they just sold out a good 12S one but they have more coming always.

I changed it to be about the motor wires looking like they did a lot of flexing which can make the motor wires get cut by shaft bits around the cable exit on the motor stator. You can put some glue there to reinforce if you want.

1 Like

Battery Hookup’s prices look really good. I’m gonna ask a couple really dumb questions, but - is it as simple as just getting their product and shrinkwrapping it? Also, the shape will be a little differentt because I want to do a “single row” to keep the clearance high/. Do you print your enclosures? If you do, do you have any advice for doing that (in the past, I’ve always been able to find a design on Thingverse and sent it to the printer. But, I’m gonna be off-script here. Any thoughts on how to build the file for the printer? THANKS!

1 Like

That’s a good thought on the wires. Frankly, the last problem I had with the board (before this problem) was that one of the motors kept changing direction. Turns out one of the phase wire connections was moving around. I was gonna just pull some length out of the wire… but, then I rode it and realized it didn’t have enough torque. So, I’ve had to refocus… Like I said, it’s been a year on this thing

Thanks again

1 Like