Neu 8057 6kW left hand drive hardtail with vesc and encoder

[district9prawn]

Finally got around to mounting a new encoder board on the bike.

I tried using position control to hold the wheel at a stop but wasn't able to get the transition nice and smooth.

Instead I configured my regen lever as a reverse throttle and set the reverse erpm limit very low. There wasn't an option in the software to use a second throttle as reverse but only a few lines of code needed changing. This enables full braking torque down to zero. Plus I also have reverse now to help move the bike around :wink:

I am happy if regen does slow me down to around 5kmh and so i never did use active braking on my adaptto controllers (only for tests) due to the fact that it does take energy from the battery (to produce reverse torque).

I never knew adaptto had active braking. Does it work smoothly?

 

[maxxximatoze]

A few more pics of the completed boards for those that cant get enough controller pics.

Hi,

What is this compact vesc ? Where can I buy it ? It is a very intersting model with the two boards sandwich, can you give me its specs pls ? Thx !

 

[district9prawn]

Hi,

What is this compact vesc ? Where can I buy it ? It is a very intersting model with the two boards sandwich, can you give me its specs pls ? Thx !

Check out team triforce uk. Assembled boards and bare pcb are for sale on the website.

I quite like the dual board design for repairability. If mosfets burn violently and wreck the pcb only the power board needs replacing. A lot easier to solder.

 

[neptronix]

I love this build.. especially the ultra steep single stage reduction.
Twice the motor of the cyclone and half the potential drivetrain issues.. rear direct drive FTW.
Please get us another video of you flogging it to top speed

I notice the motor is not completely silent ( i suppose this is to be expected ), yet you're running a VESC variant controller.
Is this one of those open cased motors, by chance?
( i'm imagining that the motor having cooling holes might make it emit more noise )

Also, that A200S controller is a barking mad little controller.. it does seem that all the innovation these days is going on in the eskateboard space..

 

[flat tire]

Badass mid drive found, ordered motor and a couple vesc to test out.

 

[district9prawn]

I notice the motor is not completely silent ( i suppose this is to be expected ), yet you're running a VESC variant controller.
Is this one of those open cased motors, by chance?

Yea the motor has cutouts front and back. Its quite audible even when coasting at speed. In the video you can just make out the motor whine as I coast down steep hills.

Badass mid drive found, ordered motor and a couple vesc to test out.

Cool which motor?

 

[flat tire]

8057

 

[madin88]

8057

Which kV?
Hopefully you get a motor with balanced bell and better bearings so you do not have to deal with same problems everyone else had in the past.
And if you buy a long shaft be aware that they all will come with 13.8-13.9mm diameter instead of 14mm.

If someone would ask me if i would recommend those large outrunner from Neumotors my answer would be that from a performance or technically view the 80xx series are good motors, but the big quality issues makes it really hard to recommend them as it cannot be used out of the box, and the other thing is the support which is like talking to a wall..
What a bummer!

 

[district9prawn]

8057

By the way if you need a shaft I can sell you one as I have some spares that I had machined in China. Its basically a CNC version of the one in my first post. Made from 304 stainless and with 12mm at the output for mounting a sprocket.

 

[flat tire]

Which kV?

75 kv. I'll run it on a bac4000 with emulated halls for deadstop performance. Those magnetic encoders are really trick. I would get one of your custom shafts but I'm setting up machine tools so I can make that stuff myself.

 

[district9prawn]

Haven't updated this thread for a while but I haven't made any changes to the bike other than switching to 12s as that pack has the high capacity cells now that I need to travel a bit further every day. Without field weakening top speed is just under 60kph which is still ok.

Recently the levers on my mt5 brake was getting spongy and I could see a bit of fluid weeping out of the lever. This seemed crap as I already replaced the plastic lever about 6 months ago for this exact same problem. Calipers were fine and I've found the magura calipers to be one of the most reliable ones on the market. I read on some German forum that shimano levers worked fine with magura calipers so decided to try the "Shigura" mod.

The oil that leaked on to the handlebar and clamp.

Bleed blocks in the calipers

Cut off the old olive and barb

Since the hose is still magura I used the magura olive and barb. The compression nut and rubber boot are the shimano one to fit the lever.

Driving the barb.

Ready to bleed.

It seems to work pretty well! I'm not a huge fan of the shimano servo wave and how abrupt it seems to make the onset of braking power but it beats the leak prone plastic magura levers.

 

[district9prawn]

Did some stuff on the bike so time for an update.

Had a chain drop recently. My fault as I was in a rush to adjust chain tension with the chain roller. Didn't loctite the new bolt and it came off a couple of weeks later. Unfortunately the chain jammed after falling off and the end of the shaft got bent!
Hard to see in the image but it created more than enough wobble in the sprocket to make it unusable.

I have spares so all good. In the mean time I've made multiple welds over the bent part and hopefully it can be machined down to size nicely.

While I had everything apart I added a chain guide at the front and a spring loaded tensioner for the rear. It helps a lot to compensate for the slight eccentricity of the rear sprocket and its disc adapter. Hopefully it also helps with chain life as I suspect the extra vibration of the chain when the tension varies causes additional wear.

On the topic of chain wear I replaced the chain and sprockets for the first time. Its been a year and 7-8000km. I've also made the gearing even faster to 6.93 (97/14) from 8.25 (99/12). The motor still runs cool and I don't have to rely so much on the dodgy field weakening I implemented.

Bonus echidna I saw while out on a ride.

 

[Okami]

Wanted to say great topic and thanks for giving introduction to higher volt vesc as well.

Too bad the uk based company who sells its logic board doesnt offer power stage part as well.

Though from what I understand nowadays it is easy to order small component assembly in china and then add the big components on your own.

At least so it happened for latest Vesc based Focer esc (designed by shaman).

Also many nice details considering the technical part.

I also initially started mid mounted RC motor build but got stuck at motor placement and motor holder fabrication.

I intended to use 25h chain and angle grinder reducer gearbox with bewel gears (so it gave nice 90 degree turn and motor could be oriented vertically).

Sprocket also on left side, right besides brake disk (with smaller diameter than brake disk)

 

[district9prawn]

I intended to use 25h chain and angle grinder reducer gearbox with bewel gears (so it gave nice 90 degree turn and motor could be oriented vertically).

An angle grinder gearbox ebike?! That I need to see.

I might give 25H chain a try when my current sprocket wears out. The finer pitch would mean I could use more teeth on the front sprocket while keeping the sprocket diameter the same. The kart chain seems like it will last forever at my current power levels.

 

[district9prawn]

While the hardware on the bike hasn't changed since the my last update, a lot of features have been added to the vesc firmware.

Since the firmware now has separate parameters for the d and q axis inductance, I thought I would try use the Ld and Lq for field weakening.

The Ld, Lq, Iq and winding resistance are used to calculate the base speed where the controller reaches its limit and the commanded Q axis current can't be maintained. The higher the Q axis current, the lower the base speed.

The graph below starts off at my motor current limit of 200A. As speed increases, motor current drops and battery limit kicks in. Unfortunately power drops off a bit before field weakening kicks in.
Iq suddenly drops to zero when I let off the throttle, but the field weakening current is maintained and decays until a safe speed is reached.

The video is on saggy 10s so fw is frequently used and id is capped at 140A.

In the video, the calculations seem to work ok and it kicks in as duty cycle nears max and the controller is running out of steam. That was my first ride on the road and I set it on the aggressive side so I didn't get unintended braking at high speed. If you watch carefully, you can also see that field weakening remains on when I suddenly back off the throttle at high speed to keep the controller safe. It also engages during braking if necessary.

[youtube]LVQi1W7kHxo[/youtube]

 

[district9prawn]

Another video playing around with 10s and field weakening. Recorded during the day this time :wink:.

In the previous video I had 100 or 110A battery limit but I forgot to include field weakening in the calculation so the true limit ended up being more like 150A. I fixed that in the firmware for this video so the controller obeys the 80A battery limit.

Now that I have the script for the overlay figured out I have a bunch of older videos I can upload too along with the ride data.

[youtube]B3uaJiIU7VY[/youtube]

 

[district9prawn]

Finally ditched the cycle analyst! For the past few years it has only been a display and I haven't actually used any of its control functions.

Also managed to changed the gauges on one screen to show the D and Q axis currents so I can see what is happening with field weakening.

 

[district9prawn]

Tweaked the fw a bit so that Iq decreases smoothly and there isn't that odd surge in power at high speed. Motor power and voltage are now more constant in the field weakening region.

 

[SRFirefox]

Very cool! Is the field weakening firmware based on the Power Designs experimental branch of the VESC firmware?

 

[district9prawn]

Very cool! Is the field weakening firmware based on the Power Designs experimental branch of the VESC firmware?

The field weakening current is calculated based on motor parameters. I think their strategy was to adjust Id to try and maintain modulation at the maximum. I did use their way of adding a control mode for ramping down the field weakening though.

If you want to mess around with it I have a fork on github. https://github.com/district9prawn/bldc/tree/field_weakening

You will also need the dev version of vesc tool 2.07.

 

[aCeMadMod]

thank you for vecs file :thumb:

 

[Hummina Shadeeba]

Is ur field weakening for the Vesc safe to use? How much faster do you go and at what cost?

 

[district9prawn]

Is ur field weakening for the Vesc safe to use? How much faster do you go and at what cost?

I've been using it almost daily since July 2020 when I posted it so it seems ok for my setup at least.

Benjamin recently put out a dev firmware with field weakening so probably best to give that a go.

 

[Hummina Shadeeba]

Bearings with normal clearance CN (as you usually get if you buy a bearing) should be installed with a press fit, otherwise there will always be a slight radial play.
If there is a loose bearing seat as it is the case on my 8057, it would be wise to use bearings with C2 clearance, or even better C2 with ABEC5 grade (which are $$).”

I think bearings with internal clearance (c2,c3,..) are intended for when the press fit is either too tight or the motor will get hot and then with expansion compress the bearings. You’d want less clearance if not tightly press fitted or not getting really hot.
If the bearing is loose in the housing retaining fluid fills the gap

 

[district9prawn]

I blew the power board on the V2 controller this week. I'm pretty sure it was my fault. I disassembled the controller to check for water ingress as I had done a lot of riding during wet weather recently. I forgot a few screws and a week later, the mosfets under the missing screws blew

Logic board was fine and easy swap to a spare power board.