RC Mid drive project - trial/freeride style

[Ohbse]

Can't believe I hadn't spotted this thread before now! Awesome work, will be a hoot to ride I'm sure. Love the battery enclosure.

Good call on the extended swing arm - not just for the practicality of motor mounting but also it has a pretty substantial impact on putting the power to use. The wheel base of the DH is limiting factor in acceleration. Any chance you have the CAD for the dropout adapters or would be willing to machine another?

 

[Stielz]

Sure, I've got the CAD files drawn up in Solidworks. Cant get any more CNC machining done unfortunately I no longer have access to that resource.

Here are the files.
View attachment DH Team Swingarm extensions.zip

I think the brake side extension is solid but cant guarantee the strength of the sprocket side - limited material in the swing arm to bolt into due to its awkward shape. Haven't given these a decent test yet.

Is it safe to assume you already have a electric DH comp / team and having trouble keeping the front wheel on the ground?

 

[Ohbse]

Sure, I've got the CAD files drawn up in Solidworks. Cant get any more CNC machining done unfortunately I no longer have access to that resource.

Here are the files.


I think the brake side extension is solid but cant guarantee the strength of the sprocket side - limited material in the swing arm to bolt into due to its awkward shape. Haven't given these a decent test yet.

Is it safe to assume you already have a electric DH comp / team and having trouble keeping the front wheel on the ground?

Beautiful, thanks. I'll have a look when I get home and have access to Solidworks. If I go down this road I will likely build out of steel so limited dimensions less of a concern. I imagine I'll have to add some speed holes to get the weight even somewhat acceptable.

That's correct - I just posted a build thread of sorts in ebike general if you're curious. Mine's hub equipped but still puts out plenty of torque.

 

[Stielz]

Unfortunately I’ve had a setback in the project.

I basically have everything ready to go except the controller which seems to have died.

I got the Arduino 0-5V to PWM converter going no problem.

Then I got the motor phase and hall sensor combination all good, spinning the motor at full speed with around 5A no load.

The next day I was spinning up the motor again, got to about half speed then it just died for some unknown reason. No smoke or fireworks just stopped working. Now there’s no startup beeps and no voltage on the hall sensor connector.

So I need to send the ESC back for evaluation so most likely will be some time before I get that back. Pretty pissed off about that.

Now I’m contemplating buying an Adaptto Max-E controller with integrated BMS. Will make my charging a whole lot simpler, is fully packed with features, and in a very power dense package. The drawback of course is the cost – around $900 USD :|

 

[Stielz]

So this is what I’ve got for the battery.

It’s made up of the Turnigy 4S 5000mAh 20C hardcase battery packs from Hobbyking.

I have 16 of these battery packs to make a total of 1.184kWh battery capacity.

I have built 4 battery ‘bricks’ each with 4 battery packs connected in parallel with the balance leads also connected in parallel.

Initially I’ll run 3 of these bricks in series to make a 12S4P battery, then once I’ve re-wound the motor for lower speed constant, higher torque constant I’ll use the 4th ‘brick’ to make a 16S4P battery.

Also there’s room in my battery enclosure for a 5th ‘brick’, should I decide I need 20S.

Here’s how I’ve done it.

Soldered up the balance lead parallel harness

Tape the battery backs into the ‘bricks’

Parallel the discharge leads. I’m using 8mm bullet connectors. To fit the four 12AWG wires into the 8mm plug I needed to get the bundle of four wires into a circular shape like this.

Then solder on the bullet plug.

And the finished ‘bricks’. Notice the draw strings so that I can get them back out of the battery enclosure.

 

[Ohbse]

Great work on the battery! Shame about the ESC letting go. I don't have any direct experience but I've read plenty of people having difficulty getting them to run reliably. The ones that seem to have the least issues and be consistently reliable are the Castle Creations HV160.

While I highly recommend the Adaptto units, I've only used them with hub motors... I'm not too sure as to the compatibility with your high RPM RC motor. I know there's a few guys now that have had some good success running them on midmount setups utilising some of the physically larger and slower spinning options like the GNG 'Big Block' but it's a road rarely traveled at this stage.

 

[macribs]

The Adaptto can handle 70,000 erpm.

 

[Stielz]

As I understand, the C80100 has 7 magnet pole pairs so each mechanical revolution takes 7 electrical revolutions.

130kv wind with 12S lipo = 5772 RPM which is 40,404 eRPM. All good there.

Say I do a 12T Delta re-wind, thats 88kv with 16S lipo we get 5209 RPM, 36,467 eRPM. All good there too.

Regarding the CC HV160, that does appear to be the pick of the bunch for RC motor controllers and people have had good results running Astros and the likes. However the C80100 has proven to be a very difficult to drive motor. I thought having the halls sensors would give it a chance but it seems not.

I think the only way to go here is a proper EV controller, so Max-E it is!

 

[macribs]

Looking forward to see the result. I really hope adaptto proves to handle Rc motors well. Cos then we for sure will see more mid drive like yours. I am such a fan of the concept. Light weight. Nimble. Yet powerful enough to get the e-grin going.

 

[Ohbse]

I think the only way to go here is a proper EV controller, so Max-E it is!

Awesome, look forward to seeing how it runs. Would be a great laugh to get our two bikes together when you're finished. Yours is significantly less compromised in terms of unsprung weight/mass on steering but mine has a lot more thermal mass and a bit more battery. I'm guessing you win in pretty much any scenario other than long distance

 

[Drum]

Hi Steilz,

Just tried to post a reply, but it didn't work so am re-posting. Hope it doesn't appear twice now...

If I understand correctly you have added Hall sensors to you RC brushless motor.. is that right?

If so, I have a Kelly controller for sensored brushless motors sitting bored down here in Otaki waiting for me to find time to get on with my build (model No somewhere in my build thread, specs on Kelly website) so if it would help you could borrow that to try.

If the smoke leaked out you would have to replace it, but if you programmed it well within its maximums I don't think it likely to explode. Not sure if the e-rpms would be as high as you want but could get it going for you, and anything you learned from the exercise would be useful to others.
Possibly you might find it was good enough (or gave you confidence that a slightly different Kelly would be good enough).

PM me if you are interested..
Dave

 

[Stielz]

Awesome, look forward to seeing how it runs. Would be a great laugh to get our two bikes together when you're finished. Yours is significantly less compromised in terms of unsprung weight/mass on steering but mine has a lot more thermal mass and a bit more battery. I'm guessing you win in pretty much any scenario other than long distance

Would be good to try this out for sure!

The motor will be my limiting factor, probably will handle 5kW continuous-ish maybe 10kW short bursts.

Where as your cro-mo will pretty much handle all you can give it.

If so, I have a Kelly controller for sensored brushless motors sitting bored down here in Otaki waiting for me to find time to get on with my build (model No somewhere in my build thread, specs on Kelly website) so if it would help you could borrow that to try.

Thanks Dave for the generous offer! I'll wait it out for my controller to arrive because I'd hate to blow yours up. Hopefully its not a huge wait

 

[gwhy!1]

As I understand, the C80100 has 7 magnet pole pairs so each mechanical revolution takes 7 electrical revolutions.

130kv wind with 12S lipo = 5772 RPM which is 40,404 eRPM. All good there.

Say I do a 12T Delta re-wind, thats 88kv with 16S lipo we get 5209 RPM, 36,467 eRPM. All good there too.

Regarding the CC HV160, that does appear to be the pick of the bunch for RC motor controllers and people have had good results running Astros and the likes. However the C80100 has proven to be a very difficult to drive motor. I thought having the halls sensors would give it a chance but it seems not.

I think the only way to go here is a proper EV controller, so Max-E it is!

I use a infineion modded 12fet for my 8085 fitted with halls @140A battery current without problems.. A lot of the problems is where the halls are placed in these motors and it dont help..

 

[Stielz]

DC power supplies ready for charging through the Adaptto.

I’m really impressed with the 12V server PSU. It’s a 100A, 1200W power supply and very compact. I’ll use this one for my initial 12S lipo pack.

And then there’s the 48V, 2950W server PSU which is an absolute monster. This will be used when I go up to 16S lipo pack with re-wound motor

 

[upek]

Charging through Adaptto is inefficient when you use 12V PSU. You won't be able to reach 70A input due to controller heat. I suggest use at least 24V.

 

[macribs]

Charging through Adaptto is inefficient when you use 12V PSU. You won't be able to reach 70A input due to controller heat. I suggest use at least 24V.

Yes closer to battery voltage is better for less heat during charging. Charging at 70A is for many people impossible due to power outlets etc.
I think many people with good server psu around 48v 1.8kw charges a 20s pack just as fast with 35 A coil because the power outlet limits the amount of amp you can put in.

 

[recumpence]

I run Meanwell 1,000 watt power supplies as chargers. They actually peak at 1,200 watts. The 48 volt power supply can be trimmed with the POT on the board to exact voltages up to 54 volts. I run the charger output through a Watts Up meter and directly into the pack. The power cord gets warm and the 10 gauge DC wire from the power supply to the pack gets very warm. This is charging a 12S pack at 49.9 volts and 25 amps. I cannot imagine how much heat 70 amps would produce. You would need 4 gauge wire or doubled up 8 gauge at least.

Matt

 

[Stielz]

Yup definitely going to have to turn the charge current way down with the 12V PSU. Don't think I'll be using the 12S configuration for long though.

Had a bit of down time while waiting for the Max-E to arrive so had a chance to draw up a cooling fan for the motor which I will get 3D printed. First I'll post it here, see if there is any suggestions on how it could be improved?

The idea is that it will press on to that aluminium end disk with an interference fit, and can put a grub screw through into the threaded hole.

 

[Stielz]

Adaptto controller has arrived

All installed and wired up.

Now just playing around with manual settings on the controller to get it running nicely.

First try riding was disappointing. Using the autodetect settings, the motor didn’t have much torque and got very hot after only a few minutes riding up and down the street at pretty low speed.

So now I’m in the process of manual tuning the controller settings. First thing changing the hall sensor angle correction. Then the second try riding was much better. Decent torque and not so much heat from the motor.

So no I have the process of attempting to optimise the controller settings using trial and error. Not to worry, I’ll get there eventually

 

[Ohbse]

I believe I recall reading the method used for the autodetection is far from perfect with motors like yours, definitely going to be a manual tune.

What sort of power profile settings are you running thus far? I'm quite curious to see what it can take!

 

[riba2233]

Yup definitely going to have to turn the charge current way down with the 12V PSU. Don't think I'll be using the 12S configuration for long though.

Had a bit of down time while waiting for the Max-E to arrive so had a chance to draw up a cooling fan for the motor which I will get 3D printed. First I'll post it here, see if there is any suggestions on how it could be improved?

Like this :

http://endless-sphere.com/forums/viewtopic.php?f=30&t=62990&view=unread#unread

Charging through Adaptto is inefficient when you use 12V PSU. You won't be able to reach 70A input due to controller heat. I suggest use at least 24V.

Yes closer to battery voltage is better for less heat during charging. Charging at 70A is for many people impossible due to power outlets etc.
I think many people with good server psu around 48v 1.8kw charges a 20s pack just as fast with 35 A coil because the power outlet limits the amount of amp you can put in.

Well actually 70 A is on secondary side, and if it's at 12 V it's only 840 W, or 4.2 A on primary side (power outlet). So it's not a problem at all. Even 70 A on 24 V, less than average hair dryer.

 

[Stielz]

I didn't realise there were that many fan designs already on here. Which is best though?
I like the look of this one http://endless-sphere.com/forums/viewtopic.php?f=30&t=62990&view=unread#p1032550

Well actually 70 A is on secondary side, and if it's at 12 V it's only 840 W, or 4.2 A on primary side (power outlet). So it's not a problem at all. Even 70 A on 24 V, less than average hair dryer

The PSU will handle it, its the wiring that is limiting. I used 12 AWG wire with a XT60 connector for the charge port. The plan has always been to go to 16S battery with 48V PSU charging in which case I'll be able to charge at the battery max charge rate no problem.

What sort of power profile settings are you running thus far? I'm quite curious to see what it can take!

Here at the profile setting I'm using initially, not by any means claiming that the motor can handle it. Need to get the motor tuning all good before I can comment how much current it will take.

 

[Stielz]

So I was having issues with the hall sensors so need to take the motor apart to check that out. Had everything I need for the re-wind so figured I might as well do that at the same time.

What a mission! I hope I don’t have to do that again.

I managed to get 14 turns of 14 AWG wire, DLRK wind, Delta terminated. That gives 76kv which is geared to do around 70 km/h max speed. I wasn’t going for a massive high speed with this build, I’m more interested in high torque and being able to go up long steep hills without overheating the motor.

76kv translates to a motor torque of 0.125 Nm/A. Pull 150A (burst) through it, that’s 18.75Nm at the motor, 140Nm at the wheel (19” moto wheel). Should be fun.

Thank to forum member Thud for the tutorial on how to do the re-wind!

 

[Stielz]

The bike is fun to ride.

Problem is that the motor gets hot very quickly. I thought this was mainly due to the setup of the Adaptto controller but haven’t been able to adjust the settings to give any significant improvement. Can’t get any response from adaptto on their recommendations on how to set up for an RC outrunner motor.

I’m wondering if my motor is just too small. Other forum members have reported the motor being able to run at around 5kW. For me its overheating on 2-3kW. Phase resistance is 42mOhm which I thought is pretty low. But with such a small thermal mass it heats up very fast.

Its not excessively high geared or anything. Running on 16S lipo geared for top speed of 70km/h. Hope I can figure out the issue soon so I can just start enjoying it.

 

[RC_guy]

I could be blind but your sprocket being mounted to the hub, does it have a one way bearing or freewheel of some sort?

Nice buid!!!