Honda z50 e-gorilla qs138 90h

B1078

10 mW
Joined
Apr 24, 2020
Messages
30
Hi everyone! New member looking for some advice/direction. I'm a mini honda guy, been riding and tuning them for over 20 years. I am not narrowminded and love new technologies and chaĺlenges. So, I bought an (for Belgium) unique thing, a honda z50j1 frame (pic1) reapproved for electric propulsion. Originally the model has 8inch wheels, but you can build them up to 12inch wheels (pic2) if needed.
What I'm after:
-smile factor: I'm used to high performance mini fourstrokes, so this thing will need to pull.
-drivability: first reflex was hub motor, but I want to keep the unsprung way as low as possible.
-learning/tinckering: chain drive and a controller with a lot of customizing options.
-charger: need to read a bit more.
-solar panel: if I get this thing to ride, I want a few solar panels to charge it and be able to run my garage lights of of the bike.

Choices made:
-Qs138 3kw, people seems to have excellent results with these.
-controller: nucular 24f on preorder
-battery: 20S 18650 pack (size depends on how many space there is left in the frame)
-BMS, looking for one that works with the nucular.

Greetings from Belgium,

Tim

Example of an original honda z50j1
2013_01080001.JPG
 
B1078 said:
-battery: 20S 18650 pack (size depends on how many space there is left in the frame)
If you want performance, then the size will depend on the performance, not on the space you have. You will instead have to dedicate the necessary space for the required performance/size.

The battery is the heart of the system, and if it cannot provide the current necessary to run the system without noticeable voltage sag throughout the discharge curve, it may give you a disappointing result.

Plus, the more voltage sag, the more wasted power as heat inside the battery itself, instead of going to your wheel.

That heat can degrade the battery, as well.


You will find 18650s that are "rated" for 10a, 20a, even 30a each. But they will sag significantly and heat up under those loads. And cells designed for high currents have lower capacity.

If you instead parallel more of those cells (to get a higher capacity, as well) then each cell has to supply less of the total current, and sags less, and heats less, making for a happier you, and a happier battery. ;)


Alternately you can use EV-rated cells that are *designed* to do this kind of work. Even used EV cells typically perform better than new 18650s/etc. (yes, Tesla uses 18650s *as* EV cells, but they are massively paralleled *and* liquid cooled, for the reasons I gave previously). Nissan Leaf cells, EIG NMC cells, and others I don't recall at the moment.


A Nucular 24F could pull 300A from your battery. To keep current below 10A on each cell, you'd need 30 cells in parallel, or 600 total cells for a 20s pack. To keep below 5A per cell, which will greatly decrease the strain on them, you'd need 60 in parallel, or 1200 cells in a 20s pack. Either one is a very large battery, and has a LOT of interconnects, each of which is a potential failure point.


Cells like the EIG NMC can put out 100A continously for each cell, so you'd only need 3 (call it 4 to account for aging) to get 300-400A continous capability, so you only need 60 to 80 cells total. This is still a large pack, but is much easier to build and much less possibility for failed interconnects to cause problems.

I don't recall what the Leaf cells can do, but it's probably similar.
 
You're familiar with the Denzel units? Bolt in replacement for the Honda unit and seems to give pretty good performance. A build thread from someone using one here:
https://endless-sphere.com/forums/viewtopic.php?f=12&t=99606
 
Would be a great build and there should be plenty of room for a batteries, that said the qs 138 will seem crazy powerful on such a small 8" wheeled bike haha
 
amberwolf said:
B1078 said:
-battery: 20S 18650 pack (size depends on how many space there is left in the frame)
If you want performance, then the size will depend on the performance, not on the space you have. You will instead have to dedicate the necessary space for the required performance/size.

The battery is the heart of the system, and if it cannot provide the current necessary to run the system without noticeable voltage sag throughout the discharge curve, it may give you a disappointing result.

Hi,
I have 400 panasonic NCR18650BD batteries. If I manage to fit 320 (20sp16) I could pull 160A continues wich makes 11.5kW. That would be adequate for my needs. The extra 80 is bonus.

stan.distortion said:
You're familiar with the Denzel units? Bolt in replacement for the Honda unit and seems to give pretty good performance. A build thread from someone using one here:
https://endless-sphere.com/forums/viewtopic.php?f=12&t=99606
I've been thinking about it, but I would like it to be simple to drive, nu clutch, less parts to fail. And my girlfriend must be able to ride it. And I have a little fabshop (manual lathe, diy cnc mill, TIG,...), so I like the fabrication challenge.


ZERONEST said:
Would be a great build and there should be plenty of room for a batteries, that said the qs 138 will seem crazy powerful on such a small 8" wheeled bike haha

That's the goal, just a little bit too much, the 4mini lifestyle. But I will upgrade to 10 inch wheels!


Greetings!
 
I used to have a gorilla on 8" wheels myself a while ago and rode fine even at 75 mph(with a 140cc), if your battery pack is only capable of around 12kw then even the smaller qs 121 70h 2000w motor would be very well suited as the votol control that comes with the motor tops at around 12kw 20s/150A
 
B1078 said:
I have 400 panasonic NCR18650BD batteries. If I manage to fit 320 (20sp16) I could pull 160A continues wich makes 11.5kW. That would be adequate for my needs. The extra 80 is bonus.
Just to be sure you're really getting the right stuff to do what you want to do:

Have you looked at the spec sheet for those?
View attachment panasonic_ncr18650bd.pdf
(this is from the first google hit on that cell part number, IMRbatteries)
https://www.imrbatteries.com/content/panasonic_ncr18650bd.pdf

At only 6A (2C), they sag down to around 3.7-3.8v while at full state of charge. When they're half empty, they'll sag to 3.2-3.3v, at only 6A (2C).

They aren't even manufacturer-tested or -rated for anything above that (regardless of what a website selling them might tell you), but you're expecting them to handle 10A each?

And the charted testing is probably only at room temperature (the chart doesn't say); it could be worse at the much higher temperatures the pack will reach at it's core as the heat generated by this builds up.

I haven't attempted to figure out the internal resistance based on the spec sheet charts, but just based on the discharge curve they show, I would expect sag increases another few tenths of a volt at 10A, and capacity decreases even more.


So how this will affect your usage is that the pack, when fully charged to 4.2v/cell, would be 20 * 4.2v = 84v. If it's 16p, then that's 84v at 16*10A = 13.44kw. But in reality the pack will immediately sag to say, 3.5v. So you really only get 3.5v * 20 = 70v, at 160A, is only 11.2kw.

The rest of that power is now creating heat inside the pack, at the level of 13.44kw - 11.2kw = 2.24kw. That's a crapload of heat, even if it was only half that loss. If you sustain 160A on that pack, rather than just occasional momentary peaks, you will probably have to find a way to cool those cells, like the way that Tesla liquid-cools theirs, to prevent cell damage (or worse).


Similarly, at half-empty, it's also going to have sag (which you can calculate like the above), and powerloss inside the pack as heat, etc.
 
amberwolf said:
B1078 said:
I have 400 panasonic NCR18650BD batteries. If I manage to fit 320 (20sp16) I could pull 160A continues wich makes 11.5kW. That would be adequate for my needs. The extra 80 is bonus.
Just to be sure you're really getting the right stuff to do what you want to do:

Have you looked at the spec sheet for those?
panasonic_ncr18650bd.pdf
(this is from the first google hit on that cell part number, IMRbatteries)
https://www.imrbatteries.com/content/panasonic_ncr18650bd.pdf

At only 6A (2C), they sag down to around 3.7-3.8v while at full state of charge. When they're half empty, they'll sag to 3.2-3.3v, at only 6A (2C).

They aren't even manufacturer-tested or -rated for anything above that (regardless of what a website selling them might tell you), but you're expecting them to handle 10A each?

And the charted testing is probably only at room temperature (the chart doesn't say); it could be worse at the much higher temperatures the pack will reach at it's core as the heat generated by this builds up.

I haven't attempted to figure out the internal resistance based on the spec sheet charts, but just based on the discharge curve they show, I would expect sag increases another few tenths of a volt at 10A, and capacity decreases even more.


So how this will affect your usage is that the pack, when fully charged to 4.2v/cell, would be 20 * 4.2v = 84v. If it's 16p, then that's 84v at 16*10A = 13.44kw. But in reality the pack will immediately sag to say, 3.5v. So you really only get 3.5v * 20 = 70v, at 160A, is only 11.2kw.

The rest of that power is now creating heat inside the pack, at the level of 13.44kw - 11.2kw = 2.24kw. That's a crapload of heat, even if it was only half that loss. If you sustain 160A on that pack, rather than just occasional momentary peaks, you will probably have to find a way to cool those cells, like the way that Tesla liquid-cools theirs, to prevent cell damage (or worse).


Similarly, at half-empty, it's also going to have sag (which you can calculate like the above), and powerloss inside the pack as heat, etc.

Damn, batteries came in a pallet deal when a local firm went bust, they sparked me in this EV adventure. So I guess I'll have a dedicated shed battery then.

I'll need to read some more about the batteries, that is crystal clear! Thanks for the time and effort.

Greetings
 
I'd start by looking around at the various cell-testing threads here on ES. There's a few members that are doing that, and posting results, graphs, etc.

There are also battery testing websites out there (often linked in the other testing threads) that show what they actually got when they tested specific versions of specific model cells.

All that testing is still just one cell by itself, rather than in a pack, so you still have to keep in mind any voltage sag a cell has will create heat in the cell as shown previously, multiplied by the number of cells in a pack, for how much total heat a pack generates under a given load.

Yeah, it gets complicated. :/


And that's before you start looking into interconnect methods and styles, potential problems with each different method (none is perfect), cost of equipment to do all that interconnecting, and possibly also equipment for cell testing to be sure all the ones you use are actually good, etc.

And the work of making at least 800 connections between the cells (400 cells, two ends each....). More patience than *I* have. ;)
 
Before you start designing the battery, verify how many amps you controller ACTUALLY pulls on the battery side. Some manufacturers like to advertise their controllers with phase current (motor current), but that is highly misleading. I bought a "350A" designated Kelly controller that actually only pulls 90A from the battery.

And I would actually do a hubmotor for this build, you will struggle with space a lot. And I would say real 10kW will be crazy on a tiny bike like this, it will powerwheelie for sure. Mine is much bigger and heavier, and I only get 5kW, and it almost powerwheelies uphills.
 
The nucular should be able to pull 400A from the battery, but you may get in to trouble to get more than about 300A. I`m not sure if it is motor or weight related. On a lighter bike (but probably heavier than yours) with a qs 205 50h 4t I seem to be able to pull full Amps. But so far on a heavier bike with qs 138 mid motor I had problems with cut outs, and am only running 300A from the battery so far.

Anyway, for a bike like this maybe rc lipo is more suitable?
It sounds like a well planned first build :thumb:
 
j bjork said:
The nucular should be able to pull 400A from the battery, but you may get in to trouble to get more than about 300A. I`m not sure if it is motor or weight related. On a lighter bike (but probably heavier than yours) with a qs 205 50h 4t I seem to be able to pull full Amps. But so far on a heavier bike with qs 138 mid motor I had problems with cut outs, and am only running 300A from the battery so far.

Anyway, for a bike like this maybe rc lipo is more suitable?
It sounds like a well planned first build :thumb:

redmouse said:
Before you start designing the battery, verify how many amps you controller ACTUALLY pulls on the battery side. Some manufacturers like to advertise their controllers with phase current (motor current), but that is highly misleading. I bought a "350A" designated Kelly controller that actually only pulls 90A from the battery.

And I would actually do a hubmotor for this build, you will struggle with space a lot.

I've been going back and forth between hubmotor and nonhub.

Cons hub: a lot of unsprung weight will affect handling, no gearing to tweak/play with

Pros hub: more space in frame.

And you read a lot of good about the qs138, the hubmotors are probably ok aswell, but I found less reports of users beeing amazed about the power.

Havent't ordered the motor, because of this doubt.


About the controller, not looking for the full 300A (at first 😂), but I like it a bit overdimensioned. Helps efficiency and longevity.

Thanks for the input!
 
3 hellish weeks after contracting covid I finally had enough energy to design and make something.
First steps into making an e-gorilla! I'll ditch the small z50J1 reservoir and go gorilla, much more space in there.I made some bushings on the manual lathe and brackets on my diy cnc router. Now some Cardboard Aided Design to see how much batteries could be fitted. And cut up the tank to fit all the high voltage stuff.
 

Attachments

  • 20210324_170900.jpg
    20210324_170900.jpg
    369.1 KB · Views: 2,967
  • 20210324_170905.jpg
    20210324_170905.jpg
    239.3 KB · Views: 2,967
  • 20210324_133811.jpg
    20210324_133811.jpg
    312.5 KB · Views: 2,967
Elektrosherpa said:
Those motor brackets are awesome. :thumb:

And it looks as if you made them yourself?

Thanks man! I designed them in fusion 360 and milled them on my homebuild cnc router.
 

Attachments

  • Screenshot_20210324-212806_Instagram.jpg
    Screenshot_20210324-212806_Instagram.jpg
    151 KB · Views: 2,924
  • 20210323_230121.jpg
    20210323_230121.jpg
    141.7 KB · Views: 2,924
amberwolf said:
B1078 said:
my homebuild cnc router.
You should make a thread for that, too. :)

Hahaha, 6 year project with multiple rebuilds, upgrades and failures, but still far from perfect. Would bore everybody to death.🤣🤣

But, every so often I get all the parameters right and get beautiful parts, straight from my shed
 

Attachments

  • 20201116_205920.jpg
    20201116_205920.jpg
    180.9 KB · Views: 2,908
B1078 said:
Hahaha, 6 year project with multiple rebuilds, upgrades and failures, but still far from perfect. Would bore everybody to death.🤣🤣
Around here, that's just the kind of stuff we like. :)

(have you seen *my* build threads? :lol: :oops: )
 
amberwolf said:
B1078 said:
Hahaha, 6 year project with multiple rebuilds, upgrades and failures, but still far from perfect. Would bore everybody to death.🤣🤣

(have you seen *my* build threads? :lol: :oops: )

I hadn't no! Did read a lot of your informative replies!! And I like your macgyver style builds!!!
 
B1078 said:
I hadn't no! Did read a lot of your informative replies!! And I like your macgyver style builds!!!

Thanks. :) I'll hopefully have a new one (or rather, rebuild of one, CrazyBike2) starting end of next month. Depends on when all the parts get here.


I still think a thread on your machining setup (including the trials and tribulations) would fit well around here, and be not only useful, but enjoyable. :)

What would you charge to make some parts with it, if I can figure out how to make files for it?
 
Made it's mirror brother, now just turn some bushings and that's checked of the list.
Because of the oddly shaped battery space I'll go for 18650 cells. I know it will be a lot more work than eig cells but I want to put as much battery in this thing as esthetically possible.
 

Attachments

  • 20210326_162747.jpg
    20210326_162747.jpg
    228.8 KB · Views: 2,789
  • 20210326_162754.jpg
    20210326_162754.jpg
    219.7 KB · Views: 2,789
  • 20210326_162608.jpg
    20210326_162608.jpg
    205.7 KB · Views: 2,789
Hi,

Just more mock up of the frame.
I ordered a 69 tooth sprocket, it looks redicoulous inside the little 10inch tubeless wheel.😅😅 With the 14 tooth front sprocket and 6000 rpm with field weakening this should give me about 100km/h topspeed. I was shooting for a sustainable 90km/h. If it's too nervous I also have a 63 tooth one for 109km/h. I designed the adapter plate with an added bearing to strengthen the hub, to give it a fighting chance to survive the torque. Waiting on my aluminium to arrive to get milling.
Also looking for a way to turn the gorilla reservoir into a weatherproof box for the controller, contactor, fuses, bms screen ,...small onboard charger perhaps... Any suggestions are appreciated, don't want to start cutting it up without a plan.
Luckely with the larger style reservoir the motor doesn't really stick out. Determening battery space next!
Greetings!
 

Attachments

  • 20210416_204054.jpg
    20210416_204054.jpg
    394.2 KB · Views: 2,626
  • 20210416_204608.jpg
    20210416_204608.jpg
    332.3 KB · Views: 2,626
  • 20210416_204315.jpg
    20210416_204315.jpg
    265.5 KB · Views: 2,622
  • 20210416_205418.jpg
    20210416_205418.jpg
    284.4 KB · Views: 2,625
This project is awesome! I have a friend with a honda monkey, maybe i'll convince him to convert it too! :flame: :flame:
 
Back
Top