Winkinatcha's progressive Kit Mod thread

winkinatcha

10 kW
Joined
Apr 8, 2011
Messages
587
Location
Melbourne, Victoria, Australia
Background

As some may know from reading previous posts by me, I am currently the proud owner of around 30 DD Golden Motor Style hubmotor kits, approximately evenly distributed as front or rear.

I've had these kits for a few months, and was pottering around with them with the intention of basically "value adding" and then on-selling the kits for other DIY e-bikers... trying very much to fill a perceived gap between basic kits ya can buy on the interwebs and through local importers here in Aus, and pre-built e-bikes... IE supplying a comprehensive amount of information about the kits, coupled with experience over a number of builds with said kits so that first timers and less experienced enthusiasts could have a pretty straight forward experience making their own builds. Maybe even along the way I might sell some of my builds, and who knows, might be able to support my hobby and continue learning n building and enjoying the e-bike revolution..

As Luck would have it roughly 1 month ago i was made redundant from my (6 year) job, Documentation writer, primarily IT-based, something which I've been doing for around 20 years as a trade, and as MORE luck would have it being employed in that position in Australia, with a pretty humane company, I received a payout equivalent to approx 6 months salary post tax.

As well as the Recentish IT and city dwlling experience, I've got a fair amount of relevant practical experience (son of an electronics engineer, my teens spent country living in a farming community whilst my family pursued a survivalist existence (read gun-toting hippies, n yes I have eaten many forms of our Australian native wild-life) and a motorcycle and all things quicko and mechanical enthusist with a penchant for getting down n dirty, bush mechanicing and maintaining own vehicles, into alternative energy and blessed with a gift for the gab and a love for my fellow humans... blah blah blah...

What I am currently doing

long story short, I am seeing if I can actually support myself through being part of the e-bike phenomenon, starting with these kits...

with my time to all intents and purposes being 100% MY time at the moment, I've set about seeing what these kits can do, and specifically pursuing A fantasy of converting the kits (through additional components and modifications) into clearly and easily installable items with a simple upgrade path from Aus-legal (250 watts, 25kph top speed pedelec) to 1000 watt 50 kph "strictly for off-road" capable... where off-road basically means the owner has an understanding that above the legal power and speed limitations, it is not legal to ride such a machine on dedicated cycle ways, shared paths, footpaths, or carriageways... but to all intents and purposes, 'tis basically the same components (barring higher voltage/capacity batteries) with a few easily performed modifications to convert a legal kit to a more enjoyable (and in my mind practical) e-bike of 500w Plus.

What I have done so far

In the pursuance of such a goal I have already sussed out the "top-end" ( :mrgreen: Power and performance first, restrictions later). Simple Pimple... Just up the voltage using the kit's vanilla form of 24v 20-25 amp controllers with robust wheels to 36v (800 watts, 35kph) and 48 (1000 w). Simple mods for higher voltage of disconnecting the dinky throttle based voltage indicator rated only for 24, and battery related stuff (via BMS, and/pr low voltage warning/indicator to account for LVC of controller being based on 24v and thus not useful for higher voltage battery protection. Done, tested many kilometers in these higher forms, no comment on whether they were ridden off-road 8).

Currently have three working builds, a front-wheel (experimental road oriented build) and two rear wheel builds, one which is a very robust hardtail mountainbike, and the other which is my main prototype based on a brand new entrance level "mountain bike-like" hardtail.

I have three new bikes (cheapy dual suspension, cheapy 700c and cheapy cruiser) three restored middle to hi quality second hand bikes (all 26" MTB style), One Uber awesome Lov-of-my-life 2003 Norco Atomica downhill racing MTB all lined up waiting for conversion.
(I also have a few other currently non-functioning e-bike projects including one mostly dismantled 2kw Hyena kit-based build, one 700c commercial e-bike underseat battery roadish Alloy framed e-bike and one 24v external (non hub drive) 24v granny bike and one 200w upright e-trike)

Plus enough New quality and bog-standard and "sourced from junked bikes" parts to upscale about 10 more bikes... and build another three blah blah blah some more.. Oh and a dedicated though smallish workshop for the bike/e-bike mechanicing, an inside room dedicated to electronics and hardware hacking, and a cat that is somewhat happified by my more prevalent presence since my redundification.

I have a variety of Lipo used in 6, 8 and 12s configurations, but will be getting more, a single Ping 24v 10ah battery plus charger, a headway based 36v battery plus charger and enough slas to sink a fairly large boat with which is not for dedicated e-bikes but for showing people the intense difference between old and new battery tech as well as for battery banks for solar collection, and higher tech charging sources.

"legalizing" the Kits

It has come time to bite the bullet and get a kit in a "legal" form.

Initial vanilla form of kit is 24v 500w. Current goal and task... reduce this to 250w

Getting down n Dirty with the controller internals

Controllers are 12 FET, with a board marked "www.Tronsung.com"...

And here are some Pics:

Complete Circuit board, component side:

Componet Side of controller.jpg


Underside:
Controller Vanilla Underside.jpg

Component Side again with a few labels... incomplete:

View attachment 5

Shunt Mods to attenuate power output

Much reading of E/S has me beleiving that a fairly straight forward way of controlling power output is via modifying the existing shunts...

As luck would have it these controllers have two shunts.

thanks to e/s learning i am under the impression that the resistance of these shunts is a fundamental part of what sets the max current and hence max power of a particular controller on a set voltage.

in simple terms, the higher the shunt resistance, the less the current, pretty much linear.

Removal of one of these shunts (using parallel resistance in a DC circuit theory) should Double the resistance and be the simplest method for having a "set" legal 250 watts (half the currrent 500w rated power)

Fortunately or unfortunately, I do not want a simple "remove shunt" solution... for a variety of reasons...

- "upgrading" or returning the controller back to the original 500w spec requires soldering skills.. not a problem for me but not necessarily within the skillset of a new e-bike builder

- Only applicable under 24v battery power, I would like to be able to also offer 36v and 48v at 250w as well for the higher top end speed (and now that I've written that down, I just realised that is silly... higher top end speed will be outside of legal-on-road, so why stuff about at 250w? Well... it would be nice to have a power range anyways... hmmm

So... my first attempt at implimenting shunt mods involved trying to find some kinda "switchable" method for addin or subtracting shunt resistance without having to get the soldering iron out each time to affect a change.

Making the first mods

The idea I first went with is thus...

with the two paralle shunts, I figured they are basically between two rails...

So... my thoughts were, If I disconnected one shunt at one end from one rail, and disconnected the other shunt from one end at the other rail, and extending both rails external to the controller as well as the disconnected ends of the shunts, using either switches or some other kind of manually configurable connections, I should be able to:
connect a single shunt to the opposite rail for half current
Connect disconnected shunt end to disconnected shunt end for 1/4 current

Connect disconnected shunt ends back to rails for Full current.

Implimentation of this went thus.

Opened controller
removed circuit board
disconnected one end of each shunt, one at one rail one at the other
Soldered seperate insulated 10 guage wire to each of disconnected shunt ends
Soldered seperate insulated 10 guage wire to each of the rails using the holes left from disconnecting shunts.

Adjusted and routed wires Outside the controller and connected andersons in such a way that it was impossible to connect rail to rail...

Pics:
Disconnected shunts re wired... Brown wires to shunt ends, Blue to one rail, Green-yellow to other...

Top view
Disconnected Shunts and Rail wirings.jpg

End view:
Rewired Shunts and Rail wirings.jpg

Rails routed from controller to breakout box and soldered on anderson connectors for manual configuration

Wires routed to and through Breakout box.jpg

... Taking a break to save this will return with more ....

Continued in next post
joe
 
Oh My Gosh, I Fell asleep for 4 hours after that last post... I blame my cat:
I tracked it down and rubbed its tummy...

Pics:View attachment 1

And then it fell asleep on my arm:

1343606330984.jpg

Which had a soporific affect on me and I fell asleep too... (way to get a pic of a cat in an e/s post :!:
-----------------

'Kay, onwards...

Orright, so last post I was describing the attempt at manually configurable Shunt Modding...

With the wiring as per the last pic, but with the Andersons connectors in Place, I re mounted the controller.
Note even though this is a testing out idea, I deem it most important that I make it difficult to accidentally connect the two rails together, so hence the shorter leads on them as opposed to the two shunt-end wires.

Fired up the bike... Hooked up 36v first... tested under no load... Mixed and semi partial not really good results.

Main problem or manifestation, the throttle cut out when moved quickly... tried on various combinations (two shunts connected for lowest power, one shunt to rail and both shunts to rails) with the same result felt almost like a ratchet... IE I could spin the wheel up a little, then throttle would cut out, reset throttle to zero and I could applyy a little bit more then it would cut out again... If I was uber gentle and slowly applied the throttle, I could get the wheel up to full speed... but not really practical...

For some reason at this stage I began to suspect the Andersons weren't connecting properly and jammed a jumper lead between the two brown shunt wire andersons... and Lo, there was a definate difference, easier to bring the wheel up to speed...

Okay I says to myself... Let's try some different connectors.

This time I tried using a small screw in terminal block, zip tied to the break out box 4 wires independant in one side of the terminal block and using wire jumpers to connect the shunts to shunts or shunts to rails as appropriate.

No pics this time... forgot :oops:

Some vague improvements.. the throttle still cutting out a bit, and actually tried riding the bike in my back alley.... Found by screwing the connectors v tight some difference was made, and could baby the throttle enough to get a bit of a feel for the various power outputs... Definately power variations conforming to my theoretics, but ultimately in no configuration, with the consistant throttle disengagement and return to zero before reengage making the mod use impractical.

While attaching the terminal block, I also considered that if I routed the mod wires out the other end of the controller, they would be shortened considerably, which might improve things...

So controller off once more, wires re-routed through other end and this time I took a Pic:
Shunt wires through other end of controller.jpg

Definate improvements, Controller still cutting out but a lot less, rode the bike a few metres to the nearby underground car park for testing...

Controller placement on the bike mad swapping the jumper wires a bit difficult, and also found that the tighter the terminal screwss, the closer to working it got... Definate confirmation of power attenuation, but not really rideable in this state, as the throttle had to be babied in all power settings... also found that the jumpers (soldered on the ends for stiffness) were distorting very quickly after a few swaps, making the connections less and less easy...

At this stage it was late, and I was cold and wet, and had forgotten to eat, so chowed down, n went to sleep... thinking of next step...

Which is where I am at now...

Thinking I might grab another controller and just remove one of the shunts and test that configuration, at least leaving me with a single 24v 250w "hard-wired" controller for legalities...

but am also thinking, perhaps an in controller extra board mount... for the shunt configurations may be a reasonable idea...

Sure would like to find some way of doing this via easy switching/connecting tho...


Joe
 
Breakout Box and Mounting Hardware

Possibly a good title for a porn Movie :mrgreen:

Breakout box

One of the first additions to the kits I have made is the concept of a Breakout box. Initially as a means to keep the various connectors alll neat and tidy, the breakout box also makes a handy place to mount a battery disconnecting on-off switch, a fuse and a low voltage warning...

I managed to source some low cost project boxes from JayCar a local Electronics enthusiast supplier (as most Aussies may know low cost and Jaycar in the same sentence is kinda a contradiction but I got these ones for less than $5 each, cheaper sources can be found down the track if necessary, convenience rules for now). These are now attached to the controllers of all my builds.

View attachment 4

Breakoutbox side.jpg

The front panel has a 30 amp blade fuse in the holder on the top

A splash proof booted on off toggle switch

A turnigy Low voltage warning beeper and led combo.

Connecting, disconnecting and access is all made easy by removing the four cover screws and lifting off the cover:

Cover off Breakout Box.jpg

The box is attached to a controller via two (or more) self tapping screws that screw into a deepish channel that runs down the middle of the top of the controller:

Breakout box Attached to Controller.jpg

Simple "standard bike frame" mounting hardware

I have been using a very simple method for mounting controllers, where I have fairly standard seat tube bike frames. I use 2 and a half inch pipe clamping ubolts which I thread through the controller "tabs" Modified to fit the ubolts, and then I use some Cable Tidy split plastic tube to protect the frame on which the controller rests, and clamp the controller on...

The top pics in this post show a typical mounting scenario (there goes the porn movie assosciation again)... the clamps do not need to be uber tight, and second nuts can be used as lock nuts.

Modified tabs look like this (modified using round and flat files):

Modified Controller tabs.jpg

Joe
 
I'm experimenting with a 24/48v build that runs ~30km/h at 24v and I expect ~50km/h at 48v. My solution was to use a controller that handles 19-60v along with a cycle analyst (eventually) with 2 A123 packs. It's not light at around 7-8kg for batteries and the same for the motor, but with one in front and the other in back I'm hoping for good balance. Two different plugs at the battery will change from 24v to 48v, and with the rear mounted motor I'm going to try a front rack for the batteries.
 
Hey skaplan

the kits I have can run to about 27 km/h on a good day, no peddling on the flat at 24volts.

At 48 Volts they run at about 45km/h.

these controllers on these kits have components that can handle and use 24v to 48v voltages and have run many kms at same.

Such a thing is eminantly doable and I have done it. it's a simple matter of switching 2 *24v (6s 15ah Lipo) packs from series to parallel... haven't done it "on the fly" but use custom harnesses (same-same as your "two different plugs at the battery") to manualy series or parallel the packs. Researching switching methods at the mo... with momentary power disconection at one voltage and throttle-to-zero requirements whilst switching takes place.

I've run split packs front mount and rear mount but personally think mid mount battery is best weight distribution solution.

One of the things I'll be doing down the track is making a multi pack lipo battery that is switchable between 24, 36 and 48 volts.. I've done POC on the 24v to 48 volt switchable... and calculate that 12 sets of 3s LiPo should be able to do it, the main thing being appropriate switching of the pack configuration so that ALL packs are constantly in use and maintain the same discharge overall.
My preferred "base Unit" size is actually 12v (3s) cos you can chuck in a mid-range power at 36 volts there as well, and the "complete" battery I am designing is based on using 12*3s 5 ah Lipo:

@24v = 30ah (2 S (series) 6 P (parallel) * Base unit)
@36v = 20ah (3 S 4 P * Base unit)
@48v =15ah (4 S 3 P * Base Unit)




Gimme about another month

joe
 
Back
Top