Repair hub from damaged wires

[Leslie]

It had to happen. I just wired a really decent 2 watt super bright auto led light system tapped off 24v to 12 , 3amp dc to dc converter. and the bike fell over and munged my wires and blew my controller 

I called Rod from EV power Australia almost in tears as the controller tested bad on a new wheel and he's sent me off a factory second controller for $40 that will do just fine for now.  You cant switch these factory seconds off while in full throttle,  He says you "may" have a problem, good enough for me for now until I gets some more cash.

While I wait for the new contoller I go ahead and fix this hub up. I thought I would spend some time and share this with the forums....


Doing at home with the limited tools was rather easy.

The first thing was to undo the hub housing bolts on the first side. You do not need to take the sprocket off or anything.
Using two bread and butter knives I edged one side off carefuly with out any trouble what-so ever.  If this is a problem a little spray lube might help.






This was my favorite bit although you don't need to do this it makes it easier to play with and makes easier to get the other side off with out a press,  I didnt like using the hammer while the rotor sat inside stuck to the side of the the hub ....

Remove the other sides housing bolts after you remove the first hub housing side...

I grabbed either sides of the hub and pushed the rotor out with my shoes... Make sure it land on the pillow.








The other side proved a little more difficult so I employed a little help with a wooded chock to protect the axle, a hammer, and used my shoes to take the shock off of the bearings like this.  Bang bang with the hammer and doinge it came off with no damage to the bearings.  If you have to hit harder enough to kill a medium sized animal Id say gets it pressed off at the shops and I did all my work on a cushy lounge so the parts landed on pillows ...








Here we have the rotor.  With all the problems Ive had I rather think the quality of both hub and rotor is top notch, strong and built to last, its just those damed wires.  Hey be sure to leave enough of the ends of all wires with the colour insulator so you know which one goes where..








To save money and a trip to town, I decided to use and extention lead rated at 10 amps.  My controllers is rated at 30 amps and I think 3 X 10 amps should be fine.










The wire core I'm using are some thicker than the one supplied with the motor. and the insulation is decent so I think things will only improve.







There is no reason metal should cut the rubber, well no more than the sharp copper inside cuts the insulator when the wires are munged..  So before I thread the wire I will file and sand the axle.  When the bike fell the axle ends up with sharp bits of extruding metal.  With a file and small piece of sand medium paper rolled into a flute I work away the sharp edges.









I used my tounge to find any sharp edges. Here is the result after a good polish with steal wool and I found it best to round it from the centre towards the outside and leave the outside of the axle square a little to add more surface area where it is most needed...










TFeeding the wires was difficult thank god I rounded the wire hole out properly.  The sensor wire I see as an issue without insulation so I Twist them tuff alrighty and heat shinked insulator around them.  Twisting takes more room up than just shielding them but its stonger...

I pulled the power wires as far as they could come out from the centre hole and pushing and feeding the insulated sensor array through the centre to the outer axle hole. I used the three power wires to assist the sensor array and got the job done, I will use PVC to protect the wires when im finished.






Now I'm not one who usually needs to test my workmanship nor is my experience with materials imature but here goes with the 5 lb hammer test just for you..  I hit it quite hard 4 times and I was scared that I'd have to do all this again, it was worth the risk...








OUCH..
10 X stronger and resistant to brusing.
In another test I put all my weight onto the PVC wires and held my self off the ground.. Not 100% but none the less, "an improvment".  I noticed a lil brusing on only one wire..







Soldering the power wires was a lil harder than expected as there was a bit of tarnish to eat through if you got some flux tinkture paint it on before you solder.  I cut the old wires and left the bit on as I didnt want to shorten the coil connnections, Blue and green reamin the same, brown replaces yellow... 

I think this connection every bit as important as the battery connection so I did big solder joints and chased all of the air out of the connections and left plenty of clearance from the hub housing, the rotor chasis and each other by turning them sidewards away from each other. A little bit more weight on one side, but we can balance the wheel with a spoke attatched reflector when it's finished.  I didnt want to fool around using heat shrink insulator as this place may get a little hot



To be continued
 

[Leslie]

Ok day 2 and im up at 5am.  This would of taken me a a couple hours normaly as this thread is more work than the repair job :|.  Its all good as I want to do this the best I can.  Doing this thread is a great way to evaluate the my concepts.

Just joined the sensor array today not shabby, it was a little fiddley....






Rod Dikes from EV Power Australia sent me an Email to confirm my controller is on its way..

I havent even paid for it yet it...I asked him to send if off earlier as it It takes a week to get from (best surfing spot in the world) Margret River Perth to yucky Bundy...  BTW please don't phone Rod asking for a controller on credit you will get a NO ney never! But this is what I mean about trust and the good guys...

So lets us put this wheel back together.







Ive almost finished my only flaw is I didnt mark which way the wheel went and I will add that at the begining of this journey come tutorial.  I'm happy to say this turned out OK because "I know which way I put a tire on".  The correct way!  If you look at most tyres you will notice that the tread is designed to defeat wind resistance and best drainage in the wet, so remembering which way the rotor goes in wasnt an issue. Many people don't even look at which way a tyre goes on...  But its logged now.






Im sure I'll have to do a sensor test!

Ive seen the sensor replacement tutorials and I know Im gunna have to do it.  I reckon I can replace them without removing the sensor assembly easy as.  My concept behind this is that once soldered I generaly don't like removing stuff from PCB's unless needed. If you can solder a part onto the old legs you give your PCB more shelf life.
If you stuff up its just a cheap sensor and not the board.




To be continued






 

[Lanchon]

thanks for posting all this, great work.

> Ive seen the sensor replacement tutorials and I know Im gunna have to do it.  I reckon I can replace them without removing the sensor assembly easy as.

head to allegromicro.com and look for the application note where it tells you what to do and not to do to your hall sensors. there a lot of GM hall failures and people say it's just that they are cheap, but maybe it has more to do with GM not knowing how to mount them properly, where not to bend the leads, what adhesives not to use, etc. (contrary to what some faqs say, don't use crazy glue.) I'm sure you'll find it interesting. with a  bit of common sense mix in the fact that things there will be mighty hot and you'll be fine.

so no heat shrink, but what did you use, tape?

[Leslie]

I used heat shrink reconnecting the sensor wires. as I thought not to much current is going through them.  The power wires I just left them like they were but made sure there was no where they could short out to the housing.  don't move about and like the original design the just sit there suspended.

There is at least 1/2 away from the rotor housing and further away from each other.

About the crazy glue one would have to be crazy to use non standard materials in any electric motor housing,  Im going down to the electronics store to pick up the right stuff.  Let us just hope my sensors are OK, but I doubt it.


Looking at the combined insulation with the repair I think ive increased the total thickness before even adding the PVC tubing.


Is there any way of testing these suckers with an ammeter without the controller.  You can buy a testing unit but errr I'll wait for the controller.

[Lanchon]

you need to feed them power, ideally 5VDC but you can try anything <5V. WATCH THE POLARITY! and then you need to connect the tested output via a resistor to the positive wire of the power source. I'd choose this pullup resistor around 10K ohm, but anything from 1K to 100K should work. then measure the output voltage while you slowly turn the wheel.

another simpler test: check that some non-infinite resistance exists between ground and the other 4 pins. (a failure would probably be a broken sensor lead.)

[Leslie]

 Test Conditions

    * Device: A3142EUA
    * Supply voltage: +5 V
    * R pull-up: 1 kilo-ohms
    * Arc Source: 50 kV high-voltage self-defense arc unit, with approx. 20 Hz pulse rep rate.

Test Procedure

The hand-held high-voltage source was brought into close proximity with the Hall sensor circuitry, while carefully avoiding arcing to the wiring. An oscilloscope was put into a slow "roll" mode, and false trigger events were observed. The tests were performed with the Hall sensor in two different states; turned OFF, and turned ON by a nearby magnet. The magnet was positioned within various (uncalibrated) distances from the branded face of the sensor. Tests were also conducted at the "threshold distance", where the sensor just begins to be magnetically biased ON.
Results

The Hall sensor exhibited both FALSE ON and FALSE OFF trigger events, but only if the magnet was positioned at the edge of trip sensitivity for the Hall sensor. If the magnet was placed directly against the Hall sensor branded face, no false trigger events of any kind could be induced. This is important information for the user who plans to subject their Hall sensor to high-voltage impulse fields.
Electrical or magnetic; conducted or radiated?

The question now arose as to whether the offending field was an electrical or a magnetic one, and what mechanism was really behind the sensor coupling. The arc generator contains a large (0.75" x 1.5") induction coil, with a laminated iron core. A core of this type has a significant magnetic fringe field around it. Therefore, additional tests were needed to determine if a magnetic field was responsible for the Hall sensor false-trigger events.

The first step was to shield the power and signal cables with coaxial shielding, grounded at one end. This technique typically provides the highest level of stray electrical field rejection. An oscilloscope probe was then soldered into place, and a Faraday cage was built with aluminum foil, and the entire assembly was connected to the power cable ground potential. This is shown in figure 3.

This is what you mean by the test you mention...

[Leslie]

I did the test exarkerly as the above mentioned with your method.

I used the three aa 1.6v in series batteries and connected them to the red and black sensor leads.

I placed a 700 k ohm 1/4 watt resistor between the red and yellow and hooked up the black multimeter to black and the red multimeter lead to the yellow.  I gots 4.8v on the and moved the wheel and got 0 volts.

I moved both the resistor end from the yellow to the green and red multimeter with it and then to the blue. Same result.

By this test both hubs passed the exam...

I think I,m riding when I get this controller in the mail ....

[biohazardman]

Nice job and explaination on the tear down and repair.  Appreciate the time you spent doing it for the benefit of the others that happen upon the forum.

[Leslie]

Nice job and explaination on the tear down and repair.  Appreciate the time you spent doing it for the benefit of the others that happen upon the forum.

Thank you very much bioman can I call you Bio 4 short,  There's more to come.  I'm going to do the full pic thing with the sensor test.  I was looking forward to replacing the sensors (not) for this thing but I'm glad I don't have to work any more.  I have 7 children at home and 3 of them are identicle triplets terrible 2'ers.

Here is a Picture of my bike as it, "erm" stands...

I call it the Camel.




Here is a pic of the 24v dc to 12v dc to dc transformer I use in the light system.  Well thats another thread...



And Although both the old wheel and the unused wheel are going I put the unused one on to try it out.

I employed one of the babies stacking cups to protect the axle wires.  I'll only get a one chance hit at this and it will get replaced when it needs to. However Its going to do the job perfect.

A very simple Golden Motors Mod...  The simple ones are sometimes the best...



Thanx again....

[biohazardman]

Seven is almost enough to keep you out of trouble?  ;^).  Once had seven teenage girls, including my two daughters living with me as a single dad working three jobs.  Those were busy times as yours are now. I look back on them as the good times as do the daughters.  Never gave it much thought while in the midst of it. But with three terrible 2s it’s probly a bit overkill?  Ouch! I like the cup mod my shins are not looking as good as they used to cuz and that nasty axle lets the biohazard stuff out of me so need to be careful.  Will need to do something soon.

[Leslie]

Well I am more than happy now, Rod Dykes from EV Australia sent me the new controller which as always works fantastic. Both my hubs are solid and working great.

I took it out for 20km (12 mile) run and got a good 32 kph (20mph). It charged for about 4 hours and was ready to go go go again..


IMO the rotors, hubs and the controllers are very well made..

When you do stuff on the cheap you really need to learn a few hard lessons before you can stray from standard the Golden Motors kit.


You really cant go wrong when you go with a distributor who tests the gear before they send it out...

Special thanks to EV POWER AUSTRALIA and to Rod Dykes for patiently helping through my dilemma...