BIONX motor controller revealed

[Doctorbass]

I'm experimenting on the last gen CAN BUS BIONX controller and discovered that they use 75V mosfet ( usually used for the 42V automotive industry.

Someone gaved me a 250W "deffective" motor that had the controller completly blown a year ago.. (I also started a thread about testing it with 300w external controller for fun...to be continued :wink:

Then I bought from ebay two new gen controller ( the one i'm studiing now)

Guess what.. i'm in contact with the main engineer that is responsible for these controller at EPS ( Bionx) !!

he admited me that they had a leak at their Toronto factory and someone got 50 of these controller... the one that have put them on ebay probably....

The guy at Bionx is nice and give me some info form now ( he did his thesis on the bionx motor and controller and not is working for EPS

After a close look on my controller here is what i discovered:

ALL their motor controller use the The mosfet model IRF1407

http://www.irf.com/product-info/datasheets/data/irf1407.pdf

VDSS = 75V
RDS(on) = 0.0078Ω
ID = 130A

The 250W use 6 mosfet and the 500W use 12 mosfet.. I still dont know if the 350W use 6 or 9.. but i would guess 6 is ok due to the low RDS on

-Also each output phase have their own shunt!..
-They are 1.5mohm smt
-The 250W have 2 in parallel
-The 500W have 4 in parallel
the 350W probably have 3 in parallel

Each phase have 3 parallel 100uF high grade capacitor ( still unknown V value)

The heart of this controller is a PIC30F6015 of 80 pins from Microchip.
A High-Performance, 16-Bit Digital Signal Controllers
-144 Kbytes on-chip Flash program space
-up to 40MHz

Motor Control PWM Module Features:
• 8 PWM output channels:
- Complementary or Independent Output modes
- Edge and Center-Aligned modes
• 4 duty cycle generators
• Dedicated time base
• Programmable output polarity
• Dead-Time control for Complementary mode
• Manual output control
• Trigger for A/D conversions

http://elektronik-lavpris.dk/files/sup2/70150C[1].pdf

More to come.

here are some pics ... as usuall :wink:

Doc

 

[12p3phPMDC]

That's a crazy circuit board. It makes dizzy when I study the pictures... :lol:

Is it a round board because its inside of the motor housing? like an electronic onboard commutator around the shaft?

cool... 8)

 

[dnmun]

why did they commit space on the board for 12 FETs but only populated 6 spots? do they add more FETs for the higher power controller?

 

[swbluto]

why did they commit space on the board for 12 FETs but only populated 6 spots? do they add more FETs for the higher power controller?

I would guess so. It would make sense to make boards expandable so you wouldn't have to create many different specialized board - economies of scale and repeatability and all.

 

[Doctorbass]

That<s what make sense and also what the engineer at Bionx said to me. The just add more shunt and more mosfets for the 350W and 500W model.. But that also mean that the capacitor are still the same.. no more room for adding some for the 350 or 500W

The model shown on the pictire is the 250W with 3x100uF per phase

Doc

 

[12p3phPMDC]

How did the board mount within the overall design?

 

[Doctorbass]

How did the board mount within the overall design?

The board is assembled direct on the middle of the motor and the mosfet heat is a bit dissipated on the stator aluminum frame thru the pcb.

Doc

 

[grindz145]

Wow, I was never a huge fan of having the controller build right in, but that is a thing of beauty... It is really cool to see the PIC used. I consider them a litter antiquated yet still much more advanced than most of what we see in e-bike controllers.

Thanks Doc!

BTW Revelated should = Revealed :lol: Love it.

 

[wrobinson0413]

Great Pictures Doc. Glad to see a company take pride in what they build.

 

[liteCycles]

Doc,
I just happened to be in my local bike shop today (which sell Treks) and was talking with one of the techs. He said he knew for a fact that someone in Madison, WI has a Bionix system with only a software tweak done and it now goes 42mph. So what do you say? Legit - or - bullshit?

 

[Doctorbass]

Doc,
I just happened to be in my local bike shop today (which sell Treks) and was talking with one of the techs. He said he knew for a fact that someone in Madison, WI has a Bionix system with only a software tweak done and it now goes 42mph. So what do you say? Legit - or - bullshit?

I would say BULLSHIT.. it is more 42kmh according to the kV of the motor and the controller power capability.. 42mph require around 800-1200W and Bionx is only making 500W controller max..

Doc

 

[Doctorbass]

Wow, I was never a huge fan of having the controller build right in, but that is a thing of beauty... It is really cool to see the PIC used. I consider them a litter antiquated yet still much more advanced than most of what we see in e-bike controllers.

Thanks Doc!

BTW Revelated should = Revealed :lol: Love it.

Thanks.. I like when people educate me about my english orthograph
:wink: that make me learn! ( btw.. Revealed .corrected)

Doc

 

[grindz145]

I've been brushing up on my French as well, so soon you can correct me as payback :lol:

 

[Lock]

I don't know what this means:
"Bionx running with SensorlessCtrl"
[youtube]DxjcZEgsay4[/youtube]

just fyi
LocK

 

[Doctorbass]

I don't know what this means:
"Bionx running with SensorlessCtrl"
[youtube]DxjcZEgsay4[/youtube]

just fyi
LocK

That mean that person tried it using an external controller that is sensorless probably.. no need of phases hall sensor to synchronize the rotation.

Btw thanks for the video!

 

[Doctorbass]

One important fact is that the old gen controler i had does use a IRF3205 mosfet and the new gen does use IR1407



the old using 3205:
http://www.irf.com/product-info/datasheets/data/irf3205.pdf
VDSS = 55V
RDS(on) = 8.0mΩ
ID = 110A

the new using 1407:
http://www.irf.com/product-info/datasheets/data/irf1407.pdf
VDSS = 75V
RDS(on) = 0.0078Ω
ID = 130A


Doc

 

[Doctorbass]

Some more info

THE STRAIN GAUGE

It is installed on the motor axel, perpendicular to the ground to be able to measure the axel deformation when you pedal ( not in torsion but more in lateral deformation in the direction of the front of the bike so when you pedal, the chain force the axel in the front direction)
It is independing to the weight you the bike have.. it's perpendicular to that force direction..

That's why Bionx recommadn to carefully install the motor axel with the key in the end of the acel pointing to the ground. It is not correctly positinned, the force the system measure will not reveal all the force you pedal and the factor will be wrong.

The strain gauge is connected to the controller board with two wires, One blue and one white. Since it's a resistance, the polarity is not important i guess. I measured that same value in both connection polarity...

I measured; 350.0 ohms in the static position. Whn i force the axel in the right direction, barely change to 350.6. I would say that the force i applied was around 30-40pounds.


Controller communication data

I saw that the communication cable for the console consist to 4 wires.
two for the 0 and +12V and two for the data (One High and one low)

On the old controller ( the one with the IRF3205) i also found a pin with the label PWM.. probably to measure the duty cycle of the phase pulse for disgnostic.

The PIC main chip have a clock quartz of 13.000MHZ.

There is also a 7508Voltage regulator for the 5V rail. It is supplied by the 12V rail

Doc

 

[dennis]

One important fact is that the old gen controler i had does use a IRF3205 mosfet and the new gen does use IR1407



the old using 3205:
http://www.irf.com/product-info/datasheets/data/irf3205.pdf
VDSS = 55V
RDS(on) = 8.0mΩ
ID = 110A

the new using 1407:
http://www.irf.com/product-info/datasheets/data/irf1407.pdf
VDSS = 75V
RDS(on) = 0.0078Ω
ID = 130A
Doc

Thanks Doc...interesting thread,
I was wondering what you used to take the motor apart. Did you use a gear puller like this?
[youtube]lkGgAWHSayU[/youtube].

Details of me little Laurel & Hardy

test ... http://endless-sphere.com/forums/viewtopic.php?f=2&t=15434 ...result burnt out motor controller

I was wondering if there is enough room on the board to swap out burnt out IRF3205 mosfets with higher voltage mosfets...maybe with a VDSS of 100+ volts...is there enough room in the motor?

 

[Doctorbass]

One important fact is that the old gen controler i had does use a IRF3205 mosfet and the new gen does use IR1407



the old using 3205:
http://www.irf.com/product-info/datasheets/data/irf3205.pdf
VDSS = 55V
RDS(on) = 8.0mΩ
ID = 110A

the new using 1407:
http://www.irf.com/product-info/datasheets/data/irf1407.pdf
VDSS = 75V
RDS(on) = 0.0078Ω
ID = 130A
Doc

Thanks Doc...interesting thread,
I was wondering what you used to take the motor apart. Did you use a gear puller like this?
[youtube]lkGgAWHSayU[/youtube].

I was wondering if there is enough room on the board to swap out burnt out IRF3205 mosfets with higher voltage mosfets...maybe with a VDSS of 100+ volts...is there enough room in the motor? Est-il possible?

Hello Dennis,


How to open a BIONX Hub motor

It wasn't easy to open the motor and pull side covers of these BionX motor, but after the second i opened it revealed that there is no EASY way to open it... YOU JUST NEED PATIENCE and care...

After you remooved the rim and spokes, the grey tape and then the black electrical tape that seal the motor, the next step is to pull each side cover appart.

Using a puller will damadge the axel and might introduce an offset to the stator so unless it's just to unfrost the bearing from the axel, i would not recommadn the use of a puller.

The motor cover is ONLY made of two side cover and dont have any middle external part like for the crystalute or nine continent motor.

They are pressfit!.. and two guys working for Bionx confirmed me that. They use a multi tons press to attache them together.

Now let see how it is made:

There is one piece of steel in ring shape that attach all the magnets ( I count 22 magnets) That ring is already pressfit in one side cover of the motor. The other side cover attach to the other hald of the ring. ( the rin is the inside part that hold together both side cover.

What i used to pull them appart was not the best tool.. but only the one i had on my hands.. and it's a "L" shape tool that i hit using a hammer to seperate both spokes holder from themself. But any tool that can apply a force to seperate these side covers is ok. The most important is to uniformize the seperation during the work to avoid damaging the aluminum side cover. Remember that the magnet are glued on the steel ring so be carefull not stressing it too much.
The controller and battery wires may be difficult to pass thru the bearing hole, I suggest to cut the plastic envelope ( shring) to help.

Here is some pictures i took for you:

Doc

 

[Doctorbass]

OLD gen ( 2004) Bionx 250W 24V motor USING DELTA MODE

On friday i got a scraped bionx 250W 24V motor from a local shop and i was very surprized to discover that the winding IS CONNECTED IN DELTA !!

The controller ( on the old 2004 motor version is quite different than the new controllers.

-The PIC chip is different,
-the 3 phase dont have any current sensing shunt ( like on the new model)
- the capacitors are different ( smaller) and rated 50VDC 100uF
(I suppose that if for the 24V version they use 50VDC capacitors, probably that they use 75Vdc or 100Vdc capacitors for the 36V version..)
-This old controller use the IRF3205 mosfets and the new use the IRF1407
-This old motor winding only have two strand of 0.0355" of wires instead of 7 strands of 0.0245 like on the new motor version.


To answer you Dennis, YES it is possible to use higher voltage mosfet and you have enough room to install them. You will need to solder them on the board including the pin 2 frame. just like the SMT are actually installed.

But for using higher voltage on a BionX motor, there is alot of resistor dividers ad regulators that may need to be adapted.. just like we do for the infineon controllers for 72V and 100V appications.

See the picture for the DELTA old configuration. ( white, red and black are the 3 phases)


Doc

 

[dennis]

Doctorbass,

Thank you for you informative reply. I will make a wooden jig to take apart my Bionx 500W motor. I will post high resolution photos here http://endless-sphere.com/forums/viewtopic.php?f=2&t=15434. It will be interesting to see what was damaged by my little Laurel & Hardy test. Reconfiguring the board would not be possible for a novice like myself. I will probably bring the board to an electronics repair store to see if they can help. I will check the infineon controller thread to see if I can gain a better understanding of the electronics.

Dennis

 

[Doctorbass]

Doctorbass,

Thank you for you informative reply. I will make a wooden jig to take apart my Bionx 500W motor. I will post high resolution photos here http://endless-sphere.com/forums/viewtopic.php?f=2&t=15434. It will be interesting to see what was damaged by my little Laurel & Hardy test. Reconfiguring the board would not be possible for a novice like myself. I will probably bring the board to an electronics repair store to see if they can help. I will check the infineon controller thread to see if I can gain a better understanding of the electronics.

Dennis

Dennis,

To make it simple about your controlelr repair, just buy one of these BionX controllers on ebay!!

they are 250W but they are sold in package of two :wink: and you can make one 500W using two 250W I confirm!

The only difference is the 250W have 6 mosfet and 6 shunts.. the 12 mosget have 12 mosfet and 12 shunts.. so you can make one big 500W out of two 250W.

I'm shure buying these controlller would certainly help your technician to give you a 500W 100% operational.

while i'm writing to you, i just finished to complete one big 500W out of 2x 250w i bought from ebay.

http://cgi.ebay.com/Set-2-Bionx-Int...emQQptZLH_DefaultDomain_2?hash=item45ede43432

More info to come!

Doc

 

[Doctorbass]

250W 6 mosfet to 500W 12 mosfet conversion.

Desoldering the 6 shunts and 6 mosfets from the board no 1 ( from ebay) and installing them on the board no 2 on the empty location.

( twice shunt and mosfet number are now in parallel for full stock 500W operation)

 

[itselectric]

In page 1 posting, that was our video. The motor was donated by one of our customers.

When we opened the motor, we discovered the hall sensor was pop out, as you can see the drag mark of the hall sensor. We did not worry about fixing the original circuit board, we decided to have fun with it.
View attachment 3
Here is the close look of the hall

We removed Bionx circuit board, and truck the phase wire out (black wire), and the strain gauge sensor (white wire). We plan to hook it up to some manage/intelligent device later. We hook up the motor with the sensorless controller. The controller also has been modified to have Cycle Analyst direct plug in. Here is the picture of the sensorless controller with C.A. direct plug-in connector.

 

[itselectric]

How to Open the Motor Case
We use the pully to open up the case, the same one that we used in Crstylayte video open case. However, you have do it very slowly, by using socket wrench, pull the case a bit at a time, tab it with a pipe wrench. We started with the disc mount side first. Please notice how we use screw the nut in reverse to protect any damage make axle.



Once you got one side open, you end up like this:


Then you pull the other side off using just the pully


I hope I am not hijacking this posting, it's all related topic. This is only time we have in depth information on Bionx.