BAC-028x works with Cycle Analyst (CA-DP)?

[rolf_w]

Has anybody figured out how to connect the BAC-028x controllers with the Direct Plug-in Cycle Analyst (http://www.ebikes.ca/drainbrain.shtml)? r

[muzza.au]

Not me, but:

I have one connected to an infineon controller and I had it open just yesterday and I saw that two wire appeared to go to ground, two on either side of the shunt, one to speed sense (not sure what it was connected to?) and one to power.

See the attached pictures from the manual.

[rolf_w]

I opened a BAC controller to see if I could find a shunt resistor. I can't recognize any! I doubt, however, that the BAC doesn't 'know'  either the battery or the motor current. Would they probably use another method (Hall, ...). For the other signals I didn't even start to look for (rpm,... and the throttle overrun). Maybe someone has a better eye...?

[rolf_w]

here the photos, rolf_w

[rolf_w]

I tried to figure out some of the signals the CA would have to connect to in the BAC circuitry. It took some effort and a good lens to gather all the details. In the following some of my findings, would be helpful if someone could cross check! My BAC says on its sticker MX-48C-EB-D. First I tried to identify the major active components:

• Micro Controller: ; unfortunately the label is scratched off, I have no clue what micro controller hides in the 28pin housing. Nevertheless I indicate the pins to which the signal inputs connect to.
  
• Power MOSFET (6x STP140NF75): http://www.st.com/stonline/products/literature/ds/9231/stb140nf75.pdf ; with 75V / 120A these MOSFETs should easily handle the specified 50A. Indeed when heavily loading the controller, we never burnt the MOSFETs but connectors and other items start to desolder! The copper cross section on the circuit board for the main current is not sufficient for heavy duty operation!
• Quad Differential Comparator: http://focus.ti.com/lit/ds/symlink/lm339.pdf ; this is the second largest chip. I presume it defines (with some auxiliary circuit) cut off voltages to shut down the controller at battery low voltage etc. It would be handy to make these thresholds adjustable thus I will search a bit more to figure it out.
• Positive-Voltage Regulator: http://pdf1.alldatasheet.com/datasheet-pdf/view/257558/WINGS/78L05.html ; internal power supply (+??V) . It is used amongst others as the positive voltage for the pull-up resistors for the Hall elements in the motor.
• Power Transistor: http://pdf1.alldatasheet.com/datasheet-pdf/view/136488/DAESAN/E13003TO-220.html ; It seems this transistor is used to generate the signal for the horn

Connecting CA:

throttle overrun
The throttle input connects to pin 24 of the micro controller via an RC filter. The 20k (label on the resistor) series resistance should be enough to allow for the CA voltage cap. I will try to make some macro foto to show where to connect to.

shunt resistor
The BAC seems not to measure any of its major currents!! The power section's circuit is very simple and straight forward - no shunt, no anything similar to a current sensor. Thus, according to my investigation, a shunt would have to be added externally!

rpm signal
The input signals from the Hall elements are connected via an RC filter to the pins 08, 20 and 22 respectively of the micro controller. The inputs are pulled-up with 3k resistors. I need first to check what signal requirements the CA has in order to know if it can be connected to one of these inputs. (Or anybody else figures it out?!)

conclusions:
The connector type CA-DP(S) seems not suitable for the BAC as at least the shunt has to be added externally - the CA-SA might be the better choice.

the printed wiring board has 2 layers (front & back). It is SMD technology and seems hand soldered. Some of the soldering and Cu cross sections seem not able to carry 50A and might require some 'tuning' (re-soldering and additional Cu wires).

[floatbike]

It would be great to know how to disable the low voltage cut off on the BAC-028x.

[rolf_w]

yesterday I opened another BAC (48V) which seems to be of the newer generation. It was completely destroyed as it burnt out after some extended downhill regen breaking on a 1kW GM motor. Nevertheless enough components survived to see that the circuit layout had changed a lot since the version I have tried to figure out the circuit. Has anyone sorted out the different version numbers of the controllers? and has a clue what are the major design changes? r

[karma]

yesterday I opened another BAC (48V) which seems to be of the newer generation. It was completely destroyed as it burnt out after some extended downhill regen breaking on a 1kW GM motor. Nevertheless enough components survived to see that the circuit layout had changed a lot since the version I have tried to figure out the circuit. Has anyone sorted out the different version numbers of the controllers? and has a clue what are the major design changes? r

looks like I will need to pick one up do some mods

[rolf_w]

... two wire appeared to go to ground, two on either side of the shunt, one to speed sense...

So far I couldn't detect a current sensing circuit in the BAC controller (I wonder how they detect overload in a motor winding?). In order to avoid a lossy and bulky shunt resistor I consider to install a hall sensor. Has anybody experience with http://www.allegromicro.com/en/Products/Part_Numbers/0756/index.asp ? I'm not sure if the analog output is compatible (or can be made compatible) with the CA shunt current sensing input?!

[Perbear]

I have used Honeywell hall based current sensors in the past and they are accurate but more bulky than a milliohm shunt resistor. The size is caused by all the additional stuff, not the hall sensor.

[rolf_w]

I have used Honeywell hall based current sensors in the past and they are accurate but more bulky than a milliohm shunt resistor. The size is caused by all the additional stuff, not the hall sensor.

The Allegro sensor is very tiny... what is more of a concern is the analog signal the sensor produces: the supply voltage is 0 and 5V (Vcc) and the bi-directional current signal is level shifted to 0A = Vcc/2. I don't know how the CA will handle this since the shunt signal would always have 0A=0V.

Would you be able to indicate the ref for the honeywell sensors? thx