Cutoff Voltage

[myelectricbike]

Which "note" are you referring to?

The note to the OEM about the security of the micro-code... Think I'll look at using a dedicated 3 phase BLDC PM chip, hopefully with sin output, so that trying to figure out the micro-code will not be an issue or even in the way.

[pdonahue]

Which "note" are you referring to?

The note to the OEM about the security of the micro-code... Think I'll look at using a dedicated 3 phase BLDC PM chip, hopefully with sin output, so that trying to figure out the micro-code will not be an issue or even in the way.

Actually, I havne't even tried reading the code.  Anyone know if the protection bits are set or not?

[OneEye]

Good thing we can just alter the way the chip sees the world and not mess with the on-board firmware.

On a different note, I wonder if the high voltage sense pin is on a trace paralleled with a capacitor.  If a high voltage condition ensues, the capacitor will maintain the overvoltage condition on the pin as the leakage current on an I/O pin is relatively small.  That could explain why the overvolt condition lasts a long time, even after the microcontroller reboots.

Could a diode on the high voltage sense line drop enough volts to keep the controller from overreacting to 50.5 volts while still keeping it from allowing say 65 volts?  Of course, that assumes someone figures out which pin is used for overvoltage sensing.

[myelectricbike]

My SLA batteries come of the charger and settle to 53.6 within an hour. I've never had any problem with cutoff. Someone familiar with the mosfets told me at one time they believed that the mosfets in the 36 volts standard Golden controller were capable of handling 58 volts so that I could power a standard 36 volt controller with 31 to 58 volts.

However, the 48 volt controller handles the extra power much better than the 36 volt controller. Acceleration is faster and top end is higher, etc. so for $78 it costs it seems like modding a 36 volt controller to do the job or a 48 volt controller is an awful lot of work for nothing, especially when it is the motor that needs help. Right now I'm putting the extra effort into upgrading the materials so I can rewind and rewire the motor.

[pdonahue]

My plan is still to buy the 48V controller, but I figure it's going to be at least a few weeks until I get a response from sales@goldenmotor and then another few weeks to ship it to Canada.  I'd like to get the 36V controller working in the meantime...

Pete

[myelectricbike]

Well certainly as back up its good to be able to have a 36 volt controller with you just in case you discover miles from home a dead cell in your 48 volt pack, etc. I'm just saying that its not the optimal plan to put it into service except as a standby unit - but then the effort might also pay off in finding out how to lower the bottom cutoff for the 48 volt controller so that you might not need to take a 36 volt controller along.

[pdonahue]

Here is the pic of the controller.  This site limits the size to 128kb so I had to shrink it down.  I'm trying to track down the rest of the pin functions and hopefully figure out how to stop the controller from cutting out.

[myelectricbike]

Hummm... Since I no longer use the power cutoff breaks I'm thinking of mounting an old fashioned keyboard lock on the end of the controller to deter theft. Do you know the the green LED is for? I can see all sorts of potential for circuit add-ons and modifications.

[pdonahue]

Re: Green LED

My guess would be that it is a status LED..  Just after power up with nothing connected it flashes 6 times then pauses, then repeats.

Pete

[myelectricbike]

Humm... a POST for a microcontroller? Humm...

[pdonahue]

Probably not a POST...  More likely used for debugging and trouble shooting during actual operation...  I've used a single LED for debugging on a few PIC projects that didn't have a serial interface or a LCD.

[pdonahue]

There is no software High Voltage Cutoff!!!  It turns out that it's actually the KA7915 voltage regulator that doesn't like the higher voltage.  It's connected to V+ via a 150 ohm resistor.   It feeds directly into the 5V regulator to feed the PIC and low voltage circuitry on the board.  The 150 ohm resistor is sized such that the voltage drop under normal conditions (i.e. Vin ~ 36V) the actual voltage that the 7915 sees will be OK.  When larger voltage is applied, the 7915 can't handle it.  It works for a few minutes then overheats.  It ends up putting out about 5V in the fault (overheated) condition.   With the voltage drop of the 5V regulator that means that about 3.2 volts ends up being applied to the 5V logic and the controller cuts out.  You then have to wait for the 7915 to cool down (not wait for the caps to discharge as I has originally thought) before the controller will work again.  I added 2 4.3V 5W zener diodes in line with the 150 ohm resistor and now the controller no longer cuts out.  I've only taken it on a short run (~5km) so far, but I'll let you know tomorrow if I make it all the way to work (15km) and back with no problems!!! 

Pete

[myelectricbike]

Inline like this...?

[Mordaz]

Inline like this...?

Probably not...  I guess he placed the zener where you placed the resistor (in your picture).  The catode goes to the 7915 in, and the anode goes to the +B.   

pdonahue, do you confirm this?

[]s
Roberto

[pdonahue]

No, not like that.


Basically replace the (150 ohm resistor):
    
   ---  (150 ohm resistor) ---

   with a series combination of the zeners and resistors:
  
   ---  (Controller V+)  ==> (150 ohm resistor) ==> (4.3V zener) ==> (4.3V zener)  ---

See the zeners in the attached pic.