Goped ESR high voltage cutoff

Does anyone know how to modify the ESR high voltage cutoff?
Perhaps adding/changing some resistor to the controller board?

I've searched all over for info on it.. The only thing I know for sure is that my goped will cutoff from high voltage at about 30v. I want to run 8cell lipo, so hot off the charger it COULD see as much as 33.6v.
(even though I charge my lipo to 4.1)

Anyone have some info about the voltage sensing on the controller board? Everything I find online points me to gopednation and that site is gone.... If I ask elsewhere, I'm always pointed here. lol.
I dont know if its important, but I have a very early goped, so it is the 1st gen board.

Nothing?
The closest I have found is this... but the poster gave up and bought a different controller:

http://endless-sphere.com/forums/viewtopic.php?f=14&t=49534&start=25

silviasol said:

dnmun said:

irfb 3205 is 55V BVds so that is 13S of lipo and you would use the 63V caps. make sure the leg spacing for the caps will work on the pcb. i suspect it is the same but verify before you order it.

you can find a lot of 13S BMS so you can bulk charge the pack using a 36V lifepo4 charger adjusted up to 54V DC. some are cheaper than others and some are made for up to 13S and some are hacked from 16S down to 13S. but some of these cheaper ones have low continuous current ratings:

http://www.bestekpower.com/481v13spcmbmspcbforli-ionli-polymerbatterypack/

Click to expand...

Sounds good. For now I will be using 8s only, like wesnewell says that should work fine. Later when I add a higher voltage motor I will swap the caps. I assume I should replace all the 35v caps, there are two larger and 4 smaller.

Thanks for the link, looks like a good store. I will order when I purchase the cells. Will I need one per pack or just one total?

wesnewell said:

8s is only 33.6V fully charged so you don't need to change anything other than the HVC if it truly has one. If you could find the LVC circuit I'd change that to 28V or at elast monitor voltage and recharge once you hit 28V under normal load or 29V under no load.

Click to expand...

I think I found them. One is fr9024n and the other ksh127. I made a picture showing how the power is sent thru the HVC. Is it possible to just remove the HVC chip or would it need to be replaced? I assume a BMS will take care of any over/undervoltage.

Will there be any issues with the motor running 8s?

LVC?
http://www.datasheet4u.net/datasheet/F/R/9/FR9024N_InternationalRectifier.pdf.html
HVC?
http://www.datasheet4u.net/datasheet/K/S/H/KSH127_FairchildSemiconductor.pdf.html

dnmun said:

if it actually has a HVC signal it would be a resistor bridge off the input voltage, with the middle going to a spot on the processor and it will have a capacitor beside it going to ground with the bottom leg of the bridge. you can spoof it by changing the equivalent resistance of the bottom leg by added another SMT resistor on top of that one to drop the voltage it would see on the pin.

Click to expand...

That would be an easy solution! Looks like are three resistors it runs thru. Basically the middle pin goes to a 35v cap then to a 4 pin chip, 1 pin to a resistor(red), middle pin to a 35v cap then the red resistor, then top pin to a fuse then two other resistors

Click to expand...

i thought this had been solved already. silvia decided to buy another controller but if someone with this controller wants to pursue finding the voltage divider we can try to figure it out.

dnmun said:

i thought this had been solved already. silvia decided to buy another controller but if someone with this controller wants to pursue finding the voltage divider we can try to figure it out.

Click to expand...

If it was ever solved, the answer has been lost with the database of goped nation.

I am all for perusing it... but I worry about blowing my controller as no one has any in stock, and I commute to work with the goped. Right now I couldnt buy a new one if I wanted to.

Stupid question... but I am not too familiar with the D-PAK style surface mount components. For whats labeled as the KSH127 below, is the entire back of the device the collector? I can see the base pin and the emitter pin.. and if I lookup the D-PAK pads for surface mount devices, I see the entire back is the 3rd connection... So am I correct the back is the collector, or is it floating somehow, with that pin just clipped?

the surface mount mosfets have the underside connected to the drain so the thermal and electrical connection is made to the pad underneath it during reflow.

no collectors on a mosfet. but if it was an npn it would be the collector.

i halfway looked at the right side of that picture from silvia and it looks like that mosfet is a switcher and the 570uH coil is the inductor for a regulated power supply for that thing. but not sure since there is another diode on the other end and it don't know where input voltage and ground is on there.

From the spec sheet I looked at, I thought that was a PNP Darlington Transistor.. spec sheet here: https://www.fairchildsemi.com/datasheets/KS/KSH127.pdf

I dont think it matters anyway. I just took a pic of my controller board. I think I may be starting from scratch, as I appear to have a different board.
This is just the underside... I dont want to try to take the board out right now, as I am at work, and its my ride home. lol.






I think this weekend I will have to pull out the meanwell, put 33.6 volts to the board, and see what happens... and just go from there.
Can anyone tell me how a HVC is probably setup? Off the top of my head I can only think of how to do one with a Zener diode.... but I dont know if what I have in mind is a common circuit design or not.

Thinking some more, I would bet that the microcontroller sees the current battery voltage, and goes from there.
I just found outthe voltage limit is adjustable with the programmer thing that I dont have... so if I find where the battery voltage goes into the microcontroller, and put a resistor in-line with it, it will shift the ADC and let me have that higher voltage... AND probably move my LVC higher too.

the processor will see the voltage from the middle of the divider bridge. it will be in the range of 1-4V i expect but that is not so important as just finding the divider. once we find it then we can adjust the bridge to bring down the voltage to the processor.

My thought has been to remove that red resistor that has the 2 pointing to it in the picture then getting the value of it and maybe with some math and guesstimates of what voltage it would read to that 16 pin chip put another resistor in that would force the chip to see a lower voltage then what it actually is.

Lately there seems to be a ton of controllers on ebay though. I have been searching every day to find a cheaper brushless that handles more watts, I found this brushed one recently. The stock throttle is going to work with any controller, you just need to open it to find which wire is which.


http://www.ebay.com/itm/E-BIKE-Motor-Brush-Speed-Controller-for-Electric-Bike-Bicycle-Scooter-36V-800W-/181607344536?hash=item2a48a42998&vxp=mtr

silviasol said:

My thought has been to remove that red resistor that has the 2 pointing to it in the picture then getting the value of it and maybe with some math and guesstimates of what voltage it would read to that 16 pin chip put another resistor in that would force the chip to see a lower voltage then what it actually is.

Click to expand...

there are several 2's on that picture. one points to the gate leg of that surface mount mosfet and the other to the leg of the cap.

the resistor divider will be right where the power comes onto the pcb from the battery. there will be two resistors in series between the input voltage and ground and there will be a trace coming from the middle of the divider which will run over to the processor. on one of the lower pins, maybe even pin #2.

if someone can take pictures of that so i can read the resistors it will help figure out what to change the resistor value to.

The high voltage resister I believe is the one in the blue box. I think the resistor divider is marked u7. The lines and numbers on the board are the way that the traces go, I suppose it is kind of hard to understand. But on the resister divider, if it is that, at u7 the bottom pin on the three pin side goes to the high voltage resistor, the middle goes to the cap under it then to the high voltage resistor and the top goes to the 101 fuse(not sure if that is a fuse) then one of the low voltage resistors then to the 470 component, then to the second low voltage resistor then to the middle pin on the fr9024n LVC.

that is a zener diode. anything beginning with a U in the label is an integrated circuit.

if either one of you can find where the voltage comes onto the pcb from the battery then take a picture of that area. look for a pair of tiny surface mount resistors which will have numbers printed on them.

the one in the blue box is a zener diode.. Im almost positive its just there to protect the controller incase of an overvoltage condition... anything over 35v will pop those caps.
However it should not fault across at the 33.6v max we need for it to safely run 8s Lipo.

It would be really nice if you could read off the numbers on the diode though.. so I can check the datasheet.. if it turns out to be a 30v zener, I'll have to replace mine before I can continue... but it should be higher than that.

where did you come up with this side about that diode? do you understand the circuit that it is a part of?

dnmun said:

where did you come up with this side about that diode? do you understand the circuit that it is a part of?

Click to expand...

Everything I know is from what I have uncovered reading online. I just received my meanwell powersupply yesterday, so hopefully this weekend I can trace out the controller myself, and test the HVC to see where it is currently set at.

why can't you guys just take picture of the resistor divider sitting on the input voltage? that diode has nothing to do with an HVC limit.

dnmun said:

why can't you guys just take picture of the resistor divider sitting on the input voltage? that diode has nothing to do with an HVC limit.

Click to expand...

He cant because he replaced his controller with an aftermarket one.
I cant as I was using it to commute to work all week, and wont be taking it apart until the weekend... lol

Going to work on it in the morning. I just finished modding the current limit on my new 480w power supply.... so I have something I can connect the scooter to.
Tomorrow I'll lug it inside... as you see from my picture earlier, the board is mounted upside down, so until I remove it, its very hard to take a pic.

I still have both mine in each of my esr's I just don't want to open the pan and mess with all the wiring right now. I am upgrading to brushless but don't have it installed yet. The picture shows pretty much everything, I think the side that is not showing is just a bunch of the small rectangle diodes or resistors(whatever those are). What would it look like?

the divider bridge will be two small 603 or 804 SMT resistors like the tiny ones in your picture. they will be right next to the spot where the voltage comes onto the pcb and one end of the two will connect to ground. there will be a trace going from the middle of the bridge over to the processor.

Well I finally got a chance to get rid of all the opinions online and actually test my own ESR.
After testing, I can tell you the the HVC for MY ESR is 33.09v. This is a solid 3v over what everyone was saying it should be.

Im a little conflicted now.. first of all, I have read from several seemingly reliable sources that at 30v my ESR will cut off. Clearly this is not the case.
at 30v, I can run 8S to 4.125... I was only planning on bulk charging to 4.1 anyway, so this may be OK... now if I balance charge, i need to make sure its not set to 4.15 or 4.2 by accident (as they're often the default values)

So. Do I modify my controller, or just call it good enough and leave it from there?

As for the controller pictures, I took a bunch, but they're huge.. so I have them hosted on my webspace. Someone post up if you can not see them for any reason.


http://fastgm.com/travis/Goped/controller/20141213_202947.jpg
http://fastgm.com/travis/Goped/controller/20141213_203009.jpg
http://fastgm.com/travis/Goped/controller/20141213_203055.jpg
http://fastgm.com/travis/Goped/controller/20141213_203157.jpg
http://fastgm.com/travis/Goped/controller/20141213_203200.jpg
http://fastgm.com/travis/Goped/controller/20141213_203210.jpg
http://fastgm.com/travis/Goped/controller/20141213_203215.jpg
http://fastgm.com/travis/Goped/controller/20141213_203251.jpg
http://fastgm.com/travis/Goped/controller/20141213_203252.jpg
http://fastgm.com/travis/Goped/controller/20141213_203254.jpg
http://fastgm.com/travis/Goped/controller/20141213_203456.jpg
http://fastgm.com/travis/Goped/controller/20141213_203502.jpg
http://fastgm.com/travis/Goped/controller/20141213_203504.jpg
http://fastgm.com/travis/Goped/controller/20141213_203508.jpg

the pictures that helped the most are the last ones with the bottom exposed. it appears in the last picture that over on the left side in the red case there is a small battery charger built into the case. is that how it charges the battery? you plug an AC cord in there? one of the red wires comes from that charger over to the solder pad marked Batt+?

on the back you can see the label Batt+ on that wide solder landing where the two red wires are soldered, then there is another isolated solder plane in the middle and another big solder plane on the left with M1+ label. that is how the current flows from the battery to the positive terminal of the motor. this is for a brushed motor obviously.

on the other side of that metal plate, on the back with the 4 screw holes with the insulating rubber shield, those solder planes are all connected together to the M1- black wires going to the negative terminal of the motor.

those relays on top connect the B+ of the battery to the motor M+ and i think, pretty sure, that is how the HVC and LVC controls are exercised.

this is not built like the normal BLDC controllers we see so it does not have the same type of microprocessor controlling the commutation of a brushless motor. so the way it sees and acts in response to HVC is not clear from looking at those pictures except i can see some traces running across the surface between the pads under the relays. the traces we are looking for will go from that Batt+ pad over to a small 8 pin IC called a dual op amp. it will have a label like 324 or something similar like 327 and the output from that op amp will go over to the transistors on the right side in the first pictures. those big diodes, like D5 are there to capture the flyback voltage spike from the relay turning off.

we need to follow any of the traces going from that Batt+ pad over to the active ICs on the front side so if you guys can look closely at any connection that goes to an 8-16 pin SOIC and has a trace going from that IC back over to the transistors. once we have a clue then we can see what the other side of the input to the op amp is and adjust that so that would raise the HVC.

or we can just solder a wire across the relay terminals and totally defeat the HVC function so you can use it with any voltage. except to go to really high voltage we need to identify the mosfets and other parts.

if you can read the labels, on that side where silvia had labeled and circled stuff, on those ICs then post them up too.

I appreciate the help.
Im going to break this down a little to reply...

dnmun said:

the pictures that helped the most are the last ones with the bottom exposed. it appears in the last picture that over on the left side in the red case there is a small battery charger built into the case. is that how it charges the battery? you plug an AC cord in there? one of the red wires comes from that charger over to the solder pad marked Batt+?

Click to expand...

Yes. the system is broken down into a simple charger board on one side, and the controller on the other. I will be removing the charging part when swapping over to lipo. My charging solutions are all external.
Between the two in the shrink wrap, is a "smart charger" circuit. and a diode.. its all a very simple thing, im not worried about that side as I can just unplug it without issue.

dnmun said:

on the back you can see the label Batt+ on that wide solder landing where the two red wires are soldered, then there is another isolated solder plane in the middle and another big solder plane on the left with M1+ label. that is how the current flows from the battery to the positive terminal of the motor. this is for a brushed motor obviously.

Click to expand...

Yup, simple brushed motor.

dnmun said:

on the other side of that metal plate, on the back with the 4 screw holes with the insulating rubber shield, those solder planes are all connected together to the M1- black wires going to the negative terminal of the motor.

Click to expand...

yup.

dnmun said:

those relays on top connect the B+ of the battery to the motor M+ and i think, pretty sure, that is how the HVC and LVC controls are exercised.

Click to expand...

This, I am not sure about. I think they're controlled by the controller for more than just the cutoffs. I think they're also how the controller "turns on".. kind of like contactors... if they were on all the time, i believe the controller would kill the batteries.

dnmun said:

this is not built like the normal BLDC controllers we see so it does not have the same type of microprocessor controlling the commutation of a brushless motor. so the way it sees and acts in response to HVC is not clear from looking at those pictures except i can see some traces running across the surface between the pads under the relays. the traces we are looking for will go from that Batt+ pad over to a small 8 pin IC called a dual op amp. it will have a label like 324 or something similar like 327 and the output from that op amp will go over to the transistors on the right side in the first pictures. those big diodes, like D5 are there to capture the flyback voltage spike from the relay turning off.

Click to expand...

this is where it gets frustrating. ALL of the ICs have no writing on them. I didnt believe it, so I dragged out the magnifier and everything. they're all shiny..and just black. no visible laser etching. check out the pics I link to below. I couldnt believe it myself. So I have no idea what chips I am looking at, just a vague idea off guesses. EXTREMELY frustrating.

dnmun said:

we need to follow any of the traces going from that Batt+ pad over to the active ICs on the front side so if you guys can look closely at any connection that goes to an 8-16 pin SOIC and has a trace going from that IC back over to the transistors. once we have a clue then we can see what the other side of the input to the op amp is and adjust that so that would raise the HVC.

or we can just solder a wire across the relay terminals and totally defeat the HVC function so you can use it with any voltage. except to go to really high voltage we need to identify the mosfets and other parts.

if you can read the labels, on that side where silvia had labeled and circled stuff, on those ICs then post them up too.

Click to expand...

we cant go too much higher on the voltage anyway... the caps are all rated for 35v. the coils for the relays are for 24v... I dont think I want to overdrive them past 35 either.
in anycase, there are no labels to read. you can really see that in the last image in the list below.


http://fastgm.com/travis/Goped/controller/20141214_203246.jpg -- overall view 1
http://fastgm.com/travis/Goped/controller/20141214_203254.jpg -- overall view 2
http://fastgm.com/travis/Goped/controller/20141214_203744.jpg -- Topside of controller 1
http://fastgm.com/travis/Goped/controller/20141214_203753.jpg -- Topside of controller 2
http://fastgm.com/travis/Goped/controller/20141214_203812.jpg -- Topside top edge 1
http://fastgm.com/travis/Goped/controller/20141214_203836.jpg -- Topside top edge 2
http://fastgm.com/travis/Goped/controller/20141214_203842.jpg -- Topside top edge 3
http://fastgm.com/travis/Goped/controller/20141214_204101.jpg -- Topside bottom edge 1

those 35V capacitors are on the 12V rail that is used to drive those relays. i think that three terminal device surface mount soldered to the surface on the right just below the B+ input is the 12V or maybe 24V regulator U6. Q9 and Q10 are the mosfets that drive the two relays and are controlled by U9. you can see the gate drive traces running from that 8 pin SOIC to the gate resistors and then to the gates of those mosfets.

i can see the programming leads to the controller too and those might allow you to reprogram the HVC.