Accidentally revert the battery polarity

hi,
I accidentally reverted the battery polarity on the ZK1218 controller for a Razor powerrider 360 when replacing a new battery .( my kid is left without his favorite ride now )
And so even if I re-installed the battery with correct polarity the pushbutton throttle doesn't start the motor.

I don't have the schematics or wiring diagram of the controller so it's time consuming to trace down the parts to fix this.

So far, I was able to locate the input cap(1000uf) that has a bulged top as a result of my goofing and replaced with a 470uf cap that I have laying around. I de-soldered the MOSFET and checked for S-D impedance and it seems to be functional as open when G-S is polarized and 0ohm when G-S is discharged.
There is a Shottky Diode that uses the same heatsink as the MOSFET but I don't know its purpose and don't know what to test on that.
There are also a few other low power diodes on the circuit board but they checked out all good.
There are two other 3-lead surface mount on the board that look like some transistors but because of its size, I am not sure how to test those either. Same with one 8-lead surface mount IC.

Any help would be appreciated.
thanks.

Without knowing the details of that specific controller, I can only give you some generalizations.

If you post good clear pictures of the board and components, we might be able to say more specifically what to test.


Often, reversing polarity on the battery inputs of controllers will damage or destroy the low voltage power supply (LVPS) that converts battery voltage down to the 12v and 5v that the brains of the controller need.

There is often a large resistor that runs from the main battery input (or a keyswitch / ignition wire that connects to battery positive) to the input pin of an LM317 (or 7812) whose output then powers the gate drivers and such, and then usually goes to a 7805 regulator that powers all the low voltage stuff like the MCU, hall sensors, throttle, etc.

You can measure for 5v on the throttle, etc, and if there isn't any, you can trace this back to the LVPS and see what is damaged there.

Sometimes, when the LVPS dies under these conditions, it *also* destroys the MCU itself, or if not that, the gate drivers for the FETs. In that case, fixing the LVPS doesn't fix the controller.

Sometimes it's just the input caps blown, like you found already, and replacing all of them fixes it.

thanks for the suggestions. I will check the 5V on the throttle but I don't think there is a LM 7812 or 317 as I couldn't see any.
I attached the picture of the zk1218 controller and it's board.

OK, this is a relay-switched controller (motor power is enabled or disabled by the relay). The FET is probably used for PWM switching of the motor, and the diode (probably two inside it) is to keep voltage spikes from the motor and/or relay from killing the FET.

What they often do is use the ebrake to cut power to the relay, whenever braking is engaged.

Sometimes they use the throttle on/off switch to send power to the relay to engage power to the motor instead, but since this uses a FET it probably doesn't use the relay this way. (often a second relay is used when this is done)

But this is a 14v relay (probably both coil and contacts), and if there's two SLA in series, then it's being used at 24v (28v fully charged), so the coil is being overheated by too much current, and eventually burns out. WHen that happens, the power can't get to the motor anymore.

A quick test of this is to apply 12v across just the relay coil and see if it goes "click". If not, the relay is toast.

If it's a two-battery system, they may be using a 14v rather than 24v relay because the voltage sag on the battery could drop so low under load when nearly dead that it would cutout or chatter otherwise. They may be using it to make it cheaper.

If it's bad you can wire across it's contact pads on the board, to at least test the rest of the system. If you don't care about being unable to cutoff power if something should fail, which might leave the motor running full blast, you can leave it that way.


(Alternately the contacts could have burned and no longer make contact, so it will still click when coil is powered, but can still test by wiring across the contacts and then the system will work).

The razor rider uses a SINGLE 12V battery so I am not sure about some of the points: can you clarify?
1.
"But this is a 14v relay (probably both coil and contacts), and if there's two SLA in series, then it's being used at 24v (28v fully charged), so the coil is being overheated by too much current, and eventually burns out. When that happens, the power can't get to the motor anymore."

2.
"
A quick test of this is to apply 12v across just the relay coil and see if it goes "click". If not, the relay is toast.
if it's bad you can wire across it's contact pads on the board, to at least test the rest of the system.
If i If you don't care about being unable to cutoff power if something should fail, which might leave the motor running full blast, you can leave it that way."
"

I am not sure I want to let the motor as soon as the power switch is turned on if this is what you mean: but this is a good idea to test if the relay is toast.

3. Also do you know what the schottky diode is for? and the PWM is generally found on switching Power supply but I am not sure what its purpose on the controller.
thanks.

forgot to mention: when I replaced the battery with correct polarity, the relay did not click.... so your analysis might be correct but I will test it out and tried to bypass it unless any new suggestion.

motoco123 said:

The razor rider uses a SINGLE 12V battery so I am not sure about some of the points: can you clarify?
1.
"But this is a 14v relay (probably both coil and contacts), and if there's two SLA in series, then it's being used at 24v (28v fully charged), so the coil is being overheated by too much current, and eventually burns out. When that happens, the power can't get to the motor anymore."

Click to expand...

This means exactly what it says. A more detailed step by step to the failure is:

If there are two SLA in series (rather than just one), the relay is being used at 24v (nearly 28v when batteries are full).

If you use a 14v relay at 24v, there is more current thru the coil than it was designed for.

If you put more current thru a coil than it was meant to carry, it will heat up more than it should.

If you put almost twice as much current it will heat up that much more, and generally wont' last nearly as long as it should.

Eventually an overheated wire's insulation will burn or melt, and the wires of the coil short together, reducing their resistance, increasing current, which increases heat; this cascades until the wire itself burns thru or desolders from the connection points.


If it's a single SLA, then the relay is being used normally, and is unlikely to fail that way, though it could still be destroyed by current spikes causing cotnact pitting or welding the cotnacts together.

I am not sure I want to let the motor as soon as the power switch is turned on if this is what you mean: but this is a good idea to test if the relay is toast.

Click to expand...

That's really all this is meant to test for--for those willing to accept the other issues or risks, then it can be used permanently--but not recommended.

3. Also do you know what the schottky diode is for? and the PWM is generally found on switching Power supply but I am not sure what its purpose on the controller.

Click to expand...

PWM is what is used to control motor speed and/or current. If it's literally only a switch for a throttle, then there may be no PWM, or only a fixed amount, and the FET may simply be used to switch the motor on because a relay generally costs more than FET (and drive circuittry) that can do the same job with no risk of burning the contacts. A cheaper relay may weld it's contacts after repeatedly switching under load, leaving the motor always on.

It's also possible it is only PWM'd to limit current at startup from a stop or in ohter too-high-a-load situations, to prevent burning out the motor.

The diode is usually to prevent the voltage spikes the mtoor generates when it's switched off from blowing up the FET(s).

Got it but I am not sure why a relay would burn due to battery polarity inversion( although I am pretty sure you are right that is toast...)

about this:
"PWM is what is used to control motor speed and/or current. If it's literally only a switch for a throttle, then there may be no PWM, or only a fixed amount, and the FET may simply be used to switch the motor on because a relay generally costs more than FET (and drive circuittry) that can do the same job with no risk of burning the contacts. A cheaper relay may weld it's contacts after repeatedly switching under load, leaving the motor always on.
"
Why would we have a relay and a MOSFET if only either one is needed to turn on the motor?

motoco123 said:

Got it but I am not sure why a relay would burn due to battery polarity inversion( although I am pretty sure you are right that is toast...)

Click to expand...

It shouldn't...but it could just happen coincidentally.

Another possibility is that the relay itself works fine, but whatever driving circuitry runs teh coil is polarity sensitive and fried, leaving the coil unable to be powered.

(the little standup scooters I've seen with this general type of controler don't use drive circuitry, just direct switching via the brake lever)

Why would we have a relay and a MOSFET if only either one is needed to turn on the motor?

Click to expand...

The relay is often used to allow the brake lever or pedal (if it has one) to shut off the power to the motor. It can also still cutoff power even fi the FET fails, because FETs often fail shorted (which allows full power to the motor).

It is possilbe to simply have the brake cutoff the drive to the FET instead, but if the FEt has failed that doesn't cutoff power in a brushed motor. (wouldn't matter with brushless, but brshed with full power just keep going)

But without a schematic (or you drawing one up based on tracing out all of the circuitry inside), we can't know exactly how this one is designed, and therefore how it really works.

Ok. In previous replies I interpreted that you meant might somehow connect to the relay directly. But if the board has other driving components which is probably the case as you can see with tiny surface mount transistors ...then it gets more tricky.

Yeah. It's probably a lot easier to just buy a new controller than to fix this one, unless you just like a challenge. I did some poking around but couldn't find anyone that's already drawn up a schematic.

Looks like they're around $15-$20, depending on shipping costs.
https://www.google.com/search?q=zk1218+razor+controller

I guess they get blown up a lot.

Yeah. I actually ordered it
I liked the challenge with the hope to get some help in the forum experts which you certainly did...

I suppose you can compare readings on a working unit with the dead one, if you really want to fix it just for the challenge.

Just don't slip with the probes and short anything out.

As soon as i get it
My frustration is how with all the advances in electronics there is still low cost protection for this scenario as this can easily happen when time comes to replace the battery.
I remember from my long dated electronic class that zener diode could be used but i haven t analyzed it yet...

There's a few easy ways to prevent this kind of thing, and essentially none of the "generic" controllers (any of the commonly available stuff) have any of that protection. There's quite a few common "oopsies" that can happen, none of which are protected against in the least. Costs money both to design them in and to pay for the extra parts--even if it is only a nickel or less per unit, if you mulitply that by 50,000 units, that's a lot of money someone could "pocket" instead.

So...these things get left out. (sometimes there are actually places on the boards for them to be installed, but instead of parts have wires jumpering the pads).

Even relatively expensive stuff designed carefully by and for DIYers, like teh Cycle Analyst, are lacking protection against really common oopsies.

Have to get fairly high up in the cost brackets, out of the DIY market completely, and well into OEM territory, to get these kinds of protections in stuff.