Contactors and their rated voltage.... Can the battery V's be higher than Contactor?

[wiztecy]

So I'm curious if there's someone who may know what's going on here. I have a 2010 Zero DS electric motorcycle which I love but it developed a glitch/cutout at times at 1600 miles. If I let the bike sit for 3 minutes or so it would work. I narrowed it down to what I believe is the contactor which I pulled the battery pack out to access. The stock contactor is rated for 48v and 200AMPS continuous. I called ALLTRAX since I was doing some tuning on their controller and they recommended getting a 400AMP contactor. When I took the contactor apart I could see arching and pitting on the disc which would cause resistance. I ordered a 48V 400AMP contactor but I just took a voltage reading from the pack and the pack is reading 56volts @75% SOC. So why would Zero put a 48V contactor in on a bike where the battery has higher V's? The next contactor up in V's is 72V.

Is it better to run a 72V contactor or am I find with the 48V contactor?

Update:
Looks like if you have a contactor that's rated lower than your battery voltage the contacts can weld and the contactor will then always be on, if the contactor is over your voltage it then may disengage when the voltage drops below the required amount to make the contact. From the Alltrax website and pdf on contactors:

"The coil voltage rating is important – use the
correct contactor voltage rating! Using 24 volts to
drive a 48 volt coil does not generate the required
force necessary to hold the contacts in place during
operation. Too much voltage will burn – or cook - the
coil wiring of which the contacts may not separate
when turned off. Contactor plunger binding or self
arcing will cause controller failure.
Choosing the correct current rating of a contactor
is just as critical as choosing the right size wire for the
motor and battery connections. The carrying current is
determined by the size and plating of the contacts
inside of the contactor. "

 

[tom66]

You sound like you're confusing coil rating and contactor interrupter rating.

The coil voltage is the voltage passed through the electromagnetic coil, which has a dc resistance. Passing current through the coil will engage the contactor. Depending on the contactor application, 12V to 24V is pretty typical. The main reason is that most vehicles have a 12V or 24V bus - like Model S - so it's convenient to use that vs generating a specific voltage. A 48V coil is odd. Most coils however are perfectly capable of operating over 25~50% of rated voltage, as they can dissipate the extra power safely. For example most 12V relays will have an absolute do-not-exceed voltage of around 15 to 20V.

Going by the fact that they appear to be using a 48V coil, the Zero bike might not use a conventional 12V system and instead run everything (lights, computers, fans, etc.) off the unregulated DC bus, which is rather odd...

The other rating is the interrupting voltage. This is the highest voltage the contactor can safely interrupt and still meet lifespan expectations. The interrupting voltage should be higher than the pack voltage.

You also have to account for interrupting power, the maximum power the contactor can interrupt. (Power = Voltage x Current.) There are usually "several-time" ratings (for example, inverter failure, software must abort, high current flowing) as well as a typical rating; the average battery bus contactor will interrupt current when it is close to zero, because the inverter will have been shut down by then and the only current flowing is leakage/parasitics/quiescent draw from standby stuff.

Pitting on the contacts suggests the 2nd is not happening reliably so you are wearing out the contactor quickly.

 

[wiztecy]

Thanks Tom! Yes I was getting confused of the coil and contactor rating. I actually thought Zero was using a 12V coil to trigger the contactor but I looked closer at the contactor and on its side it has:

Coil: 48V DC
Contact: 48V DC 200A

Below are pics of the contactor I pulled with only 1600 miles on the bike. I saw a picture of a contactor where on the contactor terminal that goes to the motor showed signs of burning due to heat which my guess is increase of resistance due to excessive pitting over time.


I found this contactor which I believe is what I'm looking for:

Albright SW200 Style Main Contactor 12V, 24V, 36V, 48V 72V

Rated Current - 400 Amps
Voltage Coil Options – 12V, 24V, 36V, 48V, 64V, 72V & 96V
Single Pole, Single Throw, Normally Open
Magnetic Blowouts
Coil Power – 11 to 15W
Rated Voltage of Contact Circuit - 6 to 80 Volts DC

Here's the contactor I had already ordered where I upgraded from 200AMPS of the stock contactor to 400AMPS, however the interrupting voltage is still below the pack voltage and destine for the same failure or worse over time:
MZJ-400 Main Contactor 36V 48V


Double-make or double-break contact
Rated load—making & breaking 400 Amps
Inrush Current 800 Amps
Voltage: 36 & 48 volt coil
Current: 400amps continuous, 600amps surge (2 seconds)


Here is the stock contactor which I pulled out of the Zero:
MZJ-200 Main Solenoid, 200 Amp, 24V, 36V 48V
Double-make or double-break contacts
These series are normally open
Current: 200 amps continuous, 800 amps surge (2 seconds)

 

[tom66]

It looks like it might be rated at 48V, 200A which means it can interrupt a TOTAL POWER of 48*200 = 9,600 watts. It seems odd that a relay or contactor is limited to total load power (after all, it can only "see" load voltage when open, and load current when closed) but it's related to arc damage from interrupting high power loads. The arcs are hotter and more damaging under higher load powers. The arcs ultimately cause the pitting you see on the contacts.

This means, assuming its interrupting voltage is rated high enough (120V is typical) it can also interrupt 84V, 100A. In the Zero bike application, the current is probably measured as sub 1 amp at the break point so the interrupting voltage is the limiting factor. Note it's typically nonlinear so it might only be 84V at 60 or so amps, and the maximum current is 200A regardless of voltage (12V/24V switching is max 200A.) The contactor is rated at the maximum permissible load.

You should try to find the datasheet for the contactor rather than a website; try contacting the manufacturer.

You could try going up in total load power so a 72V, 200A contactor is likely to last longer though you need to make sure to get the same coil rating. To be sure, you could install the contactor in the bike and use a voltmeter to measure the open and closed voltage across the coil.

But it could also be a fault in the inverter drawing excessive power in standby or a firmware bug - not shutting down inverter quickly enough - have you had any equivalents of "PULL OVER IMMEDIATELY" from your bike (in the Model S, these lead to immediate contactor disconnection, and they can only do that a few times before damaging the contactors - luckily, they replace them when they replace the pack or inverter.)

 

[wycolo]

My bike has 72 volt pack but has a separate 12v battery just for the contactor and lights. The 400 amp controller may also use 12v to operate. I'm replacing the lights with LEDs so to minimize drain on the 12v batt.

Lacking a 12v batt, your bike it seems should have a regulator to run the 48v contactor coil always at 48v from the 72v (or less) pack. Just pick a solid state regulator that can handle the coil draw (and then some).

In its favor it looks like your contactor contacts can easily be resurfaced and pressed back into service. And they probably are free to rotate to even out the surface wear.
--

 

[wiztecy]

Thanks again Tom and wycolo.

Learning more about contactors I found that Zero didn't install any pre-charge resistors on the contactor, there was only a coil suppression diode joining the coil connections. Appears that if a pre-charge resistor isn't installed the hard spike of raw battery voltage running across the contactor can damage the contacts which may be part or all of the issue I'm having. I'm curious now about the voltage running across the coil, however I don't know if that's an issue or not since I don't see any problems with heat or signs of defect across the coil contacts or wires.

Tom about the "pull over" condition, I did have the bike cut out and disconnect the contactor many times. Then in the matter of 3 minutes function again. And the kill switch/kick stand switch would not engage or cause the contactor to function or make a clank. Worse case was that I waited 1 hour for it to work again. My feeling was that it was a defective contactor and looking at the arch that that was the cause of where resistance was high causing it to fail.

I thought there could a bug also with the charge inverter possibly picking up noise as if its charging and cutting off power to the contactor but I'm seeing no error codes/lights indicating there's a fault. From the electric motorcycle forum a member said these older bikes didn't have a pattern of noise / glitches which the 2012/13 bikes did.
@wycolo, I didn't want to sand the contactor surface to have it fail and the need to take it apart in another 800 miles By the way, what type of electric motorcycle do you have?

 

[tom66]

Its more related to the inverter bus capacitance. Pre charge resistors are almost essential for any EV application. Usually it's on a second contactor so are you sure there isn't one? First contactor connects pack, resistor and inverter. After a short time, when vehicle is ready to drive (the DC bus capacitors have charged and all self tests passed), the second contactor bypasses the resistor.

Without the pre charge system destruction of the contactors is all but ensured...

 

[wiztecy]

Yep, no pre-charge resistors. Its pretty simple, one contactor and from there it goes to the controller then to the motor. Inverter is only for charging converting AC to DC to the battery. There's no converter, the battery is DC to the DC (500AMP) controller to the DC 72 Volt motor.

 

[tom66]

Yes, but any inverter will have capacitors on it, so it sure is odd that they did not use a precharge system, unless the capacitance is very small (but that causes voltage sags due to inductive wiring, plus transients and noise - it's rarely done except for very low power levels.)

 

[dhrivnak]

Many such bikes are hand built as most EV are low volume, so it is possible someone forgot to attach it or got lazy. But I agree a precharge resistor is important. If you switch out the Albright are very solid products.

 

[lolachampcar]

Most any contractor will arc if pressed into service doing low source impedance (your battery) charing of speed controller input capacitors. I've not looked at the implementation on my 13 DS but I can not imagine they do not have a pre-charge mechanism.

 

[wiztecy]

Most any contractor will arc if pressed into service doing low source impedance (your battery) charing of speed controller input capacitors. I've not looked at the implementation on my 13 DS but I can not imagine they do not have a pre-charge mechanism.

I'd hope they put a precharge resistor on the 2013's. I suspect they did and same with the 2012s. The 2012's and up had huge improvements in design, power, and reliability (especially with the motor). But you don't know until you take it apart right!

Below are pics of the pack open and you can see the contactor. The 1st one is my pic, this is after I unhooked most of the leads to the contactor. The 2nd one (MBB and contactor.pdf which is in the link below the 1st pic) is one a forum member in the motorcycle forum mailed me as a reference, its in PDF format. You can see the heated / damaged wire right where its connected to the left side of the contactor. No pre-charge resistor, only a diode across the coil.

Talking with Alltrax(controller manufacturer) and EVdrives where I bought my 48V/400Amp replacement contactor (unfortunately chinese made) they confirmed that the 48V interrupting voltage will work fine for the 56 volt battery pack. I was going to bump up to the 72 Volt version but they both confirmed it wasn't necessary. They did confirm that moving up to the 400AMP from the 200AMP will make a difference in longevity and definitly the pre-charge resistor is a must for that without this the controller can get damaged from the voltage spike and over time the contactor will get abused by it. Yes I was very surprised why there wasn't one, I'll ask Zero the next time I stop by there on my way to work in Scotts Valley to see what the deal was and also if they can confirm if they're using one now.

Yes the Albright did look very solid! If the one I ordered earlier hadn't showed up and the confirmation that it should work fine wasn't validated I was ready to pull the trigger to buy it. But now looking at sending the one I got back will be a pain. Hey, I now know for next time.

Here's the one I got:
MZJ-400 Main Contactor 36V 48V

Picked up a 480 ohm 10 watt precharge resistor and will be building the bike back up tomorrow. I did find a comm port on the MBB which is the computer with all the Zero code in it. 4-wires, I'm going to tie into this and have a cable I can connect to when the bike is built. A person was able to connect and modify some of the settings for improved performance. Also you can monitor the bike's voltage as well as seeing how balanced the pack / cells are and how far volt wise their each out of balance which is nice. I wish the Roadster had the voltage reading of how out of balance it is. Even though it say's all 0's in the grid I have a feeling that they could be still out of balance somewhat.

View attachment MBB and contactor.pdf

 

[lolachampcar]

my 13 has the sevcon under the seat and the contractor is MUCH larger and resides between the sevcon and the battery.

 

[wycolo]

> no pre-charge resistors. Its pretty simple, one contactor and from there it goes to the controller then to the motor. [wiztecy]

In that case the pc resistors could be inside the controller? On my bike I need to draw up a circuit diagram. One would assume some sort of step-start provision would soften the initial contact current. I'm not familiar with the term 'pre-charge resistors', seems innately confusing.

Thinking about it now I believe my contactor clunks ON as soon as the ignition key is turned ON. Thus it should not show any arcing at all since user would never turn key ON while at the same time twisting the hand throttle. Thus my Curtis 400A controller does all the current controlling by its own solid-state self. So why is Wiztecy's controller arcing its contacts??

Search for photo of my bike here on TMC, somewhere . . .
--

 

[lolachampcar]

Switching supplies used for DC motors all have very large capacitors on their input. The nature of a capacitor is to have extremely low source impedance which means they will take every bit of current the battery can source when initially connected until the caps charge up. Its this "filling of the caps" that draws the high current and causes arcing. The idea of precharge resistors is to slowly charge those caps for a bit before the contractor closes so Vcap is close if not exactly Vbat.

 

[wycolo]

So your DC motor does not run off the battery bank but off a switching supply? I'm confused; my bike has no charging system at all which makes it simple and easy to understand. Is it possible to separate charging from driving on the Zero? I doubt that Zero provides a block diagram but maybe you have drawn one up. 'Step-start' circuits temporarily insert resistors in the primary of power supplies that use solid state rectifiers to moderate the inrush current. Using 'charge' to describe the same thing is what is confusing ('pre-charge'). Also why are the big caps being connected except at the beginning (and disconnected at the end) of a trip?

I picked up a pair of one farad 25v caps and if I got a 3rd they could be placed in series across my 72v battery bank. Would this improve my instantaneous acceleration without undue stress? What size cap bank does the Zero use?
--

 

[wiztecy]

Here's the page where Alltrax talks about the Pre-Charge resistor:

Alltrax Systems Componets Page

PRE-CHARGE RESISTOR - provides a small trickle charge to the main filter capacitors in the motor
controller. Having these "caps"charged prevents large surge currents when turning on the car.

The motor controller has large internal filter capacitors. When the vehicle is off, these capacitors
discharge. When the Key Switch (KSI) is turned on, a huge amount of current will flow charging
these capacitors. This inrush can arc and damage the solenoid contacts and stress the controller.

Yes, as they indicated above my Alltrax 500AMP AXE (Model: AXE-4855-r7) controller on my 2010 Zero DS has capacitors. After talking to a rep, he mentioned nothing about precharge resistors being in the controller. He stated that I should put one on the controller so these caps don't get damaged over time from the current surge when key on occurs. I'm not sure what the capacitor size is in the AXE controller. I did talk to my friend who's a computer hardware engineer and he thinks that the only possibility that Zero didn't put a precharge resistor on this bike is that possibly they're worried it would have a parasitic drain to the pack when not in use. I'm going to install it and monitor my bike. The contactor is inside the pack so its not easy to monitor the idle current across the resistor nor do I want wire leads hanging down to cause possible issues / short.

@wycolo, possibly you're not seeing arching on your contactor for that your contactor is large enough and spec'd correctly so that it deals with these surges better. What voltage and amp contactor are you running? Also how many miles did you put on your bike so far?

So the DC motor in my Zero DS is fed from: Battery Bank-> Contactor -> Controller -> Motor

Below is a diagram. My contactor however does not disengage when throttle is off. Its always on unless the kill switch is triggered or kick stand sensor detects its down.

Note that Zero has an MBB which is a board with some code that limits acceleration from 0-30mph and it turns it up full power. I believe its also the BMB. I found the serial com port connection inside the pack and making a 4 wire extension that hangs outside of the pack so I may access and modify some settings that are in this MBB to get quicker acceleration off the line and get around their 0-30mph limiter. Just have to play around to find the right lines to tie into, gnd, tx, rx, vcc.

The charge system just charges the battery bank, however if there's a bug, noise, or some sort of voltage spike on the charge line the MBB may see this noise on the line as a plug in / charging state and shut down the bike. When I put the bike together I'll pull the four prong plug that the bike has which connects the charger to the bike since I'm curious if the bike needs feedback from this or if it can be pulled from the system. That way it helps debugging the shutdown condition due to a dirty signal.

 

[markwj]

Not sure if it any use to you, but the OVMS Twizy module now has some code to control the sevcon controller - things like regen severity, acceleration and top speed limit.

 

[wiztecy]

Not sure if it any use to you, but the OVMS Twizy module now has some code to control the sevcon controller - things like regen severity, acceleration and top speed limit.

Very cool! Thanks Mark. My model is the early model which doesn't have regen, just coast, but when I get my later model Zero I definitely will add that. Curious if it will have any effect to the acceleration limits on mine. I'll repost in the electricmotorcycleforum.com and let all the Zero owners know. The newer bikes have a CANBUS where mine does not. I assume this would only work on bikes with the CANBUS, correct?

 

[markwj]

Yes, can bus required. But, looking at the code may help you see what is going on.