DIY Electric Car Forums banner

Keep killing Kilovac EV200 relays

5874 Views 14 Replies 7 Participants Last post by  DanielReclik
In only 200km, I have killed two Kilovac EV200 relays, first a brand new one and now one lightly used.

I´m using one relay as the main contactor for the controller, and one relay mid-pack between battery boxes, controlled by the emergency stop circuit.

This mid-pack relay is about to give up for the second time now, and I cant understand why.

The max voltage of my system is around 360VDC, and max battery current is 400A (DMOC645), so both should be well within the specs of the EV200 relay.

http://www.rec-bms.com/datasheet/Technical_datasheet_Kilovac.pdf

Even for my mid-pack relay, I have a precharge circuit, bypassing the relay through a 500ohm resistor for a few seconds before closing it.
The reason for this are my 3 Brusa NLG513 chargers, that are directly connected to the battery at all times. I don´t know how big the capacitors are in the chargers, but I thought it couldn´t hurt precharging them as well.

Seeing as the EV200 are about 100USD each, I would think they should last longer, no?

Any ideas?

This relay is acctuated once every time I start the car, as the 12V for the emergency circuit shuts of with the key...
But as it is now, I often need to restart a few times before it latches...
1 - 15 of 15 Posts
How are the contactors failing--at the main power contacts or the drive coil?

How is the coil drive circuit energized--is it directly from the key switch contacts or a separate control circuit?

How do you charge the pack--Are the three chargers in series or parallel, do you charge the entire pack all together, or do you charge front and back pack separately?
The first one failed closed in the main contactor. Noticed that one the hard way *bzzzzt* :)

The second one doesnt always want to close it seems. I have to restart the 12V system to get it to try actuate once more.

One side of the coil is directly to ground, the other side is looped through emergency stop button, crash impact sensor and such.

The pack is charged all in one, with all the chargers in parallell. The Brusas are single phase, so I have three to utilize the three phase system here in Sweden for a total of almost 10kW charging power.
Sounds like the first one was operated (opened or closed) while current was flowing and the contacts welded.

For the second one, do you have a free-wheeling diode or resistor snubber around the drive coil to provide a current path when opening the contactor? If not then arcing will have occurred and degraded the emergency switch contacts or punched thru the insulation of the drive coil winding to short it out partially.

So during charging the 12V system must be powered up to close the mid-pack contactor--what maintains the 12V, is it a battery that may be getting run down or is weak?
The first one didn´t have a precharge, since I overlooked the capacitors in the chargers, so it might very well have gotten welded together by the inrush current to the chargers.
Also, I had used it to discharge my entire traction battery cell by cell for a bottom balancing act, and that might have taken some cycles of life out of it.

And I do not have a free-wheeling diode, the electrician here at work asked the same thing. How would such a thing be connected?

I have done most of my wiring according to EVTVs exemple for their VW Thing, but I guess they may not have drawn in all the minor components.

The 12V must be on as you say, during charging, not just for the contactor mid pack to close, but for the JLD404 to count the Ah, and also for the Orion BMS to balance the cells.
The source is a 165Ah LiFePO4 battery, that is also beeing charged the same as the traction pack, so I dont really see it getting weak.
A 500 ohm resistor may be too high ohmic value. You say for a few seconds. What is the actual voltage across the kilovac contacts when the coil is powered? What times that in your start sequence?
The voltage across the contactor should be full traction pack = 360V in worst case.

I´m not totally sure about the value, but 500ohm comes out to 0.72A at 360V, shouldn´t that be enough to charge the capacitors in the chargers in a few seconds?
I must check the actual value before going further into that discussion.

Its the first thing happening in the start sequence.

1. Turn 12V power on.
2. Contactor mid-pack closes after precharge (done with a Omron time relay with switching output. Precharge turns off when contactor turns on).
3. Key is turned, starting the precharge sequence of the main contactor to charge the capacitors in the controller.
4. Wait 10seconds and then floor it! :)
It depends on the total capacitance (F) of your chargers. Worth checking. Analog voltmeter and stopwatch works. Scope is better. Also measure voltage across resistor after transient. Some equipments have bleed resistors across caps which can cause drop across precharge resistor.
Assuming 10,000 uF (battery chargers and motor controller, DC-DC converter, etc...).

0.01F * 500 ohm = 5 second time constant

So after 10 seconds of pre-charge you should be around 300V or so...

However, at the same time those devices might be starting up and drawing some power.

You would have to place a volt meter across the contactor and measure what the voltage is after your pre-charge time. I would guess it is still around 100+ volts, when the contactor is closed.

Typically I like to keep pre-charge resistors under 50 ohms.

The Nissan Leaf uses a 30 ohm pre-charge resistor.
See less See more
WolfTronix said:
Typically I like to keep pre-charge resistors under 50 ohms.
Yes. It's a bit of a beginner's mistake to keep the pre-charge resistor's value high, so the peak current and hence peak power is low. But the resistor suffers the same energy pulse regardless of the peak current. Lower resistance means higher peak current, but for less time. Higher resistance means less peak power but for more time, and there is a greater chance that a human in the loop is getting impatient, and will cut the pre-charge time short.

Of course, there is still a limit to the peak pulse power; too much peak power and the resistor could explode. So it's important to find a pre-charge resistor that is designed to take a large power overload for a short time. TE Connectivity's HS series has a rating of 25 times nominal power for less than a second. I've used these in applications where the peak power is some 4 kW per resistor (and that application used two, one in each leg), and 200 W nominal units were needed. As far as I know (I don't work there any more), they haven't exploded yet. Other power resistors may not have such a high peak power rating.

Edit: with your three large chargers (plus no doubt a DC/DC and controller), you would have a larger pre-charge requirement than typical vehicles. So that seems like a reason for your problems. The flywheel diodes will also help protect the 12 V switches. I'm surprised that these weren't in the example schematic that you followed.
See less See more
Many thanks for all the great answears!

My thoughts on why I opted for a relative high value resistor, was to not put so much strain on the precharge relay.
I now have a industry type 11pole relay, capable of handling 10A or so.

A lower value resistor for the precharge means more current through this relay as well. And I´m actually wondering if it might be this one that has burnt its contactors and sticking a bit, never switching over to pull the main contactor.

I came to this conclusion because one time I say my full battery voltage on my meters, and when applying throttle it went away. This indicates that the precharge indeed had excecuted, charging the caps, both in the chargers and later in the controller, making it think everything was ok to run.

I must try to get some time in the garage this weekend, sorting this mess out!
My thoughts on why I opted for a relative high value resistor, was to not put so much strain on the precharge relay.
I now have a industry type 11pole relay, capable of handling 10A or so.
If it's like this, then it's rated at 28 VDC. This is probably a major source of your problems. As the contacts close and the contacts bounce, they will be arcing, and they likely won't last long. In a conversion I helped with, the pre-charge relay was another EV200 contactor. Way over-kill, but at some $50 from Ebay at the time, they were cheaper than smaller relays that could handle the voltage and 10 A or so of current. Now those cheap EV200s are gone, but EVC500s are similar and aren't too bad at about double the price.

Any relay or contactor rated at at least 400 VDC and about 10 A would do, but these are harder to find than you might think.

To draw a peak of 10 A at 360 V would require a 360/10 = 36Ω resistor; next highest E12 preferred value is 39Ω. That's 13x lower resistance than 500Ω. So your relay wasn't suffering from too much current.

But 10 A @ 360 V is 3600 W. Even with a 25x overload rated resistor like the HS series, you'd need a 3600/25 = 144 W part; next highest value would be 150 W or maybe 200 W. To use a nominal 100 W part, i.e. keep the peak power under 100 W x 25 = 2500 W, you'd want a peak current of 2500/360 = 6.9 A, or a resistance of 360/6.9 = 52Ω, use 56 Ω. Or perhaps 68 Ω for a little more margin, and in case your battery voltage might sneak over 360 V when recently charged.

A lower value resistor for the precharge means more current through this relay as well.
Yes, but that current reduces to zero very quickly. When bypassed with the main contactor soon after, there is zero voltage and current. It's the voltage that is probably killing your pre-charge relay.

And I´m actually wondering if it might be this one that has burnt its contactors and sticking a bit, never switching over to pull the main contactor.
With the 500Ω resistor, it would switch a maximum of 360/500 = 0.72 A, quickly dropping to zero. It will have a lazy life, as long as its contacts are rated for 360 VDC.

I came to this conclusion because one time I say my full battery voltage on my meters, and when applying throttle it went away. This indicates that the precharge indeed had excecuted, charging the caps, both in the chargers and later in the controller, making it think everything was ok to run.
Why later in the controller? Perhaps you separately pre-charge the chargers and controller, so you don't have to have the controller energized when charging.

Be very careful depressing the throttle when the pre-charge resistor is in circuit. With the 500Ω resistor, the power is E²/R = 360²/500 = 260 W, which a 100 W resistor could stand for a second or so perhaps. If you were using say a 38Ω resistor, you'd be dissipating around 360²/39 = 3.3 kW in the resistor. Even one second of this sort of power will cause it to open circuit or likely explode. I've seen it happen and smelled the result for a week :(
See less See more
Don't know if it is useful but I use kettle elements as my pre-charge resisters
We have the same issue. Please try GV241. It will help
1 - 15 of 15 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top