Grounding of Controller

Electric Motors and Controllers
Post Reply
mj2412   10 mW

10 mW
Posts: 23
Joined: Mar 03 2011 1:47am
Location: Austria

Grounding of Controller

Post by mj2412 » Jun 17 2020 6:18am

Hello, my self brewed motorcycle is using a Curtis 1236 controller. The bike is working, but i recognized a strange behavior at the BMS. After searching at the BMS first i relaized that i have 43 Volt between Battery minus and steel motorcycle frame. When i put a resistor (100 Ohm) between the B- of the controller and the frame the voltage drops to 0. Therefore it looks like a grounding issue. I looked into the electrical schematic from Curtis, but could not figure out how to ground the controller and furthermore is it the correct choice to ground the controller. I could easily take a cable from B- to the frame.
Any advice is highly appreciated as the bike is nearly finished and i want to do some proper test runs.
Thanks Martin

obcd   100 W

100 W
Posts: 161
Joined: Jul 16 2019 3:45pm

Re: Grounding of Controller

Post by obcd » Jun 17 2020 6:44am

I wouldn't connect battery minus to motorcycle frame. If anything else touches frame once you did this, it will cause a short and damage to the electronics. The voltage you measure can come from capacitive coupling of some components mounted on a heatsink. A voltage meter hardly produces any current flow, so it's not unusual to get a reading. You could try with a 100K resistor and see if that's enough to remove the voltage as basically it could mean something is touching the frame. And you sure don't want some live signals connected to your frame.

john61ct   100 GW

100 GW
Posts: 7454
Joined: Dec 18 2018 2:06pm

Re: Grounding of Controller

Post by john61ct » Jun 18 2020 11:40am

I've always seen DC negative returns put in common with the engine block & chassis for mobile rigs, including boats cars trucks etc.

There is of course no true "ground" as in Earth in that context.

Many in fact like to **use** the chassis as negative return path to source, but I always prefer a full circuit pair home run.

But tying all the various DC circuits' negatives to "chassis ground" so they share a common reference I always thought was a traditional "best practice".

User avatar
amberwolf   100 GW

100 GW
Posts: 32109
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Grounding of Controller

Post by amberwolf » Jun 19 2020 6:03am

The grounding of *12v* systems in automotive/etc applications is common, but it also causes assorted problems.

Any wire with damaged nsulation that carries 12v can then short to the frame where it runs next to it.

As long as there is a fuse between this wire and the source, the fuse will blow and the problem stops, and the fault can be located and repaired.

When the fuse is beyond the point the wire shorts, the wire heats and melts or burns, and if it's in a bundle it can cause a fire, destroying more wiring allowing more shorts etc. :bolt: :flame: :cry:

In a crash either of the above can happen easily...but if the frame werent' grounded, the only way a short could happen is if a ground and a 12v source were shorted together by pinching/etc., which is much less likely than just a single wire being shorted to the chassis.



Higher voltage traction packs are *not* grounded to the frame, because of the above, *and* because any insulation was damaged on a positive wire but didnt' short to the frame, and someone touched that wire while leaning on the frame, they'd get full pack voltage across their bodies. :bolt: :kff: :flame: :cry:



An issue specific to ebike/etc hubmotors (or other BLDC types) is that if you have a grounded frame, it's connected to the axle or the motor housing (and thus the stator and windings). If any damage occurs to the wires where they exit the axle or motor housing (very common) that connects a phase wire to the axle or housing, you just shorted the FETs on that phase to battery ground, and probably blew them up. So now you need a new controller....when all that would have happened without the ground is that you'd have damaged wiring, as long as only one phase wire was damaged this way (common enough)

john61ct   100 GW

100 GW
Posts: 7454
Joined: Dec 18 2018 2:06pm

Re: Grounding of Controller

Post by john61ct » Jun 19 2020 8:59am

Thanks for clarifying

User avatar
fechter   100 GW

100 GW
Posts: 15251
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: Grounding of Controller

Post by fechter » Jun 19 2020 10:16am

With EVs, you normally don't want to connect anything to the frame for safety reasons. Getting a voltage measurement to the frame is just from some stray leakage and the fact that it goes to zero when you put a resistor across it shows there is no significant connection. Try using a 100k resistor and see if it still drops to zero.

Anyway, I wouldn't worry about the stray voltage reading and I wouldn't connect the frame to the main battery.
"One test is worth a thousand opinions"

mj2412   10 mW

10 mW
Posts: 23
Joined: Mar 03 2011 1:47am
Location: Austria

Re: Grounding of Controller

Post by mj2412 » Jun 20 2020 12:23am

Thanks for all the answers.
We checked with a oscilloscope the signal at it is the PWM stray voltage from the controller. The standard voltmeter, just reads the mean value. There is nearly no current. Best reduction could achieve with a capacitor between B- and the frame. Generally i would have no issue with the grounding in the air, but the stray voltage has an effect on my BMS display. With the controller on i can not read anything on the BMS any more. I'm using the founding power 24s smart bms. Is not sold any more, but is a nice and cheap BMS in my opinion. Maybe i can solve the issue on this side. Maybe i try to put a steel mesh between both units.
Keep you updated.

User avatar
amberwolf   100 GW

100 GW
Posts: 32109
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Grounding of Controller

Post by amberwolf » Jun 20 2020 12:38am

If the BMS is wireless, you could try enclosing all of the motor cabling with a shield thats' grounded to battery ground.

If the BMS uses a cable running along the frame, try moving the cable away from the frame a little bit. Same if it's running parallel to any motor or controller cabling.

User avatar
fechter   100 GW

100 GW
Posts: 15251
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: Grounding of Controller

Post by fechter » Jun 20 2020 11:10am

A capacitor between the frame and wiring would be OK from a safety standpoint. You could also place a high value resistor across it (100K).
"One test is worth a thousand opinions"

John in CR   100 GW

100 GW
Posts: 15161
Joined: May 20 2008 12:58am
Location: Paradise

Re: Grounding of Controller

Post by John in CR » Jun 20 2020 11:56am

I've had high voltage leak to the exterior case of controllers and also a DC/DC converter before, and some of the leaky connections were enough to give a shock I could feel if I was touching the frame while connecting battery negative. I also felt a little tingle while riding in the rain one time (a bit scary knowing I had a battery just below my butt at over 120V in a downpour and still 10 miles from home), and I know I had no shorts in anywhere in the my wiring or battery to the frame. My solution is to now use rubber or plastic grommets and plastic washers for the bolts and rubber mats to make sure everything is electrically insulated from the frame.

User avatar
fechter   100 GW

100 GW
Posts: 15251
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: Grounding of Controller

Post by fechter » Jun 20 2020 2:45pm

Even with everything isolated you may still get interference to the display just from capacitive coupling between cables. Cable routing may make a difference or you may need shielded cables. Another approach is to use a common mode choke on the display cable near the display. This would be something like a ferrite torroid you pass the cable through and several turns if you can fit them. Avoid having the display cable go anywhere near the motor phase wires.
"One test is worth a thousand opinions"

Post Reply