[DavidAnnArbor]

Back in mid-February I had my level 2 charging station installed in my garage.
The requirements for the station are that it is on a separate circuit, bypassing the house breaker panel. This separate circuit in turn is attached to a separate electrical vehicle meter.
The circuit was grounded.
The circuit for the car is a 50 AMP service, using 6 AWG aluminum wire, XHHW conductivity.
I used to have an old electric dryer plug in my basement which I don't use anymore when I installed a natural gas dryer instead.
I have a 100 AMP service to my house. The central A/C goes on and off periodically at night during the summer.
The electrical work passed inspection.

The first Juicebox I received worked fine from up til June 11th and then all three lights started blinking. I believe I would get three rapid blinks. I unplugged it, and plugged it back in but it would revert back to the three blinking lights. Emotorwerks sent me a replacement. But then after a few days of not being plugged in, I plugged the original Juicebox back in again and it started working.

Emotorwerks suggested I still use the new replacement one instead, and so I sent the first Juicebox back to the company.

The second Juice box worked for around a month and a half, and then it stopped working. I now get 5 blinking lights.
It stopped working sometime in the middle of the night around August 18th, after completing a charge. I've unplugged and waited days at a time and it never started back again.

A 3rd Juicebox is being sent.

Not sure what's going on. And how to solve this problem if the 3rd Juicebox also fails.

I've used a public Level 2 charger to charge my vehicle, and have no problems using those chargers.

 

[NewsCoulomb]

I would have an electrician check out the circuit before plugging another JuiceBox in.

 

[DavidAnnArbor]

The 5 beep error code for Juicebox is:

"Check your ground wire in both the panel, and at the device termination (hard-wired or outlet plugged into) to make sure all connections are tight. If error persists after unplugging and rebooting the JuiceBox, and your car will not charge, please contact support."

 

[XJ12]

Aluminum wire? Have to be real careful about the torque on the lugs if that's the case. Aluminum can develop an oxide that would raise the resistance and might prevent a proper ground. Could be why we're getting fault codes now after several months have gone by. Would immediately get an electrician to check it out.

 

[Pike Bishop]

Aluminum wire? Have to be real careful about the torque on the lugs if that's the case. Aluminum can develop an oxide that would raise the resistance and might prevent a proper ground. Could be why we're getting fault codes now after several months have gone by. Would immediately get an electrician to check it out.

Very true. You would also want the electrician to carefully review grounding and bonding in your setup. They are not the same thing.


Your description of a separate meter for EVSE implies that the 240V circuit for the EVSE needs to have its neutral wire bonded (connected) to an earth or plumbing ground at the meter. Your Juicebox would then be connected behind a new 50A breaker. This breaker box needs to be both bonded and grounded, neutral to ground. This makes Neutral=Ground.


The thinking is that Juicebox, as do many open source EVSE, do not use the neutral wire, assuming that Ground=Neutral. The problem arises if Ground does not equal Neutral due to an error in bonding or other issue. The EVSE bases its GFI as well as internal control voltages on having Ground=Neutral. An improperly referenced ground can cause the alarms you report.


Again, earth ground resistance moves around constantly depending on temperature, rainfall, and mineralization of the soil where the ground rod is. (My suspicion is Ann Arbor, with warmer minimum summer temperatures, is causing some ground drift. You installed in February when mean temperatures were in the 30F range. In summer, you are finding temperatures in the 70F range. Those will change ground resistance.)

You may want to reference your electrician to NEC 250 (National Electrical Code).



Good luck!

 

[DavidAnnArbor]

Yes in reference to this, "Your description of a separate meter for EVSE implies that the 240V circuit for the EVSE needs to have its neutral wire bonded (connected) to an earth or plumbing ground at the meter. Your Juicebox would then be connected behind a new 50A breaker."

The electrician created a ground at the meter, and there was a new 50A breaker.

But I will reference all these ideas when I contact the electrician.

I hope this doesn't become an intractable problem for me.

 

[Pike Bishop]

Yes in reference to this, "Your description of a separate meter for EVSE implies that the 240V circuit for the EVSE needs to have its neutral wire bonded (connected) to an earth or plumbing ground at the meter. Your Juicebox would then be connected behind a new 50A breaker."

The electrician created a ground at the meter, and there was a new 50A breaker.

But I will reference all these ideas when I contact the electrician.

I hope this doesn't become an intractable problem for me.

It should not be a problem, yet you should have your electrician check it out. There are still other possibilities including the Juicebox being defective.



Taking the hypothesis that your electrician made an honest mistake and wired the EVSE disconnect as a sub-panel (a panel connected to another panel through a breaker), the electrician would not have bonded the neutral to ground. This causes a problem for the EVSE.


Here is a diagram from an electrical inspection site.

A subpanel (on the right) has no connection (bond) between a local ground and the neutral.

When Juicebox asks installers to "not use the neutral" and connect only the ground of a 4-wire circuit, Juicebox designers are assuming their EVSE will be connected directly to a breaker in a main panel (middle). In a main panel, ground is directly connected to neutral.


The problem is that all of the electricity coming into your Juicebox is referenced to the neutral line, and really wants to get back to neutral. If you connect a Juicebox to the right hand side sub-panel, you throw a very long circuit for ground to get back to neutral. This may mix up ground faults or affect any 110V used in the EVSE (for 12V DC, for example).


Again, your electrician probably did the correct connection, but if not, correcting it is just a few screw turns. It's worth a check.

 

[DavidAnnArbor]

Thank you Pike Bishop

 

[DavidAnnArbor]

It should not be a problem, yet you should have your electrician check it out. There are still other possibilities including the Juicebox being defective.



Taking the hypothesis that your electrician made an honest mistake and wired the EVSE disconnect as a sub-panel (a panel connected to another panel through a breaker), the electrician would not have bonded the neutral to ground. This causes a problem for the EVSE.


Here is a diagram from an electrical inspection site.

A subpanel (on the right) has no connection (bond) between a local ground and the neutral.

When Juicebox asks installers to "not use the neutral" and connect only the ground of a 4-wire circuit, Juicebox designers are assuming their EVSE will be connected directly to a breaker in a main panel (middle). In a main panel, ground is directly connected to neutral.


The problem is that all of the electricity coming into your Juicebox is referenced to the neutral line, and really wants to get back to neutral. If you connect a Juicebox to the right hand side sub-panel, you throw a very long circuit for ground to get back to neutral. This may mix up ground faults or affect any 110V used in the EVSE (for 12V DC, for example).


Again, your electrician probably did the correct connection, but if not, correcting it is just a few screw turns. It's worth a check.

The electrician came out today and I printed out what you wrote and showed him the diagram also.
In the diagram he said it's not a subpanel, it's its own main panel (main breaker). He said the panel is both bonded and grounded correctly, neutral to ground.
This panel is grounded both to the water meter/plumbing, and secondarily also to the ground rod outside.

He said that the aluminum has anti-oxidant compounds on all connections which he also checked again, i think i saw him adding some electrical tape.

The junction box located above the charger unit has aluminum to copper junctions, and it is dual rated, and he checked the split bull connections and they are all tight, but just to be sure he replaced them with polar taps instead of split bulls. He said the junction box is grounded.

He then suggested that maybe the problem is with the Juicebox itself. He pulled out his phone and showed me a picture of Costco's Siemens brand charger.

Just after he left, the new Juicebox arrived. I installed it, and got it up and running. I am now letting my car charge because it's a bit low, and so far everything is working.

 

[bbg523]

The electrician came out today and I printed out what you wrote and showed him the diagram also.
In the diagram he said it's not a subpanel, it's its own main panel (main breaker). He said the panel is both bonded and grounded correctly, neutral to ground.
This panel is grounded both to the water meter/plumbing, and secondarily also to the ground rod outside.

He said that the aluminum has anti-oxidant compounds on all connections which he also checked again, i think i saw him adding some electrical tape.

The junction box located above the charger unit has aluminum to copper junctions, and it is dual rated, and he checked the split bull connections and they are all tight, but just to be sure he replaced them with polar taps instead of split bulls. He said the junction box is grounded.

He then suggested that maybe the problem is with the Juicebox itself. He pulled out his phone and showed me a picture of Costco's Siemens brand charger.

Just after he left, the new Juicebox arrived. I installed it, and got it up and running. I am now letting my car charge because it's a bit low, and so far everything is working.

I had the exact same issue( 5 Beeps, stopping in middle of charge, reset clears it up for a while) with my Juice box last year when I first got it. Thought it was wiring as well.

The folks at EMotorwerks sent me out an updated unit ( new firmware) and all has been fine since.

 

[Pike Bishop]

In the diagram he said it's not a subpanel, it's its own main panel (main breaker). He said the panel is both bonded and grounded correctly, neutral to ground.

Congratulations! That is exactly correct. It is its own main panel, and you now have a solid ground-netural bond with the correct anti-corrosion. One potential problem down.


Should the Juicebox show another ground or line fault, you would then go down the fault tree to the following:

1. There is source of electrical noise in your home causing the Juicebox to detect a ground fault. This appears on several forums from Tesla to Volt to Nissan. Alleged culprits are leaf blowers, coffee makers, or other motor devices drawing 10-15A on a 110V line in the same home. GFI circuits may only need to see a few milliseconds of noise to trip.
   
   (As this appears to be related to better weather--install in February, trip in summer, is there something that turns on in warmer weather? A dehumidifier cooler or something else?)
2. Juicebox has a sensitive or poorly designed GFI circuit. This has been mentioned in several Tesla forums -- Try another maker.
3. The Juicebox GFI is defective and a new unit will fix it.

Hope it works and you are back on the road. However, high current EVSE appear to be a design challenge for the home market. A designer has to create a blend of lead and feathers: Lead to deliver 7-15kW for hours, and feathers to be so sensitive it will shut off with 5-20mA of leakage current.

 

[DavidAnnArbor]

There is source of electrical noise in your home causing the Juicebox to detect a ground fault. This appears on several forums from Tesla to Volt to Nissan. Alleged culprits are leaf blowers, coffee makers, or other motor devices drawing 10-15A on a 110V line in the same home. GFI circuits may only need to see a few milliseconds of noise to trip.

Hope it works and you are back on the road. However, high current EVSE appear to be a design challenge for the home market. A designer has to create a blend of lead and feathers: Lead to deliver 7-15kW for hours, and feathers to be so sensitive it will shut off with 5-20mA of leakage current.

Wow I don't know really know.

In the summer I typically run the central A/C on at night to cool my house down so that I can get better quality sleep. It will periodically turn on and off maybe around ten times in the course of the evenings. It's usually not on during the daytime.

A google search states that,
"Central air conditioning also runs on 230 volts. However, because they are permanently mounted, central air units are hard-wired into your electrical box. A typical 2-ton central air unit that puts out 24,000 BTUs will draw between 15 and 20 amps.
Current Draw of Air Conditioners | Hunker
https://www.hunker.com/13407208/current-draw-of-air-conditioners"

I asked the electrician if my central A/C could be the problem but he said that the A/C at most is pulling 30 amps, and my charger isn't pulling more than 40 amps so he didn't think that was a possible problem.

My wifi router only draws around 2 amps I think. It's on all seasons. My printer only draws around 4 amps at most.

I turn off my computer at night, occasionally I might forget.

The dishwasher, the gas dryer, the washing machine draw around 10 amps per appliance, or 30 amps if all 3 are running simultaneously.

But what if I had all appliances, my central A/C, and the charger all going at the same time ? That would be about 100 amps. Does overusing the electricity cause the GFI circuit to trip ?

 

[Pike Bishop]

Noise and line fault nuisance trips

Hi David,


It should not be load current, but more noise, short term voltage spikes on top of normally clean 60Hz current that can cause a nuisance trip.


Noise is usually caused by lightning, radio interference, starting a device, smaller motors or low cost appliances. This is a good diagram:

The problem is noise can fool a line fault (Ground Fault) monitoring circuit to believing that there is current leak and shutting off. That may have been the cause of your Juicebox shutting off.



Here is a simplified schematic of the monitoring circuit:

Although this is for 120V, the idea is the sensor is a toroidal (ring) transformer that generates a voltage if the current going into your charge cord does not equal the current going out of the charge cord. A small difference (<20mA) indicates a hazard. The 'solid state circuitry' in the diagram above analyzes the change and trips the relay if needed. The trip has to happen quickly, so the solid state circuitry does not have a lot of time to analyze what is happening.


The design of the 'solid state circuitry' affects the ability to reject nuisance trips.



If noise appears on one of the lines, it can generate a voltage to the solid state circuitry and cause a nuisance trip.


So much for theory, now for practice.

• Your former EVSE cut off with a 3 or 5 blink code meaning 'line fault'
• Your panel and power supply have now been checked by an electrician.
• Your problem is intermittent, and may be related to temperature, or?
• The problem of noise affecting Juicebox has been mentioned in various forums
• In this thread @bbg523 notes the same problem, resolved by upgraded Juicebox firmware
• You have identified multiple sources of noise in your house. However, most of the sources are continuous or were already operating when you first installed the Juicebox.

Again, some options

1. Your new Juicebox has upgraded software and may be more resistant to nuisance trips. Wait an see what happens. If everything is OK, it may have been the Juicebox.
2. The problem reappears quickly and you decide to chase noise. This is hard. You may want to plug your Bolt in during daylight hours, and then turn loads on and off to replicate the fault. (example-Turn on the A/C during the day while the Bolt is charging.)
3. The problem reappears quickly and you decide to find another EVSE that is less prone to nuisance trips. There are many options.

Personally, I would wait and see if the new Juicebox solves the problem.


Good luck.

 

[NY-Rob]

the idea is the sensor is a toroidal (ring) transformer that generates a voltage if the current going into your charge cord does not equal the current going out of the charge cord. A small difference (<20mA) indicates a hazard. The 'solid state circuitry' in the diagram above analyzes the change and trips the relay if needed. The trip has to happen quickly, so the solid state circuitry does not have a lot of time to analyze what is happening.

Back in the day when Juicebox was a kit that you had to assemble (yes... soldering required) they were plagued by false GFI tripping issues.

The assembly manual detailed how the charge cord cables had to pass through the toroidal (ring) transformer at right angles to it's face or you would experience false GFI trips. The manual had detailed photos on how the assembled unit should look to avoid false tripping.


Below is the first Juicebox I built, never had false tripping with it.

 

[DavidAnnArbor]

So much for theory, now for practice.

• Your former EVSE cut off with a 3 or 5 blink code meaning 'line fault'
• Your panel and power supply have now been checked by an electrician.
• Your problem is intermittent, and may be related to temperature, or?
• The problem of noise affecting Juicebox has been mentioned in various forums
• In this thread @bbg523 notes the same problem, resolved by upgraded Juicebox firmware
• You have identified multiple sources of noise in your house. However, most of the sources are continuous or were already operating when you first installed the Juicebox.

Again, some options

1. Your new Juicebox has upgraded software and may be more resistant to nuisance trips. Wait an see what happens. If everything is OK, it may have been the Juicebox.
2. The problem reappears quickly and you decide to chase noise. This is hard. You may want to plug your Bolt in during daylight hours, and then turn loads on and off to replicate the fault. (example-Turn on the A/C during the day while the Bolt is charging.)
3. The problem reappears quickly and you decide to find another EVSE that is less prone to nuisance trips. There are many options.

Personally, I would wait and see if the new Juicebox solves the problem.


Good luck.

Below is the first Juicebox I built, never had false tripping with it.

So far so good but last night the charger turned on and off multiple times.
The first charge last night was the most.
11:03pm start - added 14.592 kwh - ended 1:06 am
1:07am start - added 0.326 kwh - ended 1:19am
2:31am start - added 0.284 kwh - ended 2:37am
3:49am start - added 0.051 kwh - ended 3:51am
5:50am start - added 0.054 kwh - ended 5:54am

Does this seem right to keep starting and stopping multiple times ?

 

[Pike Bishop]

So far so good but last night the charger turned on and off multiple times.
The first charge last night was the most.
11:03pm start - added 14.592 kwh - ended 1:06 am
1:07am start - added 0.326 kwh - ended 1:19am
2:31am start - added 0.284 kwh - ended 2:37am
3:49am start - added 0.051 kwh - ended 3:51am
5:50am start - added 0.054 kwh - ended 5:54am

Does this seem right to keep starting and stopping multiple times ?

David,


How much charge did you need last night? These readings could be perfectly normal if you needed about 1/4 of the battery charged (~15kWh).


My OpenEVSE 50A will do the same thing: Charge to the setpoint (either full or hilltop reserve), and then cycle on and off while the Bolt is connected. The on-off cycles are not regular and seem to deliver <0.5kWh per cycle. It's probably battery charge balancing.


Did you have an error code on your Juicebox? I am doubtful this is a GFI fault. A GFI fault should 'latch' and stay off until you reset the unit for safety.


Take a look.

 

[NY-Rob]

So far so good but last night the charger turned on and off multiple times.
The first charge last night was the most.
11:03pm start - added 14.592 kwh - ended 1:06 am
1:07am start - added 0.326 kwh - ended 1:19am
2:31am start - added 0.284 kwh - ended 2:37am
3:49am start - added 0.051 kwh - ended 3:51am
5:50am start - added 0.054 kwh - ended 5:54am

Does this seem right to keep starting and stopping multiple times ?

That could be normal as it's checking and "topping off" if/as needed.


As Pike Bishop mentioned- if it was a GFI issue it would latch till manually reset.


I think you're good so far....

 

[DavidAnnArbor]

How much charge did you need last night? These readings could be perfectly normal if you needed about 1/4 of the battery charged (~15kWh).


My OpenEVSE 50A will do the same thing: Charge to the setpoint (either full or hilltop reserve), and then cycle on and off while the Bolt is connected. The on-off cycles are not regular and seem to deliver <0.5kWh per cycle. It's probably battery charge balancing.


Did you have an error code on your Juicebox? I am doubtful this is a GFI fault. A GFI fault should 'latch' and stay off until you reset the unit for safety.


Take a look.

That could be normal as it's checking and "topping off" if/as needed.


As Pike Bishop mentioned- if it was a GFI issue it would latch till manually reset.


I think you're good so far....

No error code. I drove 80 miles yesterday so I needed about 15kwh

 

[GJETSON]

So far so good but last night the charger turned on and off multiple times.
The first charge last night was the most.
11:03pm start - added 14.592 kwh - ended 1:06 am
1:07am start - added 0.326 kwh - ended 1:19am
2:31am start - added 0.284 kwh - ended 2:37am
3:49am start - added 0.051 kwh - ended 3:51am
5:50am start - added 0.054 kwh - ended 5:54am

Does this seem right to keep starting and stopping multiple times ?

I get those all summer. Battery cooling/conditioning.

 

[NY-Rob]

No error code. I drove 80 miles yesterday so I needed about 15kwh

Well the numbers add up then


Consumed 15kWh, replaced 15kWh.