E-MotorWerks JuiceBox - an open source 15kW EVSE

[Levi8than]

772, even "OpenEVSE" needs a reboot every once in a while, the microprocessor sometimes get hung up (usually from power line transient noise issues), its like rebooting your Windows XP machine, probably good to do once in a while

I realize this wasn't directed at my issue. But this is a neat idea. Good idea, maybe memory got corrupted, or we're hitting the watchdog timer.
I rebooted mine, too.
The good news is that my Leaf charged completely overnight, even in the rain.
The bad news is that I left a window cracked and had a puddle in the door handle and a wet seat. Sigh.

I'll take a look at the code and see what might be getting corrupted. Maybe it can be re-factored and have to re-initialize the constants every few times through the main loop. Just remember to mark them volatile sot he compiler doesn't optimize this out.

 

[greenleaf]

Very easy if you have a way of measuring it. But be very careful because there is 230-240 volts live within in this box and it can kill you.

There is a yellow pot on the main board that you turn with a screwdriver to adjust the duty cycle, while the device is plugged into the vehicle.
And you have a choice. You either need access to a scope or a multimeter that measures duty cycle (like this: http://www.amazon.com/Extech-EX330-Autoranging-Multimeter-Thermometer/dp/B000EX0AE4/) or a current clamp (like this: http://www.amazon.com/Uni-Trend-UT202A-Auto-ranging-Clamp-Meter/dp/B005FSSKJA/) or both.

In the trigger wire case, set the vehicle to charge based on a timer so it doesn't begin charging as soon as you plug it in (and risk blowing your breaker or worse).
Then as soon as you plug in, the juicebox will report the current it allows to be drawn over the pilot line.
Following the formula on this page: https://code.google.com/p/open-evse/wiki/J1772Basics, turn that pot until the duty cycle reads 26.6% for 16Amps. Then cancel the timer (or set it whatever you actually need), and let it charge as expected.

Or with the current clamp, you'll need to actually have the vehicle charging. Not waiting to charge. So you'll need to set it conservatively low to avoid blowing the breaker.
This method is less exact and may require a short ramp-up period for the vehicle to react to your change on the duty cycle pin. And just read the current on the clamp and ramp up until you reach 16 amps.

I prefer both methods. Set the duty cycle first. Then confirm with the clamp at a few points during the charge cycle.

Then you're free to set the timer or whatever you need. The Juicebox will re-sample the pot every time it starts so your new setting is locked in. I put a little hot glue on mine. It's a habit I've developed over the years from working on electronics. It's easy enough to peel off when you need to change it, but it's strong enough to keep subtle earthquakes or people bumping the box from turning your knobs.

Or if you're not comfortable doing any of this, if you ask nicely when you place your order for an assembled JuiceBox, the folks at EMW may be willing to set the current for you. It can't hurt to ask.

Thanks Joe. In one of the FAQs I read,

https://www.kickstarter.com/projects/emw/emw-juicebox-an-open-source-level-2-ev-charging-st" onclick="window.open(this.href);return false;

it was mentioned that "you will just set that limit for JuiceBox (there will be a way to adjust this in-field)". I wonder if JuiceBox might be providing a "diagnostics" mode to set this without me getting a multimeter to measure the duty cycle.

 

[Levi8than]

it was mentioned that "you will just set that limit for JuiceBox (there will be a way to adjust this in-field)". I wonder if JuiceBox might be providing a "diagnostics" mode to set this without me getting a multimeter to measure the duty cycle.

I've only dealt with the basic model. For premium, with an LCD screen, it's possible that they may allow you to set this via button presses. I can't comment on that.

 

[Raycreed]

Hey Guys, New owner of both a Nissan Leaf and a EMW Juicebox. Very happy with the leaf, and Definitly like the Juicebox, But I am having a little bit of trouble getting it to charge faster. My leaf is an SL model and has the onboard 6.6kw charger. I have my leaf plugged into my Juicebox, and everything works ok, but it will only charge up at 3.6kw. I checked the voltage and amperage, it was [email protected] amps. when I turn the POT it will go up and down up to the 14.5 amps, but won't go any higher. I have it on a 40a Breaker and all the wiring is correct, but it won't charge faster than 3.6kw, even if I turn the POT all the way to the right. Any Ideas what i should do or reasons why it won't go to 30 amp?

 

[ImpactVector]

The car dictates the amps it wants. So if your over 80% it usually lowers.

Try it when it's 50% and it should go higher in amps.

 

[Raycreed]

I'll try that again, but I've had it charge from 40% SOC, the display will still say charging at 3.6kw and about 6 hours to full

 

[Raycreed]

Yep, just tested it, Currently drawing 14.3 amps @ 55% SOC. it will not charge any faster for some reason, even though on other chargers I've drawn 6000w.

 

[Levi8than]

Yep, just tested it, Currently drawing 14.3 amps @ 55% SOC. it will not charge any faster for some reason, even though on other chargers I've drawn 6000w.

Apologies if I'm hitting anything you've already tried or that seems obvious, but I'll just throw a few ideas out there.
Do you have something that can read duty cycle? There are inexpensive multimeters on Amazon that do.
If so, 14.3 amps = 23.8% duty cycle. It's possible you have it set here.
If you're expecting 6.6 kwh @ 240 volts, you'd want a duty cycle of 45.8 or greater. I think the juicebox maxes out the duty cycle when you turn the set knob all of the way left (to zero). This is a hardwired constant. You could try this.

Next thing to check is the voltage of the line before connecting to your car and while the car is charging. It's possible that the voltage is drooping too low, that the car is limiting its amperage pull in order to not droop your voltage.

If both of these check out, then I've got nothing.

 

[GlennD]

You do not list a location. If you are in Southern California I can loan you a Nick Sayer EV Simulator II that will read the pilot and list the max current.

 

[Raycreed]

I wish i was in California! but no, I'm in Cincinnati, where an EV is basicly alien. I think whats happening is something is limiting my current, either programming in the juicebox (firmware) or or maybe (hope not) something with my wiring. My 240 line is 10 AWG @ about 6ft from my power box, but its solid insulated wire from the breaker box, not stranded wire. I have the juicebox on a 10-30 plug/socket , with 6 AWG from the plug to the evse. 3.6kw isn't bad, especially compared to the Trickle charger, but it would be nice to charge at 6.6kw..

BTW, my juicebox is version 8.3 if that matters.

 

[Raycreed]

also Levi, none of the suggestions worked that you mentioned. I'm going to try and wire up a 6-50 welders plug to it and see if it works on that outlet at 6.6kw, that way, I'll know its something at my home instead of the charger.

 

[ImpactVector]

Did you get the fdti cable? You can reflash the unit to use a different base amps.

Would be interested to see what it shows up on the screen

 

[Raycreed]

nope, just the base unit with 70a j1772

 

[Levi8than]

I'm going to try and wire up a 6-50 welders plug to it and see if it works on that outlet at 6.6kw

Your 30 amp 10-awg dryer plug + wires are fine. And your distance to breaker is so short that you shouldn't even have noticeable drop in voltage at peak amperage. Don't waste your time changing the plug; it will not change anything -- unless you have it wired incorrectly so it's actually running off 120, not 240. That would explain things. But even then you wouldn't get 3.3 kw, more like 800 watts.

You should double check this: 240 volts AC across the relay pins. 120 from the ground lug to either leg of the terminal.

Also, if your voltage remains similar before and during load, then you have issues with your wiring. Meaning that you should read 240 volts while not charging the car, and above 230 volts while the car is charging. If it's different from this, please post your numbers.

I doubt re-flashing will help fix this issue, because it's been working correctly for many iterations of the design. I'd rather look into a possible flaw in your setup.
It's possible that the wrong size resistor or pot was placed on the board, making it impossible for you to set the duty cycle properly.

These little screwdriver-turned pots are also quite fragile. I've had cases int he past on other projects where I've turned them beyond their operating range and then had them behave erratically for the rest of their life until replacing them. Yours may simply be no good. You might want to contact the EMW folks directly with your issue. Maybe they've seen it before.

 

[wwhitney]

My 240 line is 10 AWG @ about 6ft from my power box

Your 10 AWG copper wire limits you to using a 30 amp breaker, which means your EVSE should be set no higher than 24 amps (80% of 30 amps). So even if it were working properly, you shouldn't use it for 6.6 kw charging. To do 6.6 kw properly, you'd need to run #8 copper from a 40 amp breaker to a 6-50 receptacle.

Cheers, Wayne

 

[Raycreed]

Well, hopefully I'll figure this out sometime. It only drops 2-3 volts when charging. from 243v to 241v under 14.5 amps. both Legs of the 220v line are 121.5 under load from the ground peg. I've only got a fluke 77 and haven't got a multimeter that can check duty cycle yet, hope I can borrow one soon so I can check the pilot line.

Thanks for the tip wayne, maybe that's my problem all along, and its somehow just limiting me to the safest current my wire can run at the moment. (voltage detection, etc)

 

[Levi8than]

Thanks for the tip wayne, maybe that's my problem all along, and its somehow just limiting me to the safest current my wire can run at the moment. (voltage detection, etc)

Nope. That won't limit your current. Beyond voltage droop (which is nearly nonexistant), there's no way for the devices to know what breaker you have at the box. This has nothing to do with why your charger isn't working. But I'll elaborate below.

Wayne is referring to the NEC requirement that you de-rate your lines and breaker to 80% when running under continuous load. Continuous load would be something like lights. The charging of your vehicle is on the edge of what might be considered continuous, since NEC defines a continuous load as one that runs at a constant amperage for 3 hours or more.

So with your leaf you have a choice. Taking Wayne's advice is a wise, safe choice that follows code to the letter. With the 30 amp circuit, you must derate it to 80% = 24 amps under continuous load, charging at only 5.7 kW. Which I think is close enough to peak to avoid running new wires through your wall. Or run new beefier wires and a larger breaker then keep your 6.6 kW limit.


(*disclaimer: use the following information at your own risk. This is just my opinion, not anything coming from a qualified expert)
Or, if like me you disagree with the theory that this is such a big deal, you can do the math.
Your leaf will only charge at 6.6 kW until 80%. After that it goes into trickle mode at a reduced current load.
So 80% of the 24 kWh battery = 19.2 kWh. 19.2 kWh / 6.6 kW = 2.9 hours. Just less than 3 hours, the definition of continuous load.

And even then, while running at continuous load, 6.6 kW will only pull 27.5 amps, 91.6% of rated capacity.

So honestly your wires can handle a sprint of more than the recommended limit of 3 hours at 30 amps, more than 4 hours at 27.5 amps.
And even at that, the recommendations are conservative. So take my recommendations with a grain of salt and form your own opinion.

But no matter what you choose, your testing in the short term is safe and not violating any code of any kind.

 

[QueenBee]

Wayne is referring to the NEC requirement that you de-rate your lines and breaker to 80% when running under continuous load. Continuous load would be something like lights. The charging of your vehicle is on the edge of what might be considered continuous, since NEC defines a continuous load as one that runs at a constant amperage for 3 hours or more.

Just FYI they require an EVSE to be considered a continous load.

Anything worth doing is worth doing right. Just run the correct sized circuit....

Sure sounds like an issue with the duty cycle.

 

[GlennD]

I am an openEVSE person not a JuiceBox user. It seems to me that a Nick Sayer EVSE Simulator II would be useful for JuiceBox setup.

The simulator would be connected to ground and the pilot. With the dip switches off it would monitor the charging and report the max available current. You can do the same thing without a car hooked up by setting the dip switches to simulate charging. You could set the pot to your desired current.

I wish i was in California! but no, I'm in Cincinnati, where an EV is basicly alien. I think whats happening is something is limiting my current, either programming in the juicebox (firmware) or or maybe (hope not) something with my wiring. My 240 line is 10 AWG @ about 6ft from my power box, but its solid insulated wire from the breaker box, not stranded wire. I have the juicebox on a 10-30 plug/socket , with 6 AWG from the plug to the evse. 3.6kw isn't bad, especially compared to the Trickle charger, but it would be nice to charge at 6.6kw..

BTW, my juicebox is version 8.3 if that matters.

 

[GlennD]

Here is a link to Nick's site.

https://squareup.com/market/nick-sayer" onclick="window.open(this.href);return false;

I also have 1 for sale in the personal for sale thread. ( shameless plug ).