Adding a Brusa charger under the hood for '11/'12s

[metricmind]

Hi Jason,

I lost some coolant (about 2 cups worth) when I plumbed my boosters. I minimized it by lifting up only
rear of the Leaf while doing it. The only liquid that got lost was what was drained from the stock
charger, but the liquid in the plumbing remained. I re-filled the system by adding distilled water,
which diluted the solution a bit but it's OK - it will be concentrated enough to handle freezing temps here
and it doesn't freeze in Portland very often. I recently visited Leaf dealership and asked for the coolant,
he told me Leaf uses the same standard coolant as any other Nissan vehicle with ICE engine. I could buy
it there, and I probably will because when I'll put in my 3rd booster, I'll loose some more of it.

As of shutting off Brusa charger - this will be quite trivial to implement for your set up.
You mentioned that once you press a button on EVSE's plug handle but not pulling it
out yet, stock charger quits (pilot signals it to the EVSE) but BRUSA keeps going since mains are
still present. IN a simplest form all you really need is two parts - a CT loaded to a suitable reed relay.
Clamp CT around AC feed to the stock charger, so when it consumes power, relay contacts are closed
and when it doesn't, relay is off. So you just connect NO contacts of the relay between pin 2 (+12V)
and pin 3 (Power Enable) of Brusa charger, that's it. When main charger quits, slave Brusa gets disabled
and quits too even if mains are still present. So you can pull out your handle - no sparks since no current
is flowing from the charger to the battery you're not interrupting it.

As far as reducing AC current to prevent circuit breakers tripping for low power EVSEs: connecting resistor
between pins 8 and 23 will accomplish this - varying it from 0 to 4k Ohm will modulate AC input restricting
it from 0 Amps to full power Amps. R > 4k has no impact. In fact this is voltage driven input, but the other
resistor of a resistor divider is already integrated, so you need only one.

Granted you can have an array of fixed resistors and switch selecting most common "not to exceed" currents:
10A, 12A, 16A for example. Take into account current already consumed by stock charger here.

Let me know if you try any of above and need help; I'll help you to set this up and we'll describe everything
for others folks here to see and know how it is done.

Good job, I admire what you've accomplished so far!

Victor

 

[EVDRIVER]

Yes, this is off topic. So I'll be quick and this is my last reply on this topic.

Members don't have to donate, so it's donations.
Vendors (acting like ones) HAVE to donate or else, so it's invoicing, not a donation.
I'd happily donate if I wouldn't be *required* to "donate" or be banned (and I was).

It's not about $15, I just gave away $300 and have no problems with money,
it's a matter of principle.

Rules are set by someone so they can be changed by the same someone.

Yes, my business exclusively sells BRUSA chargers and not only in the US, I make
no secret of it - it is bluntly said so on my web site. (my business sells to OEMs so
this Leaf forum is very wrong place to promote anything, so I don't and never did.
I'm not posting as a vendor here even though I am one. You keep failing to make this distinction).
This means NO MATTER what I say or post, it may be taken as attention
to me -> to products -> to sales. So I can't say ANYTHING as Victor Tikhonov here
because it might benefit my business and this is WRONG.
I can't talk unless classified as vendor and required to pay (even $0.99) or in general
do something others don't have to. It's just fundamentally wrong.

Let's get back to technical fun stuff. I'm NOT paying unless I act like a vendor.
If you don't like it and unwilling to change your rules, ban me, so be it, forum loses me.
You will sure have sense of accomplishment fulfilling your "small formalities" rules
which is the most important.

Victor, I don't know who your are addressing but I can't ban you. I really don't think you were banned for not donating quite frankly but you don't seem to get that even though I keep telling you this. Just letting you know, I am not the admin so talk to the admin. IMO it seems you would rather make a stink over this then just make a simple donation or PM someone in the beginning. It's not my site and I don't care if you do or don't. If I had your business I would be glad to donate here, look like a supporter and get LOTS of free google traffic and the like, that is the value you clearly don't see. Feel free to post all you want here, clearly you are not banned now.

PS- I don't think you ever would have had to pay a dime you just needed to contact someone to get your ID changed rather than make such a big deal over something you did not even bother to understand by sending a PM, etc. I can request you be put on as a vendor and that you don't have to donate. Vendor is the name but it does not mean you have to be selling things. If you want this then send me a PM but please do not post in this thread.

Thank you.!

 

[KillaWhat]

If at ANYTIME I push in the button on the handle (but still hold it in the socket).. the Leaf STOPS PULLING current.. but the Brusa continues- there's no pilot- BUT DO NOT HOOK IT UP!!

I've pulled out a few times now (lol) and I'm sure there's a slight arc. This is only if you stop during mid-charge.

The same set of circumstances happen without the Brusa.
If you pull the J1772 while the Virgin Leaf is charging, it either will or will not shut the EVSE down based on proximity loss or pilot interruption..
Adding the Brusa does not change that, it just adds more current through the process.
I can't move my J1772 very much before that Contactor in the EVSE audibly opens.

If the Leaf signals an EVSE shutdown, there is no longer current flow to the car in any way.
I think, as a general rule of thumb, that we should stop the charge rather than pull the Hot J1772.
Good habit to get into for when we install our 12KW add-on-packs

So.... My new concern (Thanks Jason :twisted:
Maybe this is a question for Tony.
The rating on my J1772 inlet is 20 amps.
I'm going to be pushing that. :lol:

I see J1772 inlets available up to 70 amps, but cannot seem to determine what the difference is except the gauge of the wires attached to them?
I mean, the pin diameter is set by standard, right?

Finished welding the mounting stand offs on my plate for the bracket, brake tank, and that other tank today.
Busy through Monday, but Threading early next week and final mounting to follow.

 

[EVDRIVER]

Clearly one does not have to have CAN control over the pilot in this situation based on their charging location needs. As long as they have the proper EVSE for the charge rate. If using a lower power EVSE this can be an issue but can be solved depending on the implementation.

 

[GregH]

Now that we can read the EVSE ampacity on 2011/12 OBCs AND see the OBC's charge power, this is no longer an issue. The DD will only allow current from the Brusa (via CAN) if the OBC is at full power and there is spare ampacity as reported by the OBC by reading the pilot.

 

[davewill]

... The same set of circumstances happen without the Brusa.
If you pull the J1772 while the Virgin Leaf is charging, it either will or will not shut the EVSE down based on proximity loss or pilot interruption..
Adding the Brusa does not change that, it just adds more current through the process.
I can't move my J1772 very much before that Contactor in the EVSE audibly opens. ...

I don't think you understand the situation. If you press the trigger, the proximity tells the OBC to stop pulling current, which is required to happen in milliseconds. It does NOT turn off the EVSE. This is easy to see if you put a current meter on the mains. Push the trigger, the current stops, release the trigger and it starts up after a few seconds. The contactor in the EVSE never moves.

Unplugging the J1772 at all while current is flowing, will cause arcing at the J1772 connector. Worse, if the EVSE opens its contactors while current is flowing, you can have arcing or even contact welding occur inside the EVSE. Why is the Brusa NOT connected to the pilot/proximity and obeying them? It that not a operating mode?

 

[EVDRIVER]

Now that we can read the EVSE ampacity on 2011/12 OBCs AND see the OBC's charge power, this is no longer an issue. The DD will only allow current from the Brusa (via CAN) if the OBC is at full power and there is spare ampacity as reported by the OBC by reading the pilot.

Yes, this is a more elegant and proper implementation.

 

[JasonA]

I don't think you understand the situation. If you press the trigger, the proximity tells the OBC to stop pulling current, which is required to happen in milliseconds. It does NOT turn off the EVSE. This is easy to see if you put a current meter on the mains. Push the trigger, the current stops, release the trigger and it starts up after a few seconds. The contactor in the EVSE never moves.
Unplugging the J1772 at all while current is flowing, will cause arcing at the J1772 connector. Worse, if the EVSE opens its contactors while current is flowing, you can have arcing or even contact welding occur inside the EVSE. Why is the Brusa NOT connected to the pilot/proximity and obeying them? It that not a operating mode?

This is 100% correct!

KillaWatt, you have one of GregH's LeafDD's right? If you have someone sit in the car, you stand up front, you'll see that as SOON AS YOU PUSH THE BUTTON, the amps falls to 0.. but hold in the handle.. the EVSE contactor/RELAY still stays closed.. release the button...

The amps will start to flow again into the Leaf.... PUSH THE BUTTON.. amps goes to 0... wait... wait... then pull out the handle.. BAM!! the relay will open on the EVSE (this is where the Brusa doesn't know).. and where Victor just made an elegant solution...

And where GregH and EVDRIVER talked about CAN working.

This is all great info. So now, if someone wants non-CAN.. it's possible, and if they want it.. its doable with LeafDD..

 

[TonyWilliams]

Unplugging the J1772 at all while current is flowing, will cause arcing at the J1772 connector. Worse, if the EVSE opens its contactors while current is flowing, you can have arcing or even contact welding occur inside the EVSE. Why is the Brusa NOT connected to the pilot/proximity and obeying them? It that not a operating mode?

I agree. This is NOT acceptable to not comply with the prox signal. When the EVSE is hooked up, it simply uses a 150 ohm resistor to ground. When the J1772 button is pushed, 330 ohms is added in series for 480 ohms to signal immediate disconnect.

As to pulling more than 20 amps through a J1772, the real criteria is the wire size into the pins.

 

[TonyWilliams]

I think I want two liquid cooled Brusa 16 amp chargers on the Rav4. It will cut charge time from 5 hours to 3 hours.

I would have to put them both on one dedicated J1772, and keep the remaining J1772 for the 40 amp charger. Then, at Roadster stations, I can pull the full 70 amps split to two J1772's.

At RV parks, I'll pull 40 amps from one 50 amp NEMA 14-50, and 32 amps from another.

If that works, I might do another two chargers for 5 total; 104 amps means 2 hour charge at 240-250 volts, but more likely 3 hours from three 208 volts public charge stations limited to 30 amps each.

Ya, that might work for 140 miles range.

Mine will have dedicated cooling, and NOT tap into the existing cooling.

 

[JeremyW]

Next long term goal with the Brusa: figure out how to shut down the original OBC, coolant pumps, and (possibly) the DC/DC for ultra-efficient L1 charging.

 

[metricmind]

Guys,

Per IEC 61851-1 (int'l standard for J1772 compliant EVSE), the EVSE suppose to disconnect (it says "drop contactors") as soon as proximity switch is detected open -
that's whole point of proximity switch - if you push it on the handle EVSE assumes the very next thing you'll do is to pull the plug out, so it
prepares for this action by dropping current to zero. Now, if some one would play with push button opening and closing the proximity switch,
that's a lot of clicking of the relay so my guess is modern EVSEs could implement SSR for this current drop when proximity is detected
(but EVSE's MC stay closed) and mechanical MC on top of that (when CP is lost). So if you press a switch you might not hear any clicking
just because SSR takes role of mechanical contactors. This is my guess, but I know for sure no charging is possible (so no current flow and no sparks if you pull the handle) if proximity switch trigger is pressed. This is clearly defined in the standard.

Please keep in mind, Brusa chargers of course handle pilot properly but they know nothing about proximity switch - proximity detection is rather function of the vehicle
than charger (original intent for proximity was to disable inverter when it's on so there is no way vehicle can move while handle is inserted,
whether powered or not). However it could be integrated into a charger, just Brusa chose not to do it.

You CANNOT connect pilot of Brusa charger in parallel (or any other way) to the pilot of stock Nichicon charger. Each loads pilot line to the
specific voltage (6V) and if both will load it in parallel, it may not be within standard value (6V or 3V) and EVSE will shut off detecting an error.
But Brusa charger must only work as slave doubling current of stock charger, charging only when stock charger is charging and stopping when
stock stops. Stock takes care of pilot alright, so all Brusa should do is to watch stock and duplicate. There is no own pilot needed for that.
Brusa charger has 2 means to stop charging externally while in automatic mode (aside cutting off mains) - disabling by disconnecting PON pin from +12V, by shorting
pin 8 and 23 (power indicator). All you need is to detect if stock is pulling mains current or supplying current into the battery and use this criteria
to enable booster. Of course you can do it over CAN too, but you still have to detect stock charger's condition.

Finally, about EVSE power - it signals to the charger how much it's capable of by modulating duty cycle of the CP it generates.
There is simple formula to translate duty into allowed power. Normally vehicle's BMS interprets it and throttles back charger (over CAN) not to
exceed power EVSE told it it's capable of. In Leaf this function can be part of its BMS or the charger can have this functionality built in.
If you want this, you must use CAN controller (you do already) and it should continuously monitor duty cycle of CP, translate it to the requested power pull,
subtract what stock is already pulling, compute what's left for the booster(s) and finally keep sending CAN frames to the booster to maintain
computed power level. EVSE can modulate it at any time (like if it gets too hot) and the charger must obey by reducing power it draws
(in your case stock + booster combined). But you have no control over stock power regulation, so you just have to watch for allowed total
and modulate booster's power not to exceed this total.

What's the max current rating of inlet's AC contacts of Leaf's inlet is very good question. I wish I knew. It will require some digging - find inlet
model and dig into Yazaki's catalog to look at its spec. You don't want the inlet to get too hot - this is the reason I used external
twist lock 50A connector. I could route power through the same pins as J1772 AC power comes in, but it's too much for 3 boosters I got.
Might be OK for two. One booster should not be the problem, but still someone has to determine inlet's AC contacts rating - it's important for us
to find this out.

Victor

 

[JasonA]

That's a good point Victor and something to note to everyone that will be doing it in AUTO mode "tied" to the Leaf. Having the current sensor pins on the Brusa/ current board watch the Leaf goto 0 amps and then cut off.

For me right now, having set at 80% (well 79%) is a perfect solution and then shut off and the Leaf top off from there. Yesterday while on the phone with you sitting there charging in the car, I didn't realize the Leaf pulls on AVG 800-900 watts with the AC running.

So, stock charger (3.3kw) - injection is 9amps.. Sitting in the car AC running(7amps avg).. WOW!

The Brusa hooked up.. 18 amps avg injection... AC running 15 amps!!

As for the J1772 Yazaki- this is something I've had some concern over. The 2011-2012 has a 20a J1772 connector, but the 32a (7.2kw) J1772 looks no different. Saw one yesterday. The pins are the same size... they have to be to plug into any EVSE.

So the wire size on the back comes into play, the 2011-2012 Leaf uses 12ga wire and we're spliced in 3" behind the plug. So the Leaf is pulling only 16a the whole length, that short distance near the plug is getting the full 32a.

And yesterday in the hot sun, I did a full 7.2kw charge and the plug was cool. Not warm.

Unlike the Brusa :shock:

Here's a quick video as some have wanted to see it ramp up.

http://www.youtube.com/watch?v=Qo5y7cdkEps&feature=youtu.be

Tony... I'm digging the liquid cooled models!! I want a few now too!

 

[JasonA]

And here's Part2 .. The Leaf, the Brusa and Heat... (I just happen to be making a video right when the Brusa hit the sweet spot (79% pack voltage)..

Water cooled units do not apply.. but it got HOT under there!!

It was 93* outside, direct full sun and full 7.2kw charging and I had the AC on in the car the entire time.

http://www.youtube.com/watch?v=lG3iCmRlBAY

 

[davewill]

Guys,

Per IEC 61851-1 (int'l standard for J1772 compliant EVSE), the EVSE suppose to disconnect (it says "drop contactors") as soon as proximity switch is detected open -
that's whole point of proximity switch - if you push it on the handle EVSE assumes the very next thing you'll do is to pull the plug out, so it
prepares for this action by dropping current to zero. Now, if some one would play with push button opening and closing the proximity switch,
that's a lot of clicking of the relay so my guess is modern EVSEs could implement SSR for this current drop when proximity is detected
(but EVSE's MC stay closed) and mechanical MC on top of that (when CP is lost). So if you press a switch you might not hear any clicking
just because SSR takes role of mechanical contactors. This is my guess, but I know for sure no charging is possible (so no current flow and no sparks if you pull the handle) if proximity switch trigger is pressed. This is clearly defined in the standard. ...

I think you are misinterpreting the standard. Pressing the switch isn't the same as "open". Pressing the switch puts a resistance on the pilot, just as Tony said, and in response to that the CHARGER is required to drop the current to zero. The EVSE correctly does nothing at this point. The proximity/pilot circuit isn't "open" until the plug is removed from the inlet. All of this is easy to see using the actual hardware as was described above. If the EVSE tried to interrupt the current, there could be arcing inside the EVSE.

In short, the slave chargers MUST cut off quickly when the release button is pushed just like the main OBC does so that when the plug is removed there is no current flowing.

 

[gbarry42]

Jeremy: To get ultra-efficient L1 charging you have to assume the Brusa doesn't waste any power or need to be cooled. See the fans on the side and see Jason's comment about it getting hot.

Jason, Victor, Davewill, etc: What I don't quite see mentioned is this notion that when you push the latch and pull the handle out, the chargers (when properly set up) drop to zero charge current, but they are still drawing current for their internal overhead. So, you are still breaking some current when the relay drops. Whichever relay--after all, the power's got to be removed somewhere, somehow. Or is this particular thing not an issue? Maybe I'm not being clear. Is the concern still about breaking real charging current or must we worry about the residual current as well?

SIde note...when I read about the latch years ago, I didn't believe it, and had to try it out myself. The only indication of power draw was trying to see the light in the garage change brightness. It was difficult to see, but it was there. The charge current stops when pushing that latch button. And no, there are no relays clicking when pushing or releasing it.

 

[aminorjourney]

Hey guys,

Great videos Jason. Really useful, and great to see the auto ramp up/ramp down. Maybe if you set the charge to 79 percent, it would also allow for a more gentle ramp-down for folks who want to set their LEAFs to charge to 80 percent rather than 100?

Onto the temperature. Maybe you could all get an infrared thermometer on the Brusa when it's charging to see how hot it gets? I'm guessing somewhere that the Brusa says how much air it needs per minute, which should give you an idea of how much air you'll need to vent from the charger on a hot day?

I'm agreeing with you on the water cooled version. It looks like it's a sensible route. But does anyone know if cooling system will continue to work happily with an extra charger plumbed in?

With the water cooled one, there's a little less bulk too, so may be easier to fit two under the hood.

Final question: can the inverter underneath take the heat that the Brusa is kicking out?

Nikki.

 

[metricmind]

Hi Davewill,

I don't think I'm misinterpreting the standard, and I just wanted to clarify something in what you said:

. Pressing the switch isn't the same as "open". Pressing the switch puts a resistance on the pilot, just as Tony said, and in response to that the CHARGER is required to drop the current to zero. The EVSE correctly does nothing at this point. The proximity/pilot circuit isn't "open" until the plug is removed from the inlet. All of this is easy to see using the actual hardware as was described above. If the EVSE tried to interrupt the current, there could be arcing inside the EVSE.

Basically you're correct except for one [might be important] detail: Pressing proximity switch (trigger on the white plug) does not directly put any resistance on the CP (pilot line)! The proximity circuit is totally separated from the CP circuit. The way this works is as follows (suppose charging is already in progress):

- You press [normally closed] proximity switch S1, this electrically opens it. Vehicle detects this condition by measuring voltage on the proximity line.
- within 100ms vehicle tells the charger to stop charging but evidently does not open *different* internal switch S2 (S2 is actually a FET) - that's the one that loads down CP line, not proximity switch S1. The positive PWM voltage supposedly stays at 6V level to let EVSE know that vehicle is still in "Charging" state although charging current is zero, so EVSE doesn't do anything at this point - its contactors are still closed and AC is still applied to the J1772 inlet. If you change your mind and leave the plug in and release the trigger at this time, vehicle detects this and in 3...5 sec it commands internal charger to resume charging as if nothing happened.

You can check this even with Nissan trickle EVSE by monitoring AC current with a clamp meter: plug in your handle and start charging normally, you'll see yellow "charging" light come on (only after you release the trigger) and about 11.2A will be drawn by stock charger. Now press the trigger. The current immediately (in <100ms) will drop to zero because this allows you to unplug without sparks, but EVSE won't know/care and yellow "charging" light still stays on (meaning CP line is still loaded by S2 to 6V (more on this below). Well, this condition is "still charging with 0A" (which technically is different from not charging because of "Not ready" condition). If you release the switch without pulling the plug, in 3-5 sec current will raise from 0A to 11.2A again. You can modulate stock charger's current from max to zero and back by pressing the trigger switch, and yellow "Charging" light will stay on at all times. At my first opportunity I'll put the scope on CP line and check the wave form on it to see of CP line is loaded or released to 9V when trigger is open, but 9V on CP means "not ready yet" and yellow light would be off, but this is not the case. SO I'm assuming CP stays at 6V. My issue is this does not work quite as the spec in the standard says it should work - it says EVSE should disconnect in response to proximity switch. So at my first opportunity I'll put the scope on CP line and play with trigger and let you know what I observe. I'm considering to relocate twist lock inlet in the place of quick DC inlet (which in turn is going to move under hood), so all my wiring will get conveniently exposed again very soon

Victor

 

[JeremyW]

Jeremy: To get ultra-efficient L1 charging you have to assume the Brusa doesn't waste any power or need to be cooled. See the fans on the side and see Jason's comment about it getting hot.

I'm not going to assume that the Brusa doesn't waste any power. I will have to see what it's efficiency is, but I'm hoping for something in the low 90% range. The on board charger efficiency when on L1 is ~78%. This is low because of the coolant pumps that run to keep the charger cool. The Brusa needs cooling too, but those fans use a lot less power than the Leaf's pumps.

I am concerned about the heat when running full out, looks like water cooling may be worth it for some environments! :shock:

 

[JasonA]

I'm not going to assume that the Brusa doesn't waste any power. I will have to see what it's efficiency is, but I'm hoping for something in the low 90% range. The on board charger efficiency when on L1 is ~78%. This is low because of the coolant pumps that run to keep the charger cool. The Brusa needs cooling too, but those fans use a lot less power than the Leaf's pumps.

I am concerned about the heat when running full out, looks like water cooling may be worth it for some environments! :shock:

Again, they state 90.1% efficiency at full 12.5 amp output but it climbs when you lower it. Still running it at 9amp output (9 amps x 385 volts = 3500 watts) or so.. and ChargeStar is showing up to 7% loss at the top so bleeding off anywhere of 2-350 watts or so of heat.

Now I see why Nissan liquid cooled their unit. The charts and graphs are all on Victor's and Brusa site

I don't even consider L1 an option, so much power is wasted running other things. I mean, it would be nice to pull 16-20a 120v from L1 through Phil's EVSE or another unit in a pinch... but then it's still just slooowwww.

I already have a another idea/project in the works!!