Mini-QC Rapid-Charger (RC) Project for LEAF QC Port

My Nissan Leaf Forum

Help Support My Nissan Leaf Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Large datacenter UPS systems use high voltages, but then they always have very large battery systems! I've seen up to 750V!

They are basically bi-directional inverters, almost always to 3-phase 480V. As such, The usefulness of such an inverter for QC is questionable.

I've seen some single-phase UPS inverters around the 200VDC range on ebay for cheap, but anything with voltages over 400V is likely to be 480VAC.

-Phil
 
Ingineer said:
I've seen some single-phase UPS inverters around the 200VDC range on ebay for cheap, but anything with voltages over 400V is likely to be 480VAC.

-Phil
Unless the output shares ground with the input maybe you could buy two and stack them.
 
The Surplus Cell Site "Power Supply" I got is a Best Ferrups 15kW UPS.
I removed it from a decommissioned cell site building and bought it from
the owner off Ebay. He had moved the building to his yard outside Richmond, VA.
Yes, it uses 12 (oops, 10 I mean) lead acid batteries, (120 volts nominal) accepts inputs from 480 3 phase or 208
3 phase or 240 volts, and will output an array of AC voltages also(which could be converted to DC).
There is a similar one on Ebay right now, though pricey $1400 obo, item number 300921244578.
Not sure about the inputs and outputs of that one.
Mine is maybe 10 years older than it. Probably very similar.
It is far from tiny, but the components in it will put out 12kW continuously, with no problem.
It will run with no batteries, so could be shrunk down in size, considerably.
It is completely isolated and has all kinds of certifications for commercial and residential use.
The modern cell sites I see at work use 24VDC power supplies with power wires run in shielded conduit
with a fiber link. I am not sure what cellular equipment used 15kW, but the thing was
used to power something big. The climate controlled building it was in was around 40x10 feet.
I am not sure it is suitable as our power supply, but the components could be repurposed.
I have not had a chance to work with this, yet. I got an automatic transfer switch with it, which I
guess was used with a generator mounted outside the building.
Jim
 
Just throwing this out there, since the topic at hand seems to be sourcing a high-power AC-to-DC power conversion solution...

What about running power supplies in a parallel or serial configuration? Or a combination thereof? Get about 10x 3kW power supplies and you're golden. Control them individually with the main control logic that communicates with the QC CAN bus, which, assuming such digital-control power supplies exist, could use a communications protocol of some kind (I2C, SPI, RS232, etc) to control voltage and current as well as getting the status of each supply. A contactor/relay would control the final step to the QC plug, so the voltage can be cut off the moment the car requests it to do so (filter capacitors would make it difficult to control otherwise).

Tesla is already using a similar parallel topology in their Superchargers from what I understand. Running multiple smaller power supplies in series has its own set of weirdnesses (specifically, isolating the ground, which is generally pretty difficult and a little unsafe), but for something as hard to source as a 240VAC-to-400VDC supply, it might be all we could hope for :lol:

Then again, what does the EV conversion market use for their chargers? Aren't there some power supplies in that segment that would come in handy here more than in a UPS (which are DC-to-AC anyway)?

Also, I saw something a few pages ago mentioning that CHAdeMO must supply what the car requests... really? I thought I've seen 20kW, 240VAC two-phase DC quick chargers before, though I really wish I could find the page I found it on. Either way, that would indicate that the protocol does support under-running the supply. As long as current flows, technically it's still charging. Lower the current (in turn, or by way of, reducing the voltage) still makes it a valid charge condition... only if current flows backwards (which is also supported by CHAdeMO - you know, those "home power" EV units?), then it's no longer charging and I'd imagine it would have something to complain about there... :lol:
 
FalconFour said:
... Also, I saw something a few pages ago mentioning that CHAdeMO must supply what the car requests... really? I thought I've seen 20kW, 240VAC two-phase DC quick chargers before, though I really wish I could find the page I found it on. Either way, that would indicate that the protocol does support under-running the supply. As long as current flows, technically it's still charging. Lower the current (in turn, or by way of, reducing the voltage) still makes it a valid charge condition... only if current flows backwards (which is also supported by CHAdeMO - you know, those "home power" EV units?), then it's no longer charging and I'd imagine it would have something to complain about there... :lol:

I honestly can't figure out what you're actually trying to say here, but there are tiny CHAdeMO chargers. The ZERO motorcycle takes 100 volts / 100 amps (10kW). That might be the perfect LEAF charger, assuming ZERO actually makes a charger that small and you can also pump up the voltage output to 400 volts.

The Fuji charger in San Juan Capistrano is 25kW.
 
I think we need more details about the sequence and voltage/current exceptions
This isolated charger will do 417V, but I am not sure if it can do it without any load attached. (pretty sure it can since it can take a command to run at 0 amps)
I am also not sure if the 417 max voltage is sufficient for the leaf to find as an acceptable start up test voltage.
http://www.zepiaenergy.dk/media/LIC_High_Power_Charger_Spec.pdf



garygid said:
I got this question:

Can you provide a spec you are looking for?
Input is as stated 240v , 50A, what do you need
on output and what control interface do you require?
......

Output voltage a few volts, near 0, to around 450v unloaded,
preferably nearer 500. Ramp up, with no load, and ramp down,
perhaps 1 or 2 seconds each way, under control from a uProc,
peraps a command argument*2 = 510v max (2v resolution).
.....
 
I am having trouble finding technical info on these chargers to prove they are isolated, but I have been told that they are. "indeed magnetically and optically isolated."
Prices very as there are various suppliers operating under various names, it might even be possible to get them direct from the manufacturer.
But the point of my post is requesting more details on the Leaf's expectations in terms of the charger.
I am not even sure if this charger can do the job. IF it can, then we will dig further for prices and more specs.
I think the 8kw version is around $1600


The charger linked is a TCCH charger, which often gets re-branded as Chennec or Elcon.
The manufacturer link is here:
http://tccharger.com/english/Product/T35/52.html

The best info on these chargers is actually here
http://store.evtv.me/proddetail.php?prod=168vcharger" onclick="window.open(this.href);return false;
they include a controller for it, however they do not have the 417V model, though I think they can get them.
 
The first link shows the 8kW unit, but no prices and very few details.

The second says isolated, but they only carry essentially one (2?) models,
at this time: 4kW max, 258v max, 38 amp max for $1900.

I cannot get their CAN-Control manual which might tell us more
about the performance of the unit. It apears that they made it
programmable, but then run as stand-alone at the selected current,
and taper off when a voltage is reached. However, if one could
change the current during operation, it might be useful.

If you can change the current, how long before it responds,
at different load levels?

Likewise for output-voltage-changing at no load.

If we could get an 8kW for the $1600 suggested by the poster,
and mod the controls (like the 2nd might have done), we might
learn something?

Yes, more info is needed.
 
I've not looked at that particular unit, but I have examined several Elcon models. They were all isolated. Nobody (other than Rich Rudman/Manzanita Micro) sells a commercial EV charger that's not isolated. (for good reason)

They are Chinese made, and nowhere near the quality of something like a Brusa, but they work. I'm also not sure how they would fare on a HiPot test of the isolation.

-Phil
 
Has anyone looked at the NetGain charger yet ?

http://evsource.com/pulsar1.php" onclick="window.open(this.href);return false;
 
TonyWilliams said:
I honestly can't figure out what you're actually trying to say here, but there are tiny CHAdeMO chargers. The ZERO motorcycle takes 100 volts / 100 amps (10kW). That might be the perfect LEAF charger, assuming ZERO actually makes a charger that small and you can also pump up the voltage output to 400 volts.

The Fuji charger in San Juan Capistrano is 25kW.

It was in response to this post: http://www.mynissanleaf.com/viewtopic.php?f=44&t=13349&start=10#p304575" onclick="window.open(this.href);return false; - quote:
garygid said:
Since the Chademo vehicle expects the input current (thus power) to follow its requested current, and there "appears" to be no provision for providing substantially less (or more) than requested by the car, or abort the charging session.

Considering that devices exist to connect to the QC port and *consume* power from the car, there must be a whole lot more flexibility to the protocol than we're giving it credit for. ;)
 
Apparently the car could be programmed to support the CAR-to-Home protocol
but we do not know if our American LEAFs contain that programming,or how
to invoke it, right even find out if it is there. We are carefully working with
what we can observe on this "exposed interface" called QC.

The QC handshaking and communication that we observe at various QC machines,
when connected to a LEAF, has given us some insight into the typical protocol
used during a QC, but there is much to the details of the expected protocol
that we do not know. I will explain is simple, non-technical terms. :twisted:

For example, in the "introduction", where the car and QC do their "mating" and
deside if they are sufficiently compatible for intimate relations and the exchange
of vital "fluids", the QC "boasts" of the voltage and maximum current that it
can deliver. The mating "dance" includes a "show me, big boy" demonstration
of the voltage (no current) output capabilities of the QC. Assuming that the (high
voltage) leakage type tests are satisfactory, the mating suddenly gets cautiously
serious.

The car asks for 1 amp of current, and keeps mumbling "give me more", ramping
up to some level of injection that is satisfactory to both participants. Somewhat like
real life, if the car feels under-satisfied for too long, or feels that the suitor is overly
eager, the car can reject the suitor, either with a "no more" request, or something
more brutal - something like cutting it off in mid stream.

The car can quietly remember that there were slight improprieties, or "act out"
and even refuse to consider further couplings, even with gentle, long-time friends.

The car keeps a diary, noting the location of each suitor, in anticipation of
a return engagement. However, it appears that there is no actual exchange
of names, even if it was a "coupling for hire" type of session, apparently
for the traditional privacy reasons.

So, the issues of how long is "too long", and what is considered "over-eager"
is being investigated a bit, but we try to stay on the "satisfied" side of
the ritual, if we can. But, to learn more about the limits of the relationship,
we are a bit "adventurous", and often "try something new". :eek: :D

Yes, I know, I have the remnants of a twisted alien mind!
 
It says "edited". I can only imagine what it was like before the censors applied a couple of fig leaves. :oops:
 
Hmm.. yes.. but does the car actually need to see the "show me big boy!" demonstration of voltage or is this only for the DCQCs self satisfaction?
 
GregH said:
Hmm.. yes.. but does the car actually need to see the "show me big boy!" demonstration of voltage or is this only for the DCQCs self satisfaction?
I presume that the car wants to see the charger supply its voltage, so it *knows* the charger is connected, to test the isolation of the DC voltage *before* it would even allow the charger to connect to its battery pack. The Leaf must have a contactor between DC charging port and battery pack, so it will not close that contactor before it is convinced that the charger is isolated properly on the DC side.
Once the car "approves" of a connection, I expect that there are only a contactor, a shunt and a fuse between the charging port and the battery pack, in other words - as long as the car's BMS (Battery Management System) does not throw a red flag and opens the contactor, you can access the battery pack directly from the HV charging pins - whether to charge or discharge... The shunt resistor allows the BMS to measure how much current is actually going in/out.
Possibly you (the charger/home connect) will need to tell the car's BMS what it is doing and in our present quest for fast charging, you can't ramp up faster than the car allows, because it wants to keep the batteries in a safe area.
I expect that just like it ramps up quickly initially, towards the end of the charging it will also slowly start dialing back the current to keep cells from going over-voltage, until it hits a certain minimum current threshold and it considers itself "done".
For discharge there will be a similar protocol of what is allowed, reducing load towards the end of discharge to keep cells from discharging too deeply, but all of this is theory based on my own experience with (non-intelligent) chargers for various EVs.
 
Oh my gawd I lol'd. :lol: If only forums weren't so fleeting and temporary in nature... that one's a post for the hall of fame!

That said... lol... I know that's how the chargers get started, as I've watched the same thing take place at a Nissan quick charger one of the three times I've QC'd (according to the app). It showed the insulation test, amp ramp-up, etc... in very nitty-gritty terms. Is there any actual experience to support that the charge will crap out if the charger delivers less than requested? Or that the car will request more power than the charger advertises is available? If we're designing a 12kW charger, can't we just tell the car "Hi, I max out at 30 amps, so do it slowly"? I don't think the Leaf would find anything particularly dissatisfying as long as its on-board charger is a putzy 3.3kW unit. I figure any action is better than making it play with itself :lol:
 
Back
Top