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.
valerun said:
Have you powered up yours yet?
Of course, how else would I know that I can change the output to a (measured) 30V - 53V
But this is still only a single unit - before I run more in series, I will measure the timing of error to fail output active
and the timing from kill input active to no power output, so that I have an idea if that combination will protect fast enough
to avoid blowing up a stack of series supplies if one detects a failure (for example a current overload) and shuts down.
I can turn one pot and create a "fail" condition, so I want to see how fast things respond, then design a circuit to
protect the series string, probably combined with controls for output power off/on and voltage control.
 
Cor said:
valerun said:
Have you powered up yours yet?
Of course, how else would I know that I can change the output to a (measured) 30V - 53V
But this is still only a single unit - before I run more in series, I will measure the timing of error to fail output active
and the timing from kill input active to no power output, so that I have an idea if that combination will protect fast enough
to avoid blowing up a stack of series supplies if one detects a failure (for example a current overload) and shuts down.
I can turn one pot and create a "fail" condition, so I want to see how fast things respond, then design a circuit to
protect the series string, probably combined with controls for output power off/on and voltage control.

Yes, I meant in series ;-)

Regardless of the control scheme I'd suggest you put in reverse diodes. I'd use something like http://www.digikey.com/product-detail/en/STPS30SM60ST/497-12331-ND/2827147" onclick="window.open(this.href);return false; (60V 30A Schottky with forward drop of 0.6V at 30A) - assuming you can have an isolated sink for it. If mounted in the exhaust path of the cooling air, sink does not have to be large at all. This assumes that you'd want to have a limp mode in case of one supply failure. If you don't, I'd just use Phil's suggestion of a small 3-5A Schottky without a sink.
 
TonyWilliams said:
http://www.andromedapower.com/uploads/AndromedaPower-PressRelease-14Sep2013.pdf

Interesting...50 kW in this size makes one wonder why the public QC stations are so bulky?
Seems they can only pack 5.14 w/inch^3...so there is really technology out there that can achieve 10 times that ????

Or is this different technology, since it appears being meant for high-voltage DC2DC transfer of energy?

I wonder if this kind of commercial application will eventually make private use of QC economically feasible?
 
Most of the folks on this thread know Luigi who built this in Costa Mesa.

It's been around a while now, but this is the smallest package yet. This is the concept that I had last year when I drove the LEAF from Mexico to Canada... Andromeda was one of my sponsors.

The unit I intended to take would require the rear seats be folded down. Not a big deal on that trip, but the bottom line was that the CHAdeMO certification was not yet complete on that unit, called the "ORCA".

This one is far, far smaller. Fully CHAdeMO compliant and certified.

You could stick this in a UL 50 enclosures for electrical equipment for indoor or outdoor locations, Types 4, 4X, 6 or 6P and put it anywhere.

Here's some modular 3R ones that I like:

http://www.amprod.us/download/Modular_Enclosures.pdf" onclick="window.open(this.href);return false;
 
Cor said:
Of course, how else would I know that I can change the output to a (measured) 30V - 53V
But this is still only a single unit - before I run more in series, I will measure the timing of error to fail output active
and the timing from kill input active to no power output, so that I have an idea if that combination will protect fast enough
to avoid blowing up a stack of series supplies if one detects a failure (for example a current overload) and shuts down.
I can turn one pot and create a "fail" condition, so I want to see how fast things respond, then design a circuit to
protect the series string, probably combined with controls for output power off/on and voltage control.

Since Ingineer mentioned that these can be probably controlled digitally, does anyone have any idea how? On one pinout description for these type of units I saw SDA and SCLk pins (on the little pin matrix between the larger connectors, see here http://s1075.photobucket.com/user/moneypit2/media/a4875637-221-SunFireX4100X4200M2pinout_zpsb18710d7.jpg.html" onclick="window.open(this.href);return false;).
 
TonyWilliams said:
Most of the folks on this thread know Luigi who built this in Costa Mesa.

It's been around a while now, but this is the smallest package yet. This is the concept that I had last year when I drove the LEAF from Mexico to Canada... Andromeda was one of my sponsors.

The unit I intended to take would require the rear seats be folded down. Not a big deal on that trip, but the bottom line was that the CHAdeMO certification was not yet complete on that unit, called the "ORCA".

This one is far, far smaller. Fully CHAdeMO compliant and certified.

You could stick this in a UL 50 enclosures for electrical equipment for indoor or outdoor locations, Types 4, 4X, 6 or 6P and put it anywhere.

Here's some modular 3R ones that I like:

http://www.amprod.us/download/Modular_Enclosures.pdf" onclick="window.open(this.href);return false;

This is NICE! Do we have an idea on pricing?

V
 
valerun said:
This is NICE! Do we have an idea on pricing?

V

I just replied to Luigi in an email asking for the price. He claims he can hook this directly to my Rav4 EV battery (like he did with the LEAF) and I would have a two way charger; charging my car from the grid, or charging your CHAdeMO car from my battery!
 
Guess I posted in the wrong thread: http://www.mynissanleaf.com/viewtopic.php?f=26&t=11076&start=10#p326195" onclick="window.open(this.href);return false;
My guess is the price will be similar to the Mobile and Rescue, ~$24k. It has two CHAdeMO plugs and a lot of cable!

TonyWilliams said:
valerun said:
This is NICE! Do we have an idea on pricing?

V

I just replied to Luigi in an email asking for the price. He claims he can hook this directly to my Rav4 EV battery (like he did with the LEAF) and I would have a two way charger; charging my car from the grid, or charging your CHAdeMO car from my battery!
 
garygid said:
What is the weight of ORCA?
Can it be operated and get adequate air cooling
as pictured in the cargo space of the car, or
does it need to be lifted out for use?

ORCA is about 160 pounds. Most of that weight is the CHAdeMO cable and the huge heat sink.

I would imagine that you don't have to remove the unit to operate it (particularly if it is hard wired to the battery in the LEAF), but that's one heck of a heater!

Probably just leave the rear hatch open?
 
TonyWilliams said:
I would imagine that you don't have to remove the unit to operate it (particularly if it is hard wired to the battery in the LEAF), but that's one heck of a heater!

Probably just leave the rear hatch open?
Figure 50 kW max and 95% efficient, so about 2500W max of heat to dissipate worst case. Probably OK as long as the heat sinks are exposed and trunk is open.
 
drees said:
Figure 50 kW max and 95% efficient, so about 2500W max of heat to dissipate worst case. Probably OK as long as the heat sinks are exposed and trunk is open.
Is this an isolated architecture?
I am guessing it is, since it can run from 400V AC input, so it is unlikely that it can achieve 95% efficiency at full load.
At partial load possibly (the often-quoted marketing claim "up to 95% efficient") but at full load it would surprise me
and be a darn good design if it achieves much over 90%.
So, count on a 5kW heater, which is still nothing compared to a combustion engine which easily achieves 200kW heat output while delivering not much more than an average 50kW usable energy, but that is a whole 'nother discussion. So yeah - leave the trunk open and make sure the vents on the unit are not blocked.
 
Back to isolated regulating front ends for a stack of nine 48v
power supplies (1500 or 3000 watt each, for around $500
for nine, surplus). But, not a long-term dependable supply.

Are there any efficient 12 kW solutions available?

If not, then we might "waste" a lot of time reverse engineering
these surplus supplies to make them controllable.
Also, having to control nine supplies might be tricky.

Much simpler, with usable results much easier and
quicker, to have one regulating front end.
Any designs available?
 
Hey guys I'm really interested in this project, but not electrical engineer enough to help with development.

I think that by the onset of 2014 winter I will probably need to charge my 2012 more often and quicker than my stock level 2 can do the job. As of now a Brusa would do it, but your approach appeals to me more (no mods to the car).

Anyway, thanks for all your great work! I'll buy everyone a round when it's finished ;)
 
garygid said:
...If not, then we might "waste" a lot of time reverse engineering
these surplus supplies to make them controllable.
Also, having to control nine supplies might be tricky.

Much simpler, with usable results much easier and
quicker, to have one regulating front end.
Any designs available?
How many do you really need to control? I agree you need to be able to turn all nine on and off (i think you'd need THAT anyway), but you should only really need to control voltage on one (or maybe two?) of them. You should be able to get any voltage you need by turning on the correct number of supplies and then dialing the one variable supply in to get to the target voltage. You would also have to be able to shunt around a turned off supply. Power output would be down at any voltage that has a supply turned off, but is that a problem?
 
garygid said:
Back to isolated regulating front ends for a stack of nine 48v
power supplies (1500 or 3000 watt each, for around $500
for nine, surplus). But, not a long-term dependable supply.

Are there any efficient 12 kW solutions available?

If not, then we might "waste" a lot of time reverse engineering
these surplus supplies to make them controllable.
Also, having to control nine supplies might be tricky.

Much simpler, with usable results much easier and
quicker, to have one regulating front end.
Any designs available?

Gary, I mentioned earlier that we are now working on one. Feel free to ask detailed questions here if needed. I am hoping to be testing the first version this coming Sunday (at 5-6kW to start with). I'll admit this is an R&D project and not a finished design but if it works, your existing solution will work as-is. I have just started a thread on this on DIYecar, as well.

Valery
 
Valery,
Great, glad to hear that you are making progress. :D

Is this a front end to a stack of nine almost
"unmodified" server power supplies?

Are you using the 1500 watt supplies?
How many in the stack?
Any mods necessary?

What modifications did you find necessary
to series-stack the server power supplies?

This would be a very useful solution, allowing
you to offer an iso-regulator product or kit, and
provide a very reasonable path for those who
want to build and try a mini-QC solution.

Perhaps a block diagram of what you are
attempting to do, and a link to your DIY thread?

We just got a 3kW 48v server power supply to investigate.
Perhaps just scaling up the important parts of this type
of power supply would work, paying attention to efficiency issues?

I am working on the components of the Arduino Due
controller, QC interface, logger, and user monitor/input.
I am making progress with the CTE32HR touch display,
but having problems getting the CTE50 to do "touch" input.

Thanks, Gary
 
garygid said:
Valery,
Great, glad to hear that you are making progress. :D

Is this a front end to a stack of nine almost
"unmodified" server power supplies?

Are you using the 1500 watt supplies?
How many in the stack?
Any mods necessary?

What modifications did you find necessary
to series-stack the server power supplies?

This would be a very useful solution, allowing
you to offer an iso-regulator product or kit, and
provide a very reasonable path for those who
want to build and try a mini-QC solution.

Perhaps a block diagram of what you are
attempting to do, and a link to your DIY thread?

We just got a 3kW 48v server power supply to investigate.
Perhaps just scaling up the important parts of this type
of power supply would work, paying attention to efficiency issues?

I am working on the components of the Arduino Due
controller, QC interface, logger, and user monitor/input.
I am making progress with the CTE32HR touch display,
but having problems getting the CTE50 to do "touch" input.

Thanks, Gary

Thanks Gary.

It is effectively a replacement for our current buck output stage. You would connect it to the output of our PFC stage (or any suitable 400-450V DC source). The control scheme for this stage will be similar to the buck stage you are using now, with some differences on the duty cycle range, etc.

The final product here would be a kit (or assembled) version of the 12kW (in the beginning) to 25kW (real target) 2-stage supply: stage 1 PFC - our current 25kW PFCDirect unit, stage 2- a regulated isolation stage. This keeps the component count to a minimum.

Note that you can already use our non-isolated buck stage (the output stage of the supply you are currently using) with a stack of server supplies to get isolated system. So it's already a viable approach. But as has been mentioned above in this thread, it is (1) less elegant, (2) less efficient, and (3) less scalable beyond used supplies inventory.

IMO, the best approach is a 2-stage system I described above (the one we are working on now). The second-best is a stack of unmodified supplies with a buck front-end. The third-best is a stack of hacked supplies with appropriate control circuitry etc.

That said, we will be powering up the series stack of 1,500W supplies next week when we get the 15 connectors (http://parts.arrow.com/item/detail/fci/51666-001lf" onclick="window.open(this.href);return false;) that we just ordered for those. I think Cor might beat us to that - he's coming over to our office tomorrow to pick up some additional supplies for his stack and already has the connectors etc I think. Once we power this up, having an isolated combo of these + our buck stage is trivial. The problem is 75-80% combined efficiency - which really starts to matter above 10kW. As in 'you need to make special modifications to your Leaf just for cooling because no existing enclosed spaces (hood, trunk, even entire car) can take 3kW+ of heat output for hours without overheating.

On scaling up the 3kW version - my take is that it will be easier to develop a high-power isolation stage from scratch than to reverse-engineer and redesign for higher voltage / higher current. A lot of things don't simply scale - a few specific issues I foresee:
1. Optimal leakage inductance of the transformer scales quadratically with current and controlling it is not super-simple
2. Higher power will mean larger devices which will mean lower frequency / different driver circuits
3. Lower frequency has another set of effects on various parameters
4. A 10-15-year old supply is not likely to use the latest architectural options and will almost certainly have efficiency penalties relative to new design
etc.
Not saying it couldn't be done - just likely to be less efficient use of time.

Our DIYecar thread with tech discussion (just started) is at: http://www.diyelectriccar.com/forums/showthread.php/open-source-isolated-12kw-charger-89311.html" onclick="window.open(this.href);return false;

Thanks,
Valery.
 
Back
Top