EPRI's Fast, Flexible - And Cheaper - EV Charging System

[Adm]

I'll let the experts explain this one... I stopped understanding it after I read 2.4 kV.

The Electric Power Research Institute (EPRI) describes its new technology as a 2.4 kilovolt, 45kVA (45,000 volt amps for the less technical among us) solid-state, direct current (DC) fast charging system for electric vehicles (EVs).

http://energy.aol.com/2012/02/17/the-next-big-thing-epris-fast-flexible-and-cheaper-ev-cha/

 

[doug]

The article is light on details but the power doesn't sound that impressive.
45kVA=45000 Volt Amps = 45kW
That's about as good as current CHAdeMO or Mennekes 3 phase, and half what Tesla is claiming with their Superchargers. I'm not sure what battery pack can take 2.4kV. I'd need more information from a more technical source.

 

[widodh]

Most battery packs operate somewhere around 400V, so why whould you charge with 2.4kV? The only reason I could think of is that you could get 200kW over the same cable as we currently get 50kW. Cable diameters have to do with the Amps, not the kW's.

But 2.4kV is killing for a human. I wouldn't want to have a cable lying around with 2.4kV on it!

 

[GSP]

Some more technical info here:

http://www.electricenergyonline.com/?page=show_news&id=163364

Here is a free abstract for EPRI's $2500 report:
EPRI Electronic Transformer Report

The second activity was focused on building and developing a 50-kVA 2.4-kV medium-voltage fast charger for electric vehicles with support from the Tennessee Valley Authority (TVA). Initial testing and demonstration of a 10-kVA IUT-based fast charger was performed in May 2011. As part of the demonstration, the 10-kVA system was successful in charging a Nissan Leaf. The 10-kVA system was then expanded to a full 45-kVA prototype system in 2011. The Electric Power Research Institute (EPRI) is working with Enertronics and Howard Industries to finalize the packaging of the medium-voltage fast charger (FC) system. The fully packaged 45-kVA 2.4-kV IUT-FC system will be demonstrated at EPRI's Knoxville laboratory in January 2012. Additional testing at TVA's host site will be performed in 2012. A data acquisition system (DAQ) and a web-based application is being developed that can be used to store and analyze the data recorded from multiple IUT systems. EPRI and Enertronics finalized the specifications of the data feeds, the DAQ system, and the features of the web-based application platform (web browser as well as IOS system).


EPRI's charger uses 2400 VAC, 45 kVA input (probably 3-phase), and provides DC voltage to the battery according to the car's ChadeMo protocol request (command?). It is less expensive and more efficient than conventional 480VAC 3-phase ChadeMo chargers. This is achieved by eliminating at least one AC transformer, and maybe other components.

Slide 4 in Nissan's presentation on their new ChadeMo charger is helpful to see what components might have been elimined by EPRI. EPRI also mentioned eliminating the 2400 to 480 VAC utility transformer that typically would supply a small business.

http://www.teslamotorsclub.com/attachment.php?attachmentid=4259&d=1329123734


widodh - 2400 VAC requires due care, but please be just as careful with 220VAC in the home and 300 vdc in the car!

GSP

 

[eledille]

Considering that all modern EVs contain a three phase rectifier and voltage regulator in its motor drive already, more money can be saved by simply feeding 400/480 V three phase power directly from the grid into the motor controller. I believe this is what Renault is doing with the Zoe which costs GBP 17000 before incentives (w/o battery, lease GBP 70/month) and includes 44 kW AC capability. I think these guys are reinventing the wheel, but their version is not quite circular.

 

[ElSupreme]

Most battery packs operate somewhere around 400V, so why whould you charge with 2.4kV? The only reason I could think of is that you could get 200kW over the same cable as we currently get 50kW. Cable diameters have to do with the Amps, not the kW's.

But 2.4kV is killing for a human. I wouldn't want to have a cable lying around with 2.4kV on it!

But at 2.4kV and such low amps you could us thread thin wires!! What could possible go wrong.

 

[GSP]

But at 2.4kV and such low amps you could us thread thin wires!! What could possible go wrong.

The 2400 volt wiring is from the utility to the ChadeMo charger. The cable to the car is no different than other Chademo chargers.

GSP

 

[TEG]

Some more technical info here:
http://www.electricenergyonline.com/?page=show_news&id=163364

From there:

...The demonstration in Knoxville will be performed using a Nissan Leaf and a Mitsubishi iMiEV. EPRI will demonstrate the communication capability of this fast charging technology with electric vehicle battery management systems. A user interface and web-based mobile data collection system are also included in the Knoxville demonstration.
The EPRI Utility Direct Fast Charger technology was developed by a team of engineers and scientists from EPRI and Enertronics, with financial and technical support from Tennessee Valley Authority (TVA)...

Saying that the demo will use LEAF and iMiEV does suggest strongly that the final output will be CHAdeMO.
I find this interesting as EPRI is a US entitity and the SAE DC QC standard seems to be in competition with CHAdeMO.
Developments in Tennessee do seem to be shaping the future EV landscape in a big way including the largest US based EV mass production (LEAF), and adoption of charging standards that started in Japan.

EPRI has a Palo Alto lab close to Tesla. Will we get a version of this charger technology that has a Tesla plug output?
Not only being a 'breakthrough' to make utilities consider installing CHAdeMO this would have the potential to be a cheaper alternative to Tesla's upcoming Supercharger.

I wonder if Ford & Tesla feel bad that this is being demonstrated first for use with Japanese engineered cars, not one of their EV products?

 

[GSP]

Considering that all modern EVs contain a three phase rectifier and voltage regulator in its motor drive already, more money can be saved by simply feeding 400/480 V three phase power directly from the grid into the motor controller. I believe this is what Renault is doing with the Zoe which costs GBP 17000 before incentives (w/o battery, lease GBP 70/month) and includes 44 kW AC capability. I think these guys are reinventing the wheel, but their version is not quite circular.

It will be interesting to see if DC charging or 3-phase AC to the car will become the predominate quick charge method in Europe. It seems like there will be both for a while, with plenty of ChadeMo installations. EPRI's ChadeMo charger is less expensive and more efficient. It eliminates an expensive transformer that is required to provide the 400/480 V three phase power that other ChadeMo chargers require. It does not sound like a "square wheel" to me.

GSP

PS. In the US, we don't have any AC quick charging in the works. Just three flavors of DC quick charging. EPRI's more efficient DC charger makes even more sense here.

 

[GSP]

Saying that the demo will use LEAF and iMiEV does suggest strongly that the final output will be CHAdeMO.

Will we get a version of this charger technology that has a Tesla plug output?

The EPRI announcement sounds like "proof of concept" level work. I suspect they chose CHAdeMO because cars were available for tests and demonstrations. The concept is most likely technically compatible with the SAE DC and Tesla quick charging methods. Like Beta vs. VHS, the marketplace will sort out what will be offered. CHAdeMO has a head start with more vehicles and chargers, and is not a proprietary interface. Tesla will likely have a good number of vehicles in the field also. SAE DC will maybe have a few vehicles, but is not limited to one automaker, or one charger OEM.

GSP

 

[GSP]

Previous research with a smaller 10kW version, from the link I posted earlier:

"Based on successful collaborative research, EPRI developed a 2.4KV 10kW working prototype of a DC fast charger for electric vehicle application, using a technology known as Intelligent Universal Transformer (IUT) as the key component. The proof of concept of the fully functional 2.4KV 10KW medium voltage IUT based DC fast charging system was successfully demonstrated in March 2011 at the EPRI laboratory in Knoxville.

This demonstration will confirm the capability of this unit to provide a full vehicle charge. Performance of this new technology will be compared with a commercial 200V 3-Phase DC fast charger. This will be the first of a number of efforts that are planned to take this technology from laboratory prototypes to actual field demonstrations.

The IUT technology replaces both the independent power conversion units as well as the conventional transformer with a single interface system which can be used for fast charging of electric vehicles. The versatility of the IUT provides an intermediate DC bus voltage at the 400-V level that can be directly used for a DC distribution system or for EV fast charging. The medium voltage IUT-DC fast charger has the potential to achieve efficiency higher than 96% in the 10-to-90% operating range, a saving of >6% over the conventional approach. For a dedicated DC fast charger where the DC-AC inverter is not needed, the isolated DC-DC converter output can be directly used for fast charging, avoiding an additional 1.5% energy loss. The elimination of the conventional transformer also allows significant size and weight reduction on cabling and installation. The standard ANSI based 50-kW transformer weighs more than 800 lbs. The entire charging station including the low-voltage based charger will weigh more than 1000 lbs. On the other hand, the weight of the entire EPRI IUT-DC fast charger electronics is less than 150 lbs. "


The cost and efficiency advantages sound compelling to me. Not only does the charger cost (and weigh) less, and use less energy due to higher efficiency, but they also talk of lower demand charges due to the "IUT" as well.

GSP

 

[GSP]

More info here:

http://www.smartgridnews.com/artman/uploads/1/IUTFastChargeDemoSummary8-4-2011_1_.pdf

"Commercially available DC fast chargers are all low-voltage 3-phase input units that can be supplied off
208/480 V AC. These DC fast chargers require conventional three-phase transformers that convert
medium voltages (~15 kV L-L) to the required lower AC voltage. All together, a conventional DC fast
charger has the following power conversion stages:

1) AC-AC stage (3-phase distribution transformer 15 kV → 480 V AC)
2) AC-DC power electronic stage (the first stage within the DC fast charger that converts 480 V AC into
an intermediate DC voltage)
3)DC-DC power electronic stage (the second and last stage of the DC fast charger that converts the
intermediate DC voltage to the voltage required to charge the electric vehicle (EV) battery) "


That last DC-DC stage must involve DC to AC to DC power conversions.

GSP

 

[rabar10]

That last DC-DC stage must involve DC to AC to DC power conversions.

Typically, yes. The PDF linked in this post has details on Nissan's older and newer fast chargers, for comparison.

 

[doug]

So if I understand, this is using a direct 2.4kV input to convert to the DC used for EV fast charging, instead of a 480V 3-phase input.

So where is 2.4kV available?

from wikipedia:

 

[GSP]

2.4kV seems to be out there. A quick google search found this:

http://www.umass.edu/physicalplant/documents/UMA_UEM_Stds.pdf


"The Amherst Campus has two electrical distribution systems a 13.8
kV grounded wye Primary Selective system and a 2.4 kV ungrounded delta
open loop system."


I found this note amusing:

"NOTE: NO single phase loads will be acceptable
on the Medium Voltage Distribution Systems."

GSP

 

[eledille]

It will be interesting to see if DC charging or 3-phase AC to the car will become the predominate quick charge method in Europe. It seems like there will both for a while, with plenty of ChadeMo installations. EPRI's ChadeMo charger is less expensive and more efficient. It eliminates an expensive transformer that is required to provide the 400/480 V three phase power that other ChadeMo chargers require. It does not sound like a "square wheel" to me.

Thinking it through again, I realize that you're right, but I still don't think it's the best approach for charging at up to 50 kW.

PS. In the US, we don't have any AC quick charging in the works. Just three flavors of DC quick charging. EPRI's more efficient DC charger makes even more sense here.

Yes, but I think three phase AC charging is unavoidable in NA too. Any auto maker that wants to sell in Europe needs efficient on-board AC fast charging anyway, and I don't think anything can compete against reusing existing equipment for charging in addition to its original purpose. The parts cost is so low. That's only valid where 400-480V is available, of course, i.e. everywhere a current CHAdeMO charger can be installed.

On the other hand, charging at above 50 kW will require DC, so more efficient DC chargers is a good thing. What I'm trying to say is that they should aim for 75 kW and above instead of 20 to 50.

 

[GSP]

On the other hand, charging at above 50 kW will require DC, so more efficient DC chargers is a good thing. What I'm trying to say is that they should aim for 75 kW and above instead of 20 to 50.

I agree completely, especially in Europe. EPRI probably made a 45 kW demo since that is what the Leaf and the i could accept. It is just a proof of concept demo to show charger OEMs and Utilities what is possible. I doubt EPRI will make chargers, since they seem to be focused on R&D "think tank" research to help the Utilities.

In the US and Japan, we may be stuck with more expensive DC charging when more than 20kW is needed. This is due to the SAE making the unfortunate decision to not support 3-phase with the J1772 "Yazaki" connector. Once enough of these get out, it may be too disruptive to switch. Just like we are stuck with QWERTY keyboards instead of the faster Dvorak layout. On the other hand, if the cost is compelling enough, switching is possible. Three phase is widely available here. As you point out, 3-phase is available wherever a CHAdeMO charger is installed.

GSP