Charge in parallel, discharge in series

Some times, users of a high voltage battery pack (> 500 V) would rather not buy a high voltage charger, or chargers that can be put in series to achieve a high voltage.
So they may come up with this crazy idea of splitting the pack in two, connecting the two halves in parallel during charging, and in series during discharging. For example, a 700 V pack can be split into two 350 V half packs, charged with a 350 V charger, and then used with a 700 V motor driver.
That’s not too hard to do, except for the BMS, which needs to be able to handle both modes of operation. And that can be really challenging.

Here is a clever solution that uses a single contactor, 2 rectifier diodes, and a Lithiumate Pro BMS.

When the AC is available, the charger is on, and a contactor opens the series connection between the half packs.
Rectifier diodes are forward biased, effectively connecting the two half packs in parallel, and the charger charges both half-packs.
The BMS needs to be told that the half-packs are in parallel, so that it can report the correct pack voltage.

When the AC is gone, the contactor closes, and the packs are connected in series.
The rectifier diodes are reverse biased, disconnecting the charger from the half-packs.
The BMS needs to be told that the half-packs are in series, so that it can report the correct pack voltage.

The Lithiumate BMS is the only BMS that can handle strings in parallel, and specifically the only BMS that can be switched on the fly between 1 series string and 2 strings in parallel.

Posted in Uncategorized | Comments Off on Charge in parallel, discharge in series

Distributed vs non-distributed BMSs

I just saw a picture of a non-distributed BMS installed on an Enerdel battery module. We use the same modules at Elithion, with our distributed BMS, and I though it would be fun to juxtapose the two.

Distributed vs non-distributed BMS

Distributed Elithion BMS (left) and non-distributed BMS (right) mounted on an Enerdel battery module

Granted, this is an extreme case, but still, it makes the point that there are some definite advantages to installing BMS electronics directly on the cells, and minimizing the number of wires around a high voltage / high power battery.

Note how the Elithion BMS cell board tucks neatly into the recess in the battery module, with only 2, small, low voltage, communication cables coming out. On the contrary, the non-distributed BMS has more than a dozen high voltage wires going from the cells to the slave board; also, a spot must be found in which the board can be placed; it’s an accident waiting to happen.

Yet, a non-distributed BMS does have other advantages, as long as the layout is done in a way to minimize the chance of a plasma event. Indeed, the Elithion Lithiumotive BMS does include the option for non-distributed slaves.

Posted in Uncategorized | Tagged | Comments Off on Distributed vs non-distributed BMSs

Manufacturing Traction battery packs

Today we shipped our first battery pack: a traction pack for a PHEV truck, for a client in California. This marks the beginning of a new phase of Elithion, expanding from just Li-ion Battery Management Systems, to complete, large Li-ion battery packs:

  • Traction packs for EVs, HEVs and PHEVs
  • Battery modules for grid storage and telecom back-up

To do so, we created a battery department: we rented lab space, hired a mechanical engineer, signed up the services of a few, key individuals with thermal, mechanical and battery experience, and partnered with like-minded companies.

Traction pack, open

The first result of this new operation is a 20 kWh, 80 kW (180 kW peak), 350 V, liquid cooled traction pack, that includes a BMS (the Lithiumate Pro), a 3 kW charger and a 1.8 kW DC-Dc converter to generate the 12 V supply.

The pack was delivered a mere 73 days after receipt of order.

The hardest part was to identify and procure the cells. We chose pouch cells from the US company Enerdel, because of their high power handling capacity, of how well designed the battery modules were, and of their availability. They aren’t cheap, not at all, but the performance was well worth the price.

Final testing established that the pack is 98 % efficient at 250 A, which means that the liquid cooling system did not have to work too hard to extract the little heat generated.

We are proud of this first achievement, and look forward to more, successful battery projects.

Posted in Uncategorized | Comments Off on Manufacturing Traction battery packs

Orion BMS and playing fair

We are proud to offer a directory of all the Li-ion BMSs and a selector tool for all Li-Ion BMSs, (including the Orion BMS) in which we do our best to be completely fair to all, in great part thanks to the collaboration we received from other BMS manufacturers. All along, we have been blessed by a wonderful environment of cooperation among the various BMS manufacturers, in the view that we are all working towards a safer world of Li-Ion batteries.

Well, maybe I was being naive.

I feel disheartened by the advertizing results for of one particular BMS. See for yourself: Google “Elithion”, or Google “Lithiumate”, and you’ll see that the 1st result is an ad for the Orion BMS. Even though the keywords “Elithion” and “Lithiumate” are terms that Elithion has trademarked, somehow they are ending up being used as keywords that trigger an ad for Orion BMS. We tried searching for other BMSs (by brand name or by model name) and the ad does not appear. We tried searching for “Lithiumotive”, which is also an Elithion trademark, and the ad does not appear. Of all the BMS names. only the keywords “Elithion” and “Lithiumate” result in the ad.

I talked to Chris Ewert, maker of the Orion BMS, and he assures me that, no, they did not use the keywords “Elithion” and “Lithiumate” in Google Adwords, and that he is not targeting Elithion in any way. Not only that, but he also offered to add Elithion and Lithiumate in the Google AdWords campaign as negative keywords, meaning that an ad for the Orion BMS will no longer appear when searching for those terms. Thank you Chris for doing that!

Regardless of ads, if you wish to use a non-distributed BMS, by all means you should consider the Orion BMS. I understand that the Orion BMS, is a great product, and I have had the pleasure to work with the Ewert brothers before, so I do encourage those in need of a non-distributed BMS to take a look at Ewert Energy’s Orion BMS.

If, on the other side, you need a distributed BMS, then the Orion BMS will not work for you. In that case, we hope you will consider the Elithion Lithiumate BMS.

Posted in Uncategorized | Comments Off on Orion BMS and playing fair

Review of: “Thermal management of lithium-ion batteries for electric vehicles”

I truly regret spending $ 35 to read the paper “Thermal management of lithium-ion batteries for electric vehicles” by G. Karimi and X. Li.

I’ll save you the $ 35: in just one sentence, this is what the paper says:

“To cool a battery, use air (the extra effort of liquid cooling is not worth the trouble) and make sure the air flow is evenly distributed in between each cell, otherwise you’ll only cool the ends of the battery, and get a temperature gradient, which is bad.”

Well, obviously!

To reach that obvious conclusion, the researchers did do a good job at modeling the effects of cooling. But they never bothered to actually try it in practice. Had they done so, they would have realized that two of the fundamental premises of their model were wrong:

  1. Cycling a battery with a thermal gradient will unbalance it (it doesn’t)
  2. The open circuit voltage of a Li-Ion cell is temperature dependent (it isn’t)


1) Temperature differences do result in differing self discharge currents; but the effect is secondary and long term, and is completely unaffected by whether the battery is cycled or is in storage. So, while a temperature differential does unbalance the battery over the long term, in the short term, cycling such a battery will not result in any more unbalance than if it were standing by.

2) Temperature differences do result in differing cell resistances; if under load, they result in differing terminal voltages. But the Open Circuit Voltage (OCV) is not affected: some time after the load is removed, the cell voltages will return all to the same voltages, even if they are at different temperatures; that’s because their SOC levels are all still all the same (the battery is still balanced)

On top of that:

What a pain it is to pay to access a paper for just 1 day, a paper that you don’t get to keep, using a font too small to read on a computer screen (they defeated the zooming function in my pdf reader).

Posted in Uncategorized | Comments Off on Review of: “Thermal management of lithium-ion batteries for electric vehicles”

Review of “The Advanced Smart Grid: Edge Power Driving Sustainability”

Book coverArtech House just published the book “The Advanced Smart Grid: Edge Power Driving Sustainability“, by Andres Carvallo and John Cooper, two of the architects of the first implementation of Smart Grid, in Austin TX.

The book reads like an essay: it can be read cover to cover (I did so in a weekend); it is not really a reference book, as it would be hard to “look something up” in it.

The first few chapters are instructive, though the writing is hardly scintillating. Then, in chapter 4, one of the authors switches to the 1st person and recounts his personal experience in leading Austin Energy from a ho-hum power company, to the first company to create the Smart Grid concept and implement it. It may not have the impact of an action novel, yet I found myself excitedly reading the remarkable achievements of a little, city owned utility, as if I were reading about NASA’s conquest of the moon.

The authors use an interesting approach to looking into the future: in a few sections, they switch their point of view to the 2020’s, looking back to the history of how the Smart Grid evolved to “today”, what challenges were encountered and how they were solved. For example, they describe how, “back in 2016”, thousands of EVs converged to the SXSW (South By South West) fair, and how Austin Energy coped with the sudden rise in EV loads as well as the availability of energy from them. (I am looking forward to that prediction coming true, and actually driving one such EV to the 2016 SXSW.)

The book is touted as a “how to”, but if you’re looking for actual technical guidelines you may be disappointed. It would appear that the authors heard the engineers and IT staff talk about the technology and the software involved, and got a vague idea about them, but all they can do is recite its jargon without truly grasping their functionality. That lack of clarity comes across, for example, when the book discusses EVs: from reading the bulk of the book you would think that a simple EV is a energy source, not a load, which it normally is. Towards the end, the book does mention V2G (Vehicle To Grid) technology: the (rare) V2G equipped EV is indeed capable of being both a power source and a load; but that is not explained earlier, when the book categorizes EVs as a resource.

The book’s true audience is managers (power company managers, CEOs of equipment manufacturers, politicians, city councilors and managers). The staff that would actually implement a smart grid would find it useful to get an idea of the goals, but may get a misleading view of the technology used. From a technical perspective, the book has quite a few interesting errors, such as using Hertz to measure voltage, and the charmingly amusing assertion that the extra energy from a solar panel on the roof that is not used by the home is send to ground (!).

While the book has some nice looking tables and graphs, they are not totally legible due to the combination of small font size and use of gray scales resulting poor contrast. Some tables show data without explaining what the numbers represent. Some graphs give the impression that information is being conveyed by the arrangement of the terms, when in reality they are simply lists of terms in no particular orders, just arranged to they fit nicely in the figure. A few graphs imply a flow (process flow, data flow) even though either there is no such flow in reality, or, if there is, it wasn’t apparent to me after reading the text. The book would have benefited from the use of before-and-after pictures, and graphs showing the effects on Austin Energy and its consumers, as a result of implementing Smart Grid.

While the focus is primarily on Austin, the book does a good job at reviewing other Smart Grid projects in the US. Readers outside the US will have to make do with reading about the US experience.

All and all, I found this book to be well written and informative. I believe that it would be essential to managers of cities, power companies and product manufacturers, to understand the benefits of Smart Grid, be prepared for the travails of implementing it, and learn the pitfalls to avoid. Reading it would be also instructive to the staff charged with actually implementing Smart Grid, for general overview, not as a technical guideline.

Posted in Uncategorized | Comments Off on Review of “The Advanced Smart Grid: Edge Power Driving Sustainability”

Selecting the right BMS for your application

By now, dozens of Li-Ion BMSs are commercially available, making a choice somewhat daunting.

Two years ago I prepared a list of all Li-Ion BMS.

Today, with 54 options, it has become impractical to explore every item in that list.

Therefore I now prepared a parametric search utility to select the right BMS for a given application.

It allows you to chose the right, commercially available BMS, for your large Li-Ion battery pack application based on your needs.

I have attempted to be as thorough and accurate as allowed by the information available on each BMS. I contacted every company in that list, and a few have responded with corrections, which I promptly made to the data base.

I hope you find it useful.

Posted in Uncategorized | Tagged | Comments Off on Selecting the right BMS for your application

EV fires and BMSs

Unfortunately, every so ofter, and EV catches fire, and, if unattended, it may burn to the ground.

If the EV uses Li-Ion cells and a BMS, a chorus of “BMS deniers” raises its voice, expressing the view that the fact that the EV had a BMS proves that the BMS was the cause of the fire. I will not bother pointing out the logical fallacy in that point of view, but I would like to address the valid concern that others expressed, that the BMS did not prevent these fires.

What seems to be occurring in the cases, is that a BMS, while physically located in the EV, was not actually installed in such way to allow it to protect the battery pack.

Let’s take 2 recent examples: the Neil Young’s LincVolt and SRJC’s (Santa Rosa Junior College) Ford Focus EV conversion. Both had an Elithion Lithiumate BMS on board.

(Neither BMS was bought from Elithion, nor from Elithion authorized resellers; I say that because when Elithion or one of its authorized resellers sell a BMS, the customer is helped to make sure the BMS is installed correctly.)

Neil Young’s LincVolt

On Nov 9th 2010, the LincVolt (a Lincoln Continental converted to a hybrid) was plugged in and charging, when it caught fire and burned down. While an investigation was promised, it either never happened, or its results were not published.

EV on fire

LincVolt HEV burining up

Based on conversation with people close to the project, two factors caused the fire:
1) While the BMS was physically in the vehicle, it was not connected to the charger, so it had no way of protecting the battery pack from over-charge
2) The charger had never been adjusted for the battery voltage, and therefore it kept on charging the pack past its nominal full voltage.
Granted, the above findings are based on unofficial analysis and reports, and are therefore unproven; but they are the best we have to go on.
What we do know is that the LincVolt staff has very specifically said “LincVolt suffered a disastrous accidental fire stemming from human error” (1), which is a way of saying that the equipment was not a fault.

SRJC Ford Focus

On March 18th 2011, the Santa Rosa Junior College EV conversion caught fire.

Burned up EV

SRJC burned up EV

An investigation may be forthcoming.

In the meanwhile, what we know is that:

  • The EV used (2):
    • Li-Ion cells
    • A Lithiumate BMS
    • A Brusa charger
  • The 12 V battery was not installed at the time of the fire (3)(4)
  • However, a DC-DC converter was installed, and could have conceivably been powering the 12 V bus, if the ignition had been on (5)
  • The charger was in the vehicle, but was not yet functional and was not plugged into the AC at the time(3)
  • The initial location of the fire is unknown (3)
  • Other vehicles nearby were not burned (3)
  • The BMS was not connected to the charger, and could not have turned it off, even if the BMS had been powered at the time (“there was no interface with the charger to turn it off”) (4)
  • The car was not plugged in. (3)(4)

Given that the car was not plugged in, and, in any case, the charger was not operational, the charger would certainly not be the cause. Then the cells may have been the source of energy that started the fire. But not necessarily; it could have been, for example, a soldering iron that had been left on since previous work done on the vehicle.

Preventing EV fires

We know that in the LincVolt, the BMS was not at fault for the fires. In the SRJC Ford, it is very unlikely that the BMS was in any way involved in the fire, but that is yet to be proven. But that does not prove that a BMS is never a contributing factor to EV fires. The only certain thing is that not connecting a BMS properly will not protect the battery pack.

At Elithion we are very insistent that the BMS must be properly connected to the system, or the pack is not protected. The first thing you see in the Lithiumate manual, is a huge warning:

You MUST provide a way for the BMS to shut down the charger, and the motor driver, DIRECTLY!

Still, all too often BMS users assume wrongly that, just by having a BMS physically present, their pack is protected.

Not so.


(1) LincVolt Gazette, press release of Nov 16th,
(2) Class project notes
(3) Chris Jones, a volunteer working on the project, from his message to the NBEAA email list of Nov 20 2011. Chris welcomes you to contact him on this matter: see the NBEAA site for contact info
(4) Peter Oliver, from his message to the EVDL mailing list of Mar 21, 2011
(5)Peter Oliver, private phone call

Posted in Uncategorized | Tagged | Comments Off on EV fires and BMSs

New website for

After 2-1/2 years, it was time for Elithion’s on-line presence to be brought up to a high level, comparable to the level of its product line.

The first part was to create a professional looking website focused exclusively to Elithion’s business. However, we did not want to lose the gobs of information in the original site. Therefore, we split the information in the original site into the two sites: if a page related directly to the business, it went to the new site; else, it stayed with the old site.

The hardest part was to make the new site look professional. Not being a professional web designer myself, I wanted to hire the services of a professional group which came highly recommended. I had a very fruitful introductory discussion with an owner of that company, during which she gave me many great pointers (which I immediately implemented). Yet, I was never able to engage that company: I was turned down, ostensibly because the site is hand coded (that is, it does not use a Content Management System), but more likely because the web designer wanted to redo the site from scratch, while all I was asking for was a few hour of guidance.

So, I ended up doing the new site by myself.

I hope you like it, and find it useful.

Posted in Uncategorized | Tagged | Comments Off on New website for

Lithiumate DIY controller

Granted, in trying to meet all different kinds of needs, the Lithiumate BMS is a rather complicated device, possibly too complicated for the Do-It-Yourself user.

For that reason, I am now working on a DIY version of the Lithiumate controller, designed for lower cost (about 50 % of the standard Lithiumate controller) and with fewer set-up items. It is seamlessly compatible with the other components in the Lithiumate line.

Compared to the standard Lithiumate controller:

  • Uses a plastic case (i/o metal)
  • Supports 8 banks (i/o 16)
  • Has USB (i/o RS232)
  • Has only 6 connectors (i/o 20)
  • Adds a “Warning” and a “Current” output.
  • Does not include contactors; drive
  • The user interface has fewer screens and settings (~ 50 %)

The Lithiumate DIY controller will be offered exclusively by EVolve Electrics.

Posted in Uncategorized | Comments Off on Lithiumate DIY controller