Using Diodes in a Lithium Ion Battery Bank

Design Tools and Resources
, ,
1

We have not worked with Diodes before and wanted opinions on a design idea that we have.

We are installing (8) lithium ion batteries (12V 100AH each) in parallel in an RV. In order to reduce the risk of circumfluence, we want to use diodes to be sure that the power from the battery bank goes in only 1 direction (charging or discharging). We are thinking of using VS-QA250FA20. Would this make sense? Is there a better configuration or component choice?

If anyone has any feedback, we would very appreciate your comments.

Thank you,
Alalwani

2

Greetings,

When connecting batteries in parallel like this, it is important to ensure that they have an equal state of charge at the time they are being connected. One should also use cells of the same size, brand, model, age, etc.

When connected in parallel, batteries will tend to charge and discharge evenly, unless there are significant differences in the resistance of the wiring running to each cell, or major differences in the characteristics of the batteries used. The parallel connection naturally solves some of the problems that come from series connection, and in many ways is much simpler. There is little or no need to worry about charge flowing from one battery into another once they are connected, as long as the wiring resistance to each is more or less identical, stays intact, and similar cells are used.

It would probably be bad for example, if a person were to connect 7 fully-charged cells in parallel, and then try to add an 8th that was completely discharged. In this case, there would be a large transfer of charge, which might exceed the charge rate limits of the discharged cell and cause damage. Again, this is most likely to happen at the time of assembly. It could also occur later if, for example, a poor connection would cause one of the batteries to be disconnected for a time while the rest of the bank was charged or discharged.

Because of the size of the batteries being used here, I would suggest that it might be a good idea to provide over-current protection per-battery in order to protect against this sort of situation. It is possible that this sort of protection has already been built into the cells themselves, but this is not guaranteed. If you would provide a link to documentation for the specific lithium batteries you plan to use, I would be better able to offer suggestions for suitable fuses.

Please be aware that lithium batteries of this scale should be treated with great care. A battery bank of the kind described could make a length of 12AWG wire glow like a light bulb filament, and you wouldn’t want to be near it if it ever catches fire…

3

That be right for that application. It has 2 diodes in 1 package.

4

Hello Rick,

Thank you very much for your response. Our design would use 8 of our batteries - so they would be of the same manufacture, age, and capacity, etc. They would all be installed at the same time as well.

Attached is our PCM specs. Our PCM already includes over current protection. I think this would satisfy what you would need to recommend fuses.

If not, just let me know & I can send over other documentation.

Thank you,
Avanti

PCM_2020.docx (536 KB)

5

As I am not an Engineer, I am not able to help you with sizing the fuses.

6

@alalwani It appears to me that the document posted describes a protection circuit module designed to provide management functions for 4 series cells. Based on what you’ve mentioned, I understand this to be a module that you plan to use to assemble of each of your 8 batteries from a number of smaller cells, such as the common 18650-type. Is this understanding correct?

If so, I would suggest that this module may not suitable for the sort of application that you describe, and may represent an unacceptable level of risk if used in such an application.

There are several reasons for this, the first being that the under-voltage protection level is given as 2.0v. A 2.5 to 3.0v figure is more commonly suggested for most lithium cells, with a 2.0v cut-off offering a significant risk of cell damage as a result of over-discharge. When used for long-duration discharge applications as in an RV, the chances of over-discharge are greater than when used in an automotive starting application, where a brief discharge cycle is typically followed immediately by recharging.

Another reason is the relatively high discharge rates permitted by the module; 150A continuous, 500A peak, 600A fast-protection limit. 8 such arrays in parallel would produce a maximum 1200A continuous discharge rate. Availability of high currents at relatively low voltages increases the risk of arc faults, a condition in which an electric arc occurs (due to wiring defects or some other cause) and is maintained while conductors melt, set things on fire, etc. because the current flow is too low to cause over-current protection circuitry to activate. As I understand it, the system proposed would be able to deliver approximately 14kW on a continuous basis, or 48kW for several seconds given the current protection limits indicated. This seems rather excessive for an RV application. A good design would be to set suitable per-battery electronic current limits in the PCMs based on the maximum application requirements and number of parallel batteries to be used, and incorporate single-acting fuses with limits above the electronic limit in order to provide protection in the event of PCM failure. With the electronic limits in the PCM set as high as they are, this does not seem practical given the cost of suitable fuses.

The temperature protection limits also seem rather high; two sets of such limits with an upper tolerance of 90°C are described, though the information provided does not clearly indicate the conditions under which these limits apply.

Finally, there is an apparent typographical error in the specification table; a plus/minus 25V tolerance on a value of 3.6 as indicated would not be acceptable from an engineering standpoint… Some may choose to assume this is an error, but as written it -is- the specification for the device, and would allow the supplier to deliver product that provides no meaningful cell balancing whatsoever while remaining within published specifications.

image
image905×37 10.5 KB

Overall, I could see the PCM mentioned as possibly being suitable for assembling batteries for automotive starting-type applications, but I would not recommend its use for an RV storage application as mentioned.

7

Hello Rick,

Thank you for this very thorough response. I enjoyed reading it and it has stimulated questions that I do not think would have occurred to me otherwise.

The one thing to bring to your attention is that our cells are prismatic cells - we do not have 18650 cells.

When we assemble batteries in an RV, we include a 200A breaker. So, even though there are no fuses in the battery, we do have protections outside of the battery. Then the batteries are connected to a 3000W inverter.

Each of our batteries already have the PCM assembled with 4 prismatic cells wired in series.

Is there a battery that you know of that you would consider for an RV application or a battery that approaches the requirement? I am curious because for what we are doing - much of our power technology is derived off of automotive technology.

Thank you,
Avanti

8

I was guessing a bit at the nature of your operation based on the info provided; it seemed as if you were assembling individual cells and PCM devices to create your own battery packs, but if you’re purchasing these as pre-assembled units, the situation looks a bit different.

I’m more familiar with the underlying technologies and engineering concerns involved here than with currently-available products in this market space, so I don’t really have much to offer in terms of suggestions for which end-product battery assembly might be better than another.

I’d recommend taking a close look at the interrupt ratings on that 200A breaker you mention. DC is harder to break than an equivalent AC, so devices only qualified for use with the latter might hold unpleasant surprises in store. Adding supplementary per-battery protection using something like a 50A fuse on each would provide some backup to that breaker as well as some assistance in the event of a PCM failure.

As to the over-discharge question, the inverter’s low-voltage cutoff function (if present) may help mitigate the issue depending on where it’s set. If there are “12V” loads with access to the battery array independent of the inverter, they’d offer an opportunity for users to circumvent that protection inadvertently.