Victron vs HA-02 Battery Balancers

[Don]

When I bought my batteries, I searched battery balancers that were available at the time and the best I could find was the Victron. I bought 3 of them and installed them. They have been running since day one. To date I am very happy with them, but on the other hand I have new batteries as well. I have been following various threads about battery balancing and mid point problems, etc. The more I look at he HA-02, the more they impress me. They can balance in series and parallel at the same time. To top it all, they are about half the price of the Victron. I am beginning to doubt if I got the right battery balancers? 

[___]

I  believe the Vickies are pass-the-parcel types. They'll probably last forever, but I tend to agree that the HA02 is better value. I won't know for sure until I've had my hands on both.

In this case I would say that the best balancer is the one you already paid for :-)

[superdiy]

11 hours ago, plonkster said:

I won't know for sure until I've had my hands on both.

Those HA-02 balancers are potted and I've done a fair bit of research in the past about ways to dissolve potting resin and most of the stuff is either difficult to get hold of and/or really nasty chemicals. I suspect that they are using a uP programmed to work in a similar fashion to the LTC3305. 

[___]

I think the blue boxes are also potted internally. The mppts certainly are, to get a proper IP rating.

Sent from my GT-I9195 using Tapatalk

[___]

I wonder about the high-pitched whine some reported. That could indicate a DC-DC converter. The number of wires hanging out of the HA02 also seem to indicate it's a different beast on the inside.

This kind of topology (siphoning off off a high cell, turning it into magnetic energy at some high frequency, then and back into electrical energy across an isolation barrier to put it in another cell) does of course pay a efficiency penalty. I believe most off the shelf converters are around 90% efficient. I would expect a single pass-the-parcel balancer to be more efficient, 100% of the passed current will end up in the target cell. So perhaps we should not be too quick to dismiss it as inferior.

I do however wonder about what happens when you have several balancers bypassing parts of the string, each with their own voltage measurement (that will differ slightly). The one thing you don't want is for all the bypass switches to be on simultaneously. Yes, it is current limited (so nothing really bad will happen), but imagine it does happen: Well then you could pass a couple amps right across the bank, which will of course affect efficiency.

As long as one bypass device is always off, the squeaky wheel gets the grease. I would expect each device to use a comparator circuit with a tiny bit of hysteresis (batteries are allowed to be a very tiny bit out of kilter, just enough to avoid ringing on the comparator), so it is probably fair to assume that such a condition (all bypasses on) is extremely unlikely.

Which brings me back to my previous point: Perhaps we should not be too quick to dismiss it as inferior. Less flexible maybe, but it also has less wires to deal with! :-)

[superdiy]

49 minutes ago, plonkster said:

The number of wires hanging out of the HA02 also seem to indicate it's a different beast on the inside.

Which brings me back to my previous point: Perhaps we should not be too quick to dismiss it as inferior. Less flexible maybe, but it also has less wires to deal with! :-)

The amount of wires on the HA-02 is because of the fact that it technically has 4 "isolated" circuits inside and therefor can be connected to serially or parallel connected batteries - one set of cables per battery. I don't really understand why you would want to balance parallel connected batteries, since they should technically be balanced in any case since they are connected in parallel, but that's a discussion for another day.

With the Victron balancers, you have a one shared connection for two serially connected batteries - by having the shared connection you don't have the option to balance parallel connected batteries or say every second battery in a large serial string. Just also keep in mind that you need 3 Victron balancers or 1 HA-02 to balance 4 serially connected batteries with technically 8 wires for the HA-02 and 9 wires for 3 of the Victron balancers. 

[___]

1 minute ago, superdiy said:

by having the shared connection you don't have the option to balance parallel connected batteries or say every second battery in a large serial string.

True, but as you just admitted earlier, why would you want to do that? I can see a case to be made in some special setups, but most people will actually connect the balancer between adjacent cells so that there is no benefit to the isolated nature of the beast.

2 minutes ago, superdiy said:

Just also keep in mind that you need 3 Victron balancers

True :-)

What we need is a simpler/cheaper balancer for 48V banks.

Then again, people are going to blow them up because they don't understand how it works and cannot follow instructions. At least with the HA01/2, the isolation means there is no danger of that. There is only the danger of de-balancing the bank! :-)

[___]

So I quickly sketched out a good old circuit with two comparators, a resistor divider bridge and quickly decided that not only is this too much effort, I'm probably going to end up with more components too. Much simpler to take a small MCU, like an attiny25 (which costs like R20) and write a tiny bit of code that simply checks that the reading on one pin is more or less half of what is on the other, or more specifically, that |x/2-y| < ε.

The attiny should be sufficient. Only need 4 pins, and this chip has 6 (5 if you prefer to keep reset intact). 2k code space should be enough for this simple task.

Then probably switch a mosfet or two, with some inherent current limiting. This will be the difficult bit that requires the hard work to avoid burning things down :-)

[Don]

2 hours ago, superdiy said:

Just also keep in mind that you need 3 Victron balancers or 1 HA-02 to balance 4 serially connected batteries with technically 8 wires for the HA-02 and 9 wires for 3 of the Victron balancers. 

The price of the HA-02 just came down a lot more. The price of the HA-02 is about half the price of a Victron balancer. In the example above, you only need 1 x HA-02 vs 3 x Victrons to do the same job. Therefore the HA-02 is actually 1/6th the price of a Victron. That makes the Victron very expensive. 6 x more expensive compared to the HA-02.

[___]

1 hour ago, Don said:

That makes the Victron very expensive.

Not arguing at all. Thinking that if you can make it that cheap with an isolated architecture, then the simpler more efficient pass-the-parcel should be even cheaper when made in China, or perhaps even at home :-)

[McWidowmaker]

So, if you have 3 x 200ah 48V battery bank you need 9 balancers? And if you have 24 x 2v cells that you wish to balance you are going to need how many? Wondering if this is really worth while. By how much would it really extend your battery life by and what would the cost vs benefit be? Anyone made some sums regarding this?

[Chris-R]

If you are talking about 3 x banks consisting of 4 x 12v batteries each, you will only require 3 x HA-02 balancers. 3 x 680 = 2040 + courier + vat = R 2428.20 delivered to your home. You could add more balancers with different configurations and fuses and a lot of wiring etc, but just the 3 basic HA-02 will do the job fine.

Cheers

[Chris Hobson]

18 hours ago, plonkster said:

Not arguing at all. Thinking that if you can make it that cheap with an isolated architecture, then the simpler more efficient pass-the-parcel should be even cheaper when made in China, or perhaps even at home :-)

My HA01s are pass the parcel and 3 were about R400 each, so more expensive than a HA02. Admittedly they came from Lithuania - but they were the cheapest I could fine at the time. Chris-R is by far the cheapest price I have seen for a HA02 and he needs to be thanked for the service to forum members. I hope that the increase in the number of HA02s out there will start generating the volumes to justify his time and trouble. 

Happy customers are good adverts and @incagarcilaso glowing reports are beyond my expectations based on the size of the bank, its age and the fact that for a short while the HA02 were making the situation worse until the wiring had been corrected.

[incagarcilaso]

On 4/3/2017 at 8:13 AM, Chris Hobson said:

My HA01s are pass the parcel and 3 were about R400 each, so more expensive than a HA02. Admittedly they came from Lithuania - but they were the cheapest I could fine at the time. Chris-R is by far the cheapest price I have seen for a HA02 and he needs to be thanked for the service to forum members. I hope that the increase in the number of HA02s out there will start generating the volumes to justify his time and trouble. 

Happy customers are good adverts and @incagarcilaso glowing reports are beyond my expectations based on the size of the bank, its age and the fact that for a short while the HA02 were making the situation worse until the wiring had been corrected.

Yes, and the last 3 weeks have shown them to be excellent. Here is an update for those who are looking at this technology: Within 3 days (yes, just 3 days on a bank of 24 1500Ah 2V cells!) the mid-point had been reduced from 4.6% (very poor condition) to be in the 0.2-0.4% in the last stages of bulk charge, although from measurements some of the cells were still clearly with a much higher and some with a much lower voltage than others (remember that because of the nominal voltage for operation of the HA02s I have to wire them up to pairs of cells not individual cells). This means that when there is a high load when the batteries are fully charged and that suddenly switches off, I still get the jump to a 2.6% deviation for a while. However, the HA02s continute to work and over the last 4 weeks they are now improving these voltage differences in the individual cells so that when this high load switch off occurs I'm not getting that jump in deviation as before. I've also noticed that they have changed the deviation on the strings so that sometimes the string that always had the lower voltage has the higher voltage - this is clearly when the boxes are passing more current to the lower voltage cells to equalize everything. Hoping that in the next few weeks all the voltage differences will be ironed out and I won't have any seriously damaged cells. The 4 cells with very low voltages are now perfectly balanced with voltages in the average range. They higher voltage cells are still slightly higher but nowhere near as much as before. I'm not expecting miracles so very happy to wait another few weeks for HA02s to complete their work with this very large, deep charge, battery bank. Just fabulous little black boxes!