Jump to content

Adding batteries to an existing battery bank


DeepBass9

Recommended Posts

3 minutes ago, viper_za said:

Only thing worrying me a little here is how would it take begin configured to float and seeing the voltages are way above it when the other charger is charging the bank

Shouldn't be a problem. It has to deal with the possibility that an AC charger is in circuit, and that the AC charger might decide it's a good day for equalisation. :-) IT should just shut down until the voltage drops to the floating voltage it is configured for. At least, this is what a PWM will do. An MPPT will likely push the panels up to Voc so they can't produce anything.

Link to comment
Share on other sites

1 minute ago, plonkster said:

Shouldn't be a problem. It has to deal with the possibility that an AC charger is in circuit, and that the AC charger might decide it's a good day for equalisation. :-) IT should just shut down until the voltage drops to the floating voltage it is configured for. At least, this is what a PWM will do. An MPPT will likely push the panels up to Voc so they can't produce anything.

Now for me to find a cheap reliable 48v PWM charger then  and test this theory ;)

What will definitely happen is AICC will not know where the additional power comes from going into the bank or even out if draws pulling directly from the PWM array

Link to comment
Share on other sites

2 minutes ago, The Terrible Triplett said:

Plonkster, not sure if this may help, the smaller controller was the MPPT, ignored most of the time for the larger one was a PWM. 

Oh I can absolutely see this happening. Let's say you have both configured for 28.8V absorb, 27.6V float, but your larger charger is calibrated 0.05 higher than the smaller one.

In the early mornings, I would expect both to charge, until you hit absorption voltage. Now the smaller charger sees battery voltage of 28.85V and shuts down completely, so the other one does all the work. Now lets say the larger charger switches to float. By this time the smaller charger may already be in float... or if it isn't, guess what: Larger charger is floating at 27.65V, so once again it has nothing to do.

Add a load that's enough to pull it down below 27.6... and both should wake up. If it doesn't... well that would make an interesting case to investigate.

Link to comment
Share on other sites

Thys AICC wont know. Neither will the BMV. It will just see the volts go up. We tested it on Sarels system when he still had a WRND rhat still worked. He has to many panels for 3 Axperts so they put 3 panels on one of them. 

Link to comment
Share on other sites

I want to add something on this matter as well. Traditionally, battery balancers existed on 24V trucks to help  balance the two batteries and in many cases did work quite well with old and new batteries. The main thing to keep in mind though, the new battery won't necessarily last as long as it would have, if it was used together with a new set of batteries. 

So, yes, it does work a bit, but to some degree. 

But if you used something like the M-Solar range or "traction cell" batteries, you could easily replace a "busted" cell in the battery and have little side affects - fewer than with sealed lead acid. 

The main problem is that the older batteries will take longer to recharge, and the new one will then "over charge" even with a balancer it does that, yet somewhat less. 

Link to comment
Share on other sites

  • 2 weeks later...

Some further musings on this. One possible solution (maybe impractical) is to split your battery bank and then have two mppts charging each part. In that way you could mix old and new batteries or even different types of batteries. You would just need a transfer switch to physically switch your inverter from one bank to another. I'm not sure how you would decide how to do that. Maybe one day on one bank and one day on the other, with two days backup power available, Not sure if that would work in reality.

Link to comment
Share on other sites

3 hours ago, DeepBass9 said:

Some further musings on this. One possible solution (maybe impractical) is to split your battery bank and then have two mppts charging each part. In that way you could mix old and new batteries or even different types of batteries. You would just need a transfer switch to physically switch your inverter from one bank to another. I'm not sure how you would decide how to do that. Maybe one day on one bank and one day on the other, with two days backup power available, Not sure if that would work in reality.

This could very well work, but then you either need to run 2 inverters (very easy todo even on single phase, but would add about 30% cost) or have a "battery management system" which could switch between the two battery banks: i.e. either have a manual change over switch of sorts, or use something like an Arduino / Raspberry Pi which can measure the volts of both banks and switch over to the higher bank with a solid state relay, or perhaps even a switchover relay. 

The problem is, you're working with DC currents so you need to use something that can handle DC volts and currents. For a 5Kw/48V inverter you need to look at getting at least a 180A relay, or solid state relay todo this. I would aim for a 250A if possible. 

Something like this: https://www.omron.com/ecb/products/dry/3/g9ec-1.html  

Link to comment
Share on other sites

For other people (like me) whom are wannabe Fierce Releasers of Smoke: Bad idea to swap banks under load. VERY BAD.

Switch EVERYTHING off, then swap over ... and do NOT forget I said EVERYTHING ... think of say a charge controller(?) that may be connected separately. :D

Daai spark is bl.ks.ms (add i for .)  seer as jy dit verkeerd doen ... Been there, done that.

And the devices that allows you to swap between banks, I have seen a video on YouTube afterwards why it is a such bad idea under load. Cannot find it quick enough.

So IF you want to do this, you have to do this, switch EVERYTHING OFF, then swap, the startup again. But things like the BMV, temp sensors and all those little connections to change each time, eish, for me personally, it is a question of time before I let the smoke out yet again.

Link to comment
Share on other sites

1 hour ago, The Terrible Triplett said:

For other people (like me) whom are wannabe Fierce Releasers of Smoke: Bad idea to swap banks under load. VERY BAD.

Switch EVERYTHING off, then swap over ... and do NOT forget I said EVERYTHING ... think of say a charge controller(?) that may be connected separately. :D

Daai spark is bl.ks.ms (add i for .)  seer as jy dit verkeerd doen ... Been there, done that.

And the devices that allows you to swap between banks, I have seen a video on YouTube afterwards why it is a such bad idea under load. Cannot find it quick enough.

So IF you want to do this, you have to do this, switch EVERYTHING OFF, then swap, the startup again. But things like the BMV, temp sensors and all those little connections to change each time, eish, for me personally, it is a question of time before I let the smoke out yet again.

TTT you'r right! I totally forgot to add: "Don't do this under load!!!"

I think the idea could very well work, but it should then perhaps rather be seen as a backup storage system, than a live storage system. i.e rather only switch to the 2nd bank when the first bank is depleted. 

Or rather have two smaller inverters, one connected to each bank, along with a seperate MPPT connected to each bank. Perhaps, if you really want to, split the loads in the DB board as follows: lights on one system and plugs on the 2nd system. 

Link to comment
Share on other sites

22 minutes ago, SilverNodashi said:

Or rather have two smaller inverters, one connected to each bank, along with a seperate MPPT connected to each bank. Perhaps, if you really want to, split the loads in the DB board as follows: lights on one system and plugs on the 2nd system. 

EXACTLY what I did. 2 separate systems.

Main system is NEW batteries own controller and panels, secondary system that powers the Pi and lights and laptop, on older batts, own controller and panels.

Secondary system has a 420ah 12v batt bank of older maintenance free batts, not near enough for the 200w panel to charge IF depleted, so I just manage the load to keep it within the panels abilities, like less than 20% DOD at all times. MPPT controller gets it to float at least every 2nd day. :D 

Have to go an put that fixed 200w panel up ... maar ai, wat gaan ek DAN weer breek. 

Link to comment
Share on other sites

4 hours ago, The Terrible Triplett said:

EXACTLY what I did. 2 separate systems.

Main system is NEW batteries own controller and panels, secondary system that powers the Pi and lights and laptop, on older batts, own controller and panels.

Secondary system has a 420ah 12v batt bank of older maintenance free batts, not near enough for the 200w panel to charge IF depleted, so I just manage the load to keep it within the panels abilities, like less than 20% DOD at all times. MPPT controller gets it to float at least every 2nd day. :D 

Have to go an put that fixed 200w panel up ... maar ai, wat gaan ek DAN weer breek. 

The main problem with this is twofold: Extra wiring in / to the DB board (and hopefully the lights and plugs have separate neutrals), which might costly depending on how far away the inverter is. 

And you should preferably not use anything smaller than a 2Kw / 3Kw inverter per "phase" otherwise you will always be limited to that load capacity regardless of how many panels and batteries you have. i.e. if your kettle is 2Kw and your microwave is 800w, and you do use them together, install at least a 3Kw inverter with a high surge. 

 

BUT, you now have some nice redundancy, i.e. if one circuit / phase goes out for whatever reason, the other will still work. so if you replace batteries, or break an MPPT ( ;) ) or your PV's goes bust, you still have some plugs / lights. 

Link to comment
Share on other sites

I've had another hair-brained idea, which is basically to use the old bank as a "power bank", that is you dribble over excess charge during the day, and dribble it back at night. The only problem with that idea is that it is TERRIBLY inefficient, probably as low as 50% round trip :-)

Link to comment
Share on other sites

48 minutes ago, SilverNodashi said:

The main problem with this is twofold: Extra wiring in / to the DB board (and hopefully the lights and plugs have separate neutrals), which might costly depending on how far away the inverter is. 

And you should preferably not use anything smaller than a 2Kw / 3Kw inverter per "phase" otherwise you will always be limited to that load capacity regardless of how many panels and batteries you have. i.e. if your kettle is 2Kw and your microwave is 800w, and you do use them together, install at least a 3Kw inverter with a high surge. 

BUT, you now have some nice redundancy, i.e. if one circuit / phase goes out for whatever reason, the other will still work. so if you replace batteries, or break an MPPT ( ;) ) or your PV's goes bust, you still have some plugs / lights. 

Naaa ... the wiring was sorted YONKS ago. Lights power source are out of the board, still on the trips in the board. I have them on a switch with one point wired to Eskom, other feed has a 3 point plug on for i..e. inverter feed so I can swap sources between Eskom and anything else I want.

WhaTTT, 2-3Kw ... lights are on 350VA Victron inverter and computers, TV, routers and a fridge on a 1600VA inverter, never exceeding 750w. 

If power is off, microwave is off ... NOU gaan ons BRAAI!!! :D 

KISaS - Keep It Small and Simple.

Link to comment
Share on other sites

When I run my house with 6 people off 1 x 1600VA and 1 x 350VA inverter, off grid, I think I would have done well. :D

I am not phased to worry about ovens, stoves, hairdryers, microwaves, vacuum cleaners and such on solar. They use a minuscule amount of watts per month compared to the loads that are needed and on 24/7. 

The ONLY issue I need to resolve on this path ... the kids computers. They need high end computers for the games they play. Note: I rather invest in that power consumption to keep them home, off the road, out of the pubs ... but if Eskom is off, tough ... they can sit and chat with us at the braai. :D

Yes I can go all out and put it all on 5 Axperts with 100 000w panels on the roof ... but I rather spend funds on reducing the load strategically, as they break. Still a while to go for all the stuff to be as economical as it can be. 

Once all is settled, then I add more batteries for the load with 2 days backup, a LOT more panels, and I only have to swap one charge controller with a Victron one like Plonkster has.

EDIT: ok, maybe a controller or two extra. :D

Link to comment
Share on other sites

Just a thought here, why not consider diode separation? If you had your load on the back end of a set of diodes coming from each bank they would effectively not see each other. you would have to charge each bank separately and you would have some losses in the diodes. in your case 6kva max load roughly 125A dropping over 0.6 V would mean a max loss of about 75W but i recon under normal loads it will only be around 8W loss. got some nice some nice Schottky diodes for a project the other day that i think could do nicely. you would need at least 8 of them per bank and you'd have to make a chunky PCB.

http://www.vishay.com/docs/94145/16ctq060.pdf

anyone else think this is a viable option?  I think the main issue would be that you wont be able to charge the batts from the inverter.

Link to comment
Share on other sites

The diode drop of 0.6 is enough to cause issues with an a absorption voltage that would be too low. Schottky is an option, but the best option would be a MOSFET, because the act resistive and causes the least voltage drop.

Sent from my GT-I9195 using Tapatalk

Link to comment
Share on other sites

1 hour ago, Weasel said:

some reverse polarity protection circuits

Exactly like that, though I must admit, I only recently read about this way of doing it. Apparently you do it for low-voltage applications, 3.3V and lower, because a diode voltage drop is just too much, so conventional reverse-polarity protection can't be used.

Thing with MOSFETs, they have that inherent body-diode which conducts the other way, but doesn't have quite the response you might get from a schottky, so there might be some caveats here. I believe the way that's solved is to use two back-to-back mosfets, a layout that is especially common in pwm charge controllers, but that in turns adds a bit of complexity in driving them both (typically pwms use an opto-isolator to drive the other one).

Of course, that's still very crude. Ideally you would want the ability to run absorption on one battery for a longer period than on the other, for example. Really, you want to count amp-hours in and out of both.

So that gets me back to just charging them separately, and tying them together at discharge time.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...