MPPT 250/100 Absorption timer with NIFE

Question

question

bbobkins asked Feb 29, '20

MPPT 250/100 Absorption timer with NIFE

The Absorption timer for Lead Acid battery types is set at the beginning of every day. However this doesn't work well for NI-FE (nickel iron) as the voltage will rise once a load is disconnected. The batteries can be 100% depleted (down to 9.5V), the load will then disconnect and the voltage will rise to as much as 13V. If the Absorption timer was to be set at that point in time any charge controller will act as if the batteries are basically fully charged.

What I need is for the charge controller to keep track of the lowest observed voltage over the entire discharge cycle then use that value to set the Absorption timer. I'm guessing nothing like this exists right now, but that would be nice thing to have in the future.

For now however, can I do this via a raspberry pi through either ve.bus or ve.direct? If i were to manually set the Absorption timer Automatically myself, I assume I can use something like Linux VictronConnect to issue commands every morning from my raspberry pi?

A complication is I need 2 MPPT charge controllers in parallel so I would need to issue the commands to both controllers around the same time, and I would assume via ve.bus? Or would I have 2 ve.direct devices?

Obviously there are a lot of moving parts here, and I have a lot more research to do, but is what I'm proposing here even possible or can anyone here offer reasons why this is not such a good idea. :)

Thanks in advance.

MPPT Controllers battery charging

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Modifications Space

Node Red Space

Answer 1 · 2020-02-29T05:45:01Z

JohnC answered · Feb 29, 2020 at 05:45 AM

Hi @bbobkins

I haven't a direct answer for you, just considerations for your research.

There's a 'Tail Current' setting to drop out of Absorb, but will also include concurrent loads.

Also a 'Re-bulk voltage offset', which you might be able to work into multiple shorter Abs sessions.

More real-time than predictive options, but something you might explore.

Love to hear how you go with these batts..

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Answer 2 · 2020-03-01T01:52:21Z

bbobkins answered · Mar 01, 2020 at 01:52 AM

Ok thanks for that, yeah I suspect there is multiple ways I can go about this. It doesn't need to be very smart, and there is no such thing as overcharging these batteries, just how often you need to water them (a solution like what I'm proposing would fix that).

I'm currently using Epever charge controllers, and I'm thinking I could even control a rely on the solar input. Disconnecting the solar for a brief period should bump the controllers back into Absorption (this could be done multiple times a day as needed). This should work for both Epever and Victron controllers. Obviously this is a bit hacky and I would be more than happy to switch over to Victron if I can get control of the chargers. Ultimately If this works I'd like to work with Victron to create a NIFE charge profile that can be built in to these controllers.

2

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

JohnC ♦ commented · Mar 01, 2020 at 03:16 AM

Yeh, your system must be reasonably substantial when you talk a 250/100, and if that's at 48V then it screams a battery monitor.

So let's not muck about. A BMV712, a GX box (or a rPi loaded with VenusOS), and VRM portal access (free) will give you full monitoring. Graphs, alarms, n'all. 'Control' would need further definition, but there's a fair bit of that available, depending on exactly what you want to do with relays or suchlike.

I reckon VRM is perhaps Victron's greatest feature, and I can't live without it..

This would be just a kickoff point for you initially, but I really suggest a closer look is well warranted.

0 ·

bbobkins JohnC ♦ commented · Mar 02, 2020 at 04:40 AM

Actually I'm talking 2x 250/100 which is what I have now but with Epever. My system is currently 12V 1000AH but eventually I will upgrade this to 24 then 48V by simply adding more cells.

A battery monitor is useless as they know nothing about NIFE and I don't need many of the features you would need with other battery types. NIFE requirements are simple and quite dumb really, but unfortunately there is no specific setting for them which is what I'm trying to rectify.

As for VRM well I rolled my own, its way better but I may be biased there.

0 ·

question