question

augusto avatar image
augusto asked

Hot swapping solar panels to MPPT 250/100

Quite special setup here, I know.

An MPPT 250/100 will charge 19.2 kWh of lifepo4 batteries and a Multiplus II will give me AC.

To charge event on the cloudiest days the solar field is greatly oversized (18kW) and the same PV is used directly to power other loads (water or space heaters in DC straight from the panels). The idea is to connect/disconnect strings in real time to the MPPT as required in order to ensure that the Isc on the MPPT is always low enough and to let the excess PV go to the heaters without affecting MTTP or any component of the electric chain.

The swapping will be performed by an external device driving one relay per string (on the positive pole). The logic will depend on SOC, current delivered by the MPPT and potential errors. In a very simplified version it could behave like this:

  1. Battery not full and charging current < 50A for more than 30 seconds: connect one more string to the MPPT and wait 30 seconds for MPPT or anything else to stabilize (if it's the first string being connected enable the MPPT also).
  2. Battery not full and charging current >90A: immediately disconnect one string and move it to the water heaters, wait 3 seconds before testing again.
  3. Battery full or MPPT not charging: disable MPPT and move all strings to the heaters.

Each string will deliver 1500W maximum (aka 30A to 35A on a 48V battery) so it's important, to ensure stability, that the high threshold and the low threshold are more than 35A apart (in this example 40A has been chosen).

Now the 1 million dollar question: will the MPPT 250/100 or the PV be annoyed by said circustances?

I guess that a selected power point will get disturbed when swapping panels but it should return to the MPP at some point, my concern is if these disturbs could damage the MPPT or trigger a fault. It could look to the MPPT like a very fast cloud covering or uncovering the sun... veeery fast!

I'm not interested in optimizing the speed of charge. The MPPT of course won't be delivering its maximum current at any time in this setup.

Thanks

MPPT Controllers
4 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

Kevin Windrem avatar image Kevin Windrem commented ·

I wonder if it might be simpler to simply connect your loads to the battery bank as a "DC SYSTEM". Your logic for when to connect/disconnect would probably be similar and you wouldn't need to worry about switching PV strings.

I looked into what it might take to switch PV strings in/out of circuit and it seemed that relays that would handle the fairly high DC voltage/current arc might be expensive. But this might also be true for switching your heaters on/off the battery bank.

0 Likes 0 ·
augusto avatar image augusto Kevin Windrem commented ·

Thanks Kevin for your contribution.

The problem with the DC system solution is that batteries are 48V and to power standard heaters I'll need another power converter, also to get the 18kW PV in heating power once the battery is full the current from the batteries would be ridiculously high and, finally, the MPPT charger won't be able to keep up with the discharge being imposed on the batteries.

The idea to connect PV straight to the heaters was to avoid additional high power electronics and be able to handle big amounts of power that otherwise would remain unexploited. The array would be composed of 12 strings at 225V Voc and 9.6A Isc so relays for that task are pretty standard.

0 Likes 0 ·
ripper avatar image ripper augusto commented ·

225V AC is standard. You are gonna have DC! Nothing standard about that. If you confuse them, they are going to weld together either in use or burn the contacts off the instant you are trying to disengage. Even electronic relais, so called solid state relais, are far and few between with these kind of voltages.

0 Likes 0 ·
augusto avatar image augusto ripper commented ·

Ups, I haven't thought that DC was so much more complex than AC concerning relays but you're right, they are quite difficult to find and pricey. I got confused because some cheap relays indicate 250 V AC/DC in the datasheets.

Back to the original idea which was to create an electronic device using MOSfets (pretty much what's inside of a solid state relay) and some recirculation diodes to a bank of capacitors kept a little higher than Voc but lower than the breakdown voltage of the mosfets to absorb the load dump currents when disconnecting panels without affecting the MPPT during normal use.

Even so the issue will probably move downwards to the (most likely mechanical) thermostats of the heating elements. Those will weld or melt when disconnecting. I guess I'll have to integrate a thermostat sensing input in the DC switching device and leave the heaters always connected to it.

If you come up with other reasons why the whole thing is a bad idea please let me know.

0 Likes 0 ·
1 Answer
ripper avatar image
ripper answered ·

I ve had something similar, Mppt and resistance heater were parallel on pv array and resistance was switched on and off by SSR. That sure confuses it and leads to severe drop in performance in low light due to low pv voltage but it tries to come up with the best solution right away. It also overshot max rated A in rare circumstances in full sun light when doing a sweep which I had never seen before. Got a 100 20 which sits at 19.7A all day if necessary. Clocked it at 21A briefly with the resistance on. Removed it as was not worth it in the end.

2 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

augusto avatar image augusto commented ·

That is very interesting. My concern is if the sudden transition of PV power would cause spikes of current on the MPPT that make it fail. The duty cycle at which the MPPT operates would cause suddenly much more (or less) power to go through until it finds its new point. But if you've used it for some time I feel a bit less concerned.

I don't understand why you removed it, though. You could have turned off the heaters when light was low to improve low light performance.

0 Likes 0 ·
ripper avatar image ripper augusto commented ·

Was connected to an air heater to heat a cabin in winter time especially. 2kWp at 65V, Resistance 19 Ohm iirc to get about 250Watts of the 600W rated max out of it.

When I had 550W from the Mppt electrical I had that, and that was in summer . Anything less it went down to nothing or 100W at best in winter and took the electric watts down to nothing too. So I never saw this thing actually doing much and didnt want to leave it running unsupervised when untested. Played around for half a year and ripped it out.

Also didnt want to blow the Mppt, because it was properly confused.


0 Likes 0 ·