Ironman

Turns out that I have a 1500W borehole pump - even larger startup surge... Heaters are like sweets - if you have it at your house, you will eat sugar. If not, you just get used to it.

So I got the Carlo Gavazzi ET112 working today, using the Venux GX to report back tot he VRM. Connecting direct to the Venus GX from my phone is the best thing! I get a second by second update on power consumption. Walking around the house and testing every single thing is very eye opening: - A hairdryer (1700W) vs hair straightener (35W) ... - Borehole pump: 1100W - Geysers 1800W each - lights: many LEDS are 60W, old style lights (50W 12V) add up very quickly. I immediately took all the heaters from my property and put them on the street. Whoever can afford to use them can take them.

I have a borehole with a 1.1kw single phase motor 50l pressure vessel and a pressure relay that switches it on when water pressure gets too low. It has a large startup surge that will cause issues. is there a control circuit that uses something like this schneider unit to soft start the motor? It is a soft starter for that size, and needs a dc or ac control signal. My pump’s current control circuit includes a phase angle detector to protect against running dry. So, preferrably need a whole control circuit for a borehole pump, but it must eliminate the startup surge.

So is the BMV700 seen as something different than a Pylontech BMS by the Venus devices? I lost a whole set of Royal batteries (8 * 105ah) due a burnt contactor on a Microcare 60A MPPT. The batteries could not disconnect from the tiny load of a 12V alarm system, and ran completely dead. We don't visit that house in summer due to Malaria. Could have been dead for 4 months - we will never know. Replaced the contactor and the MPPT is still going three years later.

I would worry if using lead acid with a BMV... Would it keep the batteries alive? You're gonna have to explain a bit more?

First parts:

What is the difference between Bluesolar and Smartsolar MPPT? Can I run multiple MPPTs from one Venus GX? How will the reporting work? Will it sum the outputs? Can you see each separately? Can I run mutiple ET112 meters on one Venus GX? One for the main supply and one of the *Always on* circuits? Or does the Multiplus II measure the *Always on* output separately from the inverter output (which will also feed back into the rest of the loads)?

@phil.g00: Conceded. Using the same panels, the power delivered at a fixed voltage varies almost linearly with irradiance. So there I was In the Linder Auditorium, listening to the Johannesburg Philharmonic Orchestra playing Chopin's 2nd piano concerto - as one does on a typical Wednesday night - and I just kept thinking about the curves on the PV panels - the graph that @plonkster posted is the same one from the Canadian Solar 300W specs, with the red curve for full irradiation. I kept thinking - mmmm take the Voltage from the Vpmax on the red curve straight down to the grey curve - and the power you get is almost a direct linear reduction at the same voltage. So I was completely wrong, and it was so frustrating because I had to wait until the end of the concert to come home, and online again... And by that time Plonkster had given the full explanation. Cool. I love this forum. It really is like that meme where my wife looks at me and thinks "I wonder if he is thinking of other women?", but I'm thinking of PV power curves. Luckily my wife knows me really well by now.

If you adjust the sun angle between identical panels, the one with the lesser sunlight will act very much like a panel with lower Vpmax and Ipmax. Please compare a panel's delivery graphs between 1000W/sqm and 800W/sqm irradiation. We are talking about mounting strings at different angles here.

I agree with @plonkster: I have tested this: When running two 90W panels in parallel with two 140W panels. The 90 panels have a lower Vpmax and Ipmax than the 140W panels. In full sun, When I blanket the 90W panels, I only lose about 15W of total power. When I blanket the 150W panels, I get almost 180W from the two 90W panels. So the 90W panels contribute way less than they should, because the voltage is held too high by the 140W panels. They only start delivering current when the voltage is brought down to their Vpmax level.

If you have one string facing east, and another facing west and you parallel the two strings on one MPPT, this will happen: When one side has much more sun that the other, the voltage will be tied together, and the MPPT will track the power point of the side with the most sun (the hot string). The voltage will be lifted to the max power point of the hot string, and the same voltage will be applied to the cold string - higher that it would have for its own maximum power point. The cold string will not deliver any power like that. Having two smaller MPPT's, one for each string will be more efficient because every string will have its own max power point tracked. Two MPPT's will always deliver more energy than a single MPPT when the two strings have different solar lighting times / profiles / shading. I have studied this issue over many years, running mismatched panels in parallel, with partial shading also complicating the issue. Ideally you should have one MPPT for every angle, or for every string that has any reason to differ its max power point from the other strings.

Agreed

I keep thinking of what we, PV enthusiasts, are doing to the grid when we employ grid-tied PV installations. Taking things to extreme - what load would the grid see if EVERYBODY had this installed: 1) A hybrid, grid-tied PV system, using a Carlo Gavazzi energy meter to block any grid feedback 2) A hybrid system that sells excess PV energy back to the grid In case one, there is a time delay in the feedback loop that causes current to be used from and pushed back to the grid. You will have almost no base load, except for extreme spikes in both directions: If the consumer puts a kettle on, current will be drawn from the grid for a period until the hybrid inverter starts taking over - decreasing current to almost 0. Conversely, when he puts the kettle off, the hybrid inverter will push current back to the grid for a period, before the CG feedback loop reduces the current feed, decreasing the current feed. Multiply that with 100 000 consumers... You will end up with a very low base load and current spikes (positive and negative) being generated by every one of these consumers, geographically distributed ... I just think that seems to be a recipe for disaster? The low base load will decrease the income that the grid can receive to such a level that they will not be able to support the technical infrastructure. In case 2, there will be less spikes, and smoother energy delivery in aggregate, but you will get a different problem: the duck curve, where the grid has to deliver much more energy early mornings and during the evenings. The only real solution for this curve is grid-scale energy storage. Which seems great, but I don't think we have costed that into the energy delivery pricing. Who is going to pay? What is the solution? I don't know, but I think we need to be aware that in 10, 20 years time when there are 40% grid tied PV installations on a grid, the energy landscape will have to look very different.

Any update on the 5kva Multiplus II stock?

Can you generate a graph that shows your depth of discharge for every day? The worst time is right before your start charging in the morning, especially if the previous day was cloudy. If you regularly discharge them to 50% or lower, they may not even make one year. There are many examples like that on these forums. I have a bank of 12 of the same Royal batteries in an off grid system. But mine gets charged fully and discharged only about 3% every day. We use the pace about 8 weekends per year, when the batteries discharge to about 50%. The batteries are now 4 years old. They are my second set, the first one lasted three years. But that was due to a Microcare MPPT battery contactor that oxidised, overloaded and burnt and stopped charging the batteries. One 100% discharge - batteries dead.

Thanks for all your input.

My kitchen is very for from the solar heated geyser. What options are there for a very small geyser, or any other way of instant hot water? I am changing to a grid-tied hybrid PV solution, so the kitchen sink must not draw high power for short periods. 4kw is way too much. I would prefer under 2kw, for longer periods. So it seems I am looking at a small under counter geyser, with a small element, maybe connected to dumb timer that switches it off overnight? Kwikot has a 10 liter that uses 1500W - at R2450 Lorenzetti has a instant heater that uses 4400W for R1300 - too much power! Heattech as a 15 liter that uses 1500W, at R1899 Seems like the Heattech is the way to go. Any experience with these? Plus point: They are a South African manufacturer. Any other alternatives I should consider?

...but then the system will never be legal, over 4.6 kw, grid tied, 80amp supply. So I don’t think that is an option. I got a quote today for the energy meter / Venus GX, and as soon as sparky is ready to split the db, I will purchase and get that installed to measure. Then I will have a very good idea of the 3kva / 5kva decision. Thanks for all the help.

I don't think that is how the MP II works... If I am wrong, will the people with Victron ESS knowledge please jump in and correct me? I am going to move my critical load to the /*always on*/ output of the Multiplus II: Lights (LED) Computers/TV/Internet, cellphone chargers Freezers / Freezers Borehole... These should be less than 2400W total - if not - that is a case for getting the 4000W model. The point is - I have to make peace with the fact that in a loadshedding mode, only these items will have power. I want these items to be able to be carried 24/7, as longs as there is sunshine during the day, so I don't want to overload this circuit. All other loads: oven, geysers, coffee, hair dryer etc - they will be dead - the Mupliplus II will disconnect itself from the main DB where they are drawing power. Geysers at least have direct solar heating too. However, while there is power from outside, all the non-critical loads will be supplemented by the MP II from battery AND/OR solar, up to 2400W (or 4000W) - to save money! And I think that if I know there are multiple loadshedding periods coming up, I can always temporarily increase the reserved SOC from 30% to say 70% to help the batteries to survive the virtual off-grid situation.

I am starting this whole process with just the energy meter / Venus GX / VRM portal just to measure. I think that a balanced overview of my loads will help a lot in shifting loads to different timeslots and deciding between the MPII 3VKA or 5KVA.

The Multiplus 5 KVA has a maximum continuous capacity of 4 KW.

...just for interest - how much energy do you discharge from it daily?

Cool! Thanks.

As you mentioned - I like doing my homework. So... Would it make sense to start with this only: Rewire main DB, add /*Always on*/ DB Install the Carlo Gavazzi energy meter in the main DB, connected to a Venus GX device -> internet... Questions: Will I be able to get usable time graphs on the VRM portal of the actual consumption that way? That should also help inform my further buying decisions. The Carlo Gavazzi page only references CCGX - will it work as well with Venus GX? I see some South African users on the VRM portal demo page - showing actual usage data for today! Are they aware that the word can see their stuff??? As a side note: I have probably been into solar PV longer than some of you - I started running a small off-grid system in Sabiepark (no Eskom allowed) in the early 1990s! Some history: The system started with 4 x 50W Siemens Panels, A 30A linear voltage regulator 4 x 180AH 12V batteries bought second hand