diy solar

diy solar

feedback on first solar design?

owad

New Member
Joined
Nov 27, 2021
Messages
8
I'm designing my first solar power system. I have the solar panels already (found a good price on them locally), but haven't purchased anything else yet. Could I get some feedback on the diagram below? Do the parts I'm choosing look reasonable and properly specced?

This would be for an in-house system. Half interested in just playing around, half in keeping essentials running during a power outage. In the latter scenario, I could dramatically cut back on water use and turn off the toys, to keep the freezer going. The freezer is out on the back porch, where it will need more power in summer than winter.

I like the ElectroDacus and the idea of being able to add more panels affordably using it, but I think the voltage on the CS3U-355PB panels is too high to make using the DSSR20 reasonable for them. Is the SmartSolar MPPT Tr 150V 60A a good choice here?

My thought with the battery charger is that I'd like the system to be usable in a power outage that comes after a few cloudy days.

I was leaning towards the EVE 230s because they were priced very reasonably at powerwholesale.net, but then registration wasn't working on their website, there were email issues, and I was told credit card payments are blocked, which is making me uneasy.


solar.png
 
I love looking at people's SBMS0 designs. They are some of the most fun/interesting to me.

I'm a bit rusty (in general, but in particular my knowledge of the SBMS0) but I've got a few thoughts:

1. If your only reason for the Battery Protect is a way for the SBMS0 to switch loads, you could optionally choose to consolidate two devices into one by using one of Victron's DC-DC converters for your 12V loads. Like their Solar Charge Controllers, some or all of the Victron buck/boost converters can be remotely switched, using one of those could save you a component. You'd still have to figure out the 5V though (or could just place the 5V converter downstream of 12V converter 24V --> 12V --> 5V not the most efficient but maybe the simplest)

2. Not sure if the Giandel inverter is a solid choice. First, IIRC they are not an easy option to use with the SBMS0 (don't have the right remote on/off capability I think--possibly I'm misremembering). Beyond that, it looks like you are using mostly quality stuff, Giandel would be a bit of a break from that.

3. It doesn't look like you have a high AIC main battery fuse. I know the limitations/quirks of the SBMS0 force you to get a little more creative with this, but its an important safety consideration for LiFePO4 banks.

4. You are pretty overpaneled, 355 * 8 = 2840 / 25.6 = ~110A (or ~105A at 27V). There are two things you want to consider here (1) can the controller handle it, Victron Controllers are some of the best when it comes to overpanelling from what I've read, but you'll want to make sure you are within the max input voltage and input current limits, (2) are you willing to give up that much potential power in peak solar conditions?

5. No reason not to add a charger, but the logic of "My thought with the battery charger is that I'd like the system to be usable in a power outage that comes after a few cloudy days." may or may not make sense. If this is just a backup system sitting in reserve until a power outage, assuming there are no loads connected. If I misunderstood and the system is not just a backup, then yeah, an AC charger may make sense.

6. What's the thinking behind the one lone DSSR20 with 100W panels?
 
I love looking at people's SBMS0 designs. They are some of the most fun/interesting to me.

I'm a bit rusty (in general, but in particular my knowledge of the SBMS0) but I've got a few thoughts:

1. If your only reason for the Battery Protect is a way for the SBMS0 to switch loads, you could optionally choose to consolidate two devices into one by using one of Victron's DC-DC converters for your 12V loads. Like their Solar Charge Controllers, some or all of the Victron buck/boost converters can be remotely switched, using one of those could save you a component. You'd still have to figure out the 5V though (or could just place the 5V converter downstream of 12V converter 24V --> 12V --> 5V not the most efficient but maybe the simplest)

Yes, remote on/off for the SBMS0 was my reason for the Battery Protect. Here in particular, but also in general, can I just use a relay? Can I use a relay to kill power to the inverter? It doesn't seem like people do that with the SBMS0, but I'm not clear on why.

3. It doesn't look like you have a high AIC main battery fuse. I know the limitations/quirks of the SBMS0 force you to get a little more creative with this, but its an important safety consideration for LiFePO4 banks.

Thank you.

4. You are pretty overpaneled, 355 * 8 = 2840 / 25.6 = ~110A (or ~105A at 27V). There are two things you want to consider here (1) can the controller handle it, Victron Controllers are some of the best when it comes to overpanelling from what I've read, but you'll want to make sure you are within the max input voltage and input current limits, (2) are you willing to give up that much potential power in peak solar conditions?

This is a big over-sight on my part: I was calculating the input current.

I was originally planning 2000W of panels, but then I ended up finding this setup for less money. So I went with it, but it's pushing all my other costs up, of course. I guess this mean's I'd want the SmartSolar MPPT Tr 150V 100A, which is a big jump in price.

What's your opinion on using the DSSR20? The current is just a little too high, so I think I'd need one for every panel, which is unfortunate, but still a lot cheaper that the Victron charger. Distance from where I want to put the batteries to the charger is going to be about 75 ft, which I think makes my 72-cell panels a more reasonable choice for ElectroDacus, but also means I'm going to be spending a fortunate in wire – unless I can put the DSSR20's out at the panels?

5. No reason not to add a charger, but the logic of "My thought with the battery charger is that I'd like the system to be usable in a power outage that comes after a few cloudy days." may or may not make sense. If this is just a backup system sitting in reserve until a power outage, assuming there are no loads connected. If I misunderstood and the system is not just a backup, then yeah, an AC charger may make sense.

I'd like to run the DC load 24/7. If it's practical to always run the pump off the system, I'd like to do that too. I think I'm under-specced to put the freezer on full time. I haven't tried to generate any data yet on how much the pump actually runs. The freezer is going to use a lot more power in the summer, but I won't be able to measure that until summer.

6. What's the thinking behind the one lone DSSR20 with 100W panels?

I already have the panels and I wanted to play with a DSSR20. And I'd still be able to generate a bit of power with it, if something happened to my main array.
 
Post the actual specs for your panels. 355W panels can't have the values you posted for Vmp and Imp since Vmp x Imp should equal the panel wattage. 39.2V x 11.73A = 460W which obviously isn't 355W.

You might want to split your panels into two groups each with their own charge controller. You could do two sets of 2S2P each on a 150/60. If the panel Voc is about 42V or less you could use two 100/50 charge controllers but it depends on the coldest temps you will see and the exact Voc.
 
It's actually 39.4V, not 39.2V. I was looking at the wrong panel when I wrote 39.2.

The panels are bifacial. 11.73A is the maximum current at 30% gain. 9.02A is the current from front side only. I doubt I'm going to get much, if any, value out of the bifacial aspect, but I thought it necessary to account for it.

Screen Shot 2021-12-11 at 5.46.55 PM.png
 
OK, the Voc is 46.8V. That's too high to have two in series with a 100V charge controller unless you NEVER get below 75ºF. So you would need two 150/60 charge controllers with two sets of 4 panels in 2S2P each. That would let you use the full 8 355W panels with your 24V system.
 
Two 150/60 SmartSolars would cost $1080, while eight DSSR20 would cost $308. At that price difference, it seems I'd be better off putting the money into more panels, no?
 
Too many panels (total wattage) is why you need too many charge controllers.

You could build a 48V system then your 8 355W panels would only need one 150/60 charge controller instead of 2.
 
Yes, remote on/off for the SBMS0 was my reason for the Battery Protect. Here in particular, but also in general, can I just use a relay? Can I use a relay to kill power to the inverter? It doesn't seem like people do that with the SBMS0, but I'm not clear on why.
Actually I believe both the DSSR and the Victron Battery Protect, are essentially fancy relays (Solid State Relays). But neither could be used with an inverter directly. Not sure if you could use a traditional mechanical relay of some sort, I don't see why not. But you may be better off just choosing a different inverter that can be switched by the SBMS0 directly, Victron inverters can, Samlex inverters may be able to, not sure what other brands can but I'm sure there are many, Dacian lists some examples in the SBMS0 manual, and there is more info on his message board.

Possibly one of the reasons you don't see relays used more with Dacian's BMS is that part of the appeal is the simplicity/logic of switching the chargers and inverter directly is what brings many people to it in the first place, so you'll get a natural bias towards that approach.

I was originally planning 2000W of panels, but then I ended up finding this setup for less money. So I went with it, but it's pushing all my other costs up, of course. I guess this mean's I'd want the SmartSolar MPPT Tr 150V 100A, which is a big jump in price.
Yeah, the larger Victron units get pretty damn pricey
Although.. A quick peak at amazon has me thinking the 100A model is actually a better value than the 60A ($640 for 150V/100A, vs $540 for the 60A) There are multiple listings for the 100A at different price points so you'd want to double check to make sure the model would work for you, but it appears to be sold by Victron's official store.
What's your opinion on using the DSSR20?
I've not seen enough data to have a strong opinion. I like the simplicity and low cost of the DSSR20's. I'm not fully convinced of Dacian's argument against MPPT for the use-cases that matter to me. I think his argument is strong based on the assumptions: (1) space is not a limitation (2) panels can be sourced cheap (3) you are designing for the longterm (4) cost and/or simplicity are the priorities.
The current is just a little too high, so I think I'd need one for every panel, which is unfortunate, but still a lot cheaper that the Victron charger.
Probably this won't be helpful but I recently read that there should be a higher current DSSR coming down the pipeline (50A I believe). Also with the 20A model, I believe this rating can be improved upon with active cooling.
Distance from where I want to put the batteries to the charger is going to be about 75 ft, which I think makes my 72-cell panels a more reasonable choice for ElectroDacus, but also means I'm going to be spending a fortunate in wire – unless I can put the DSSR20's out at the panels?
Hmm yeah, this is the other limitation/tradeoff of the DSSR approach. For longer wire runs (75ft between panels and batteries = 150ft roundtrip which is what you would use to calculate the voltage drop) this could get expensive.
I'd like to run the DC load 24/7. If it's practical to always run the pump off the system, I'd like to do that too. I think I'm under-specced to put the freezer on full time. I haven't tried to generate any data yet on how much the pump actually runs. The freezer is going to use a lot more power in the summer, but I won't be able to measure that until summer.
Ahh i see, then yes you are right, an AC charger for back to back to back days of shade probably makes sense.
I already have the panels and I wanted to play with a DSSR20. And I'd still be able to generate a bit of power with it, if something happened to my main array.
As good a reason as any :)
Actually when I was looking into building a system with the SBMS0 I was thinking about doing something similar. I'm not sold on Dacian's pitch about the logic of non-mppt over mppt outside of his specific design model (I see the logic for his design conditions/assumptions, but mine were different). And as mentioned I haven't really seen enough data to dismiss or confirm my hesitation/outstanding questions, so I was considering doing something similar to what you are doing running an MPPT and DSSR side by side. That project has been shelved for now though.
 
Two 150/60 SmartSolars would cost $1080, while eight DSSR20 would cost $308. At that price difference, it seems I'd be better off putting the money into more panels, no?
Factor in the difference in wiring and the additional DSSR's that would accompany the additional panels, and if the DSSR's still come out ahead and you have the space, than probably yes, the more economical decision would likely be the DSSR route. I think this right here is the heart of Dacian's reasoning. Cost savings put towards more panels will be a better value than the price premium for MPPT/higher efficiency.
 
Back
Top