diy solar

diy solar

Has this been discussed somewhere? Are there inverters that automatically start up when connected to 12V?

HRTKD:

I would love to buy a Victron, but unfortunately that's not in my budget. I have to stay within what is actually possible for me at this time.
 
Everyone:

There is a 1200W / 1500W @15min / 2400W peak inverter by EPEVER that has a fairly low standby power use of less than 12W, which comes to 0.288kWh if left on 24/7, which is fairly low, especially given the price of this inverter.

However the 3 x 160W panels I currently plan to use will generate around 1.2kWh per day in winter where I live (4 hours of sunshine), so this would cost me about 25% of my "harvest" just to have the inverter on standby. To make it worse, I will only get 1.2kWh a day on average in winter - there are a few times in winter where there will be no real sunshine for two or three days in a row.

So if I went with this inverter and leave it on 24/7, I would want to add one more panel just to cover the inverter.

However space to put panels in this application and location is very limited, so this is challenging.

Also, if in future I need to replace the inverter, I may well only find an inverter that has a higher standby consumption than this one, and then I'm back to square one, needing to design a solution then to switch off power to the inverter automatically when 240V power is not needed.
 
Everyone:

I am a bit surprised that, at least as far as I can see, there doesn't seem to be a readily available solution to deal with the issue that leaving an inverter on standby in small solar installations would eat a very major chunk of the power such a small install can generate.

The golden ticket here seems to be what some of you mentioned, an inverter that goes into a very low / near zero consumption "sleep" mode unless and until a load above a certain power draw is connected.

If I can find an inverter that does that and that fits in my budget.

I would be super grateful if you can point me to any inverters that might fit the bill - 12V, 220/240V, around 1500/2000Wpeak or better 2000W/3000Wpeak, and with a very low consumption "automatic sleep" function as described above.

As before, I really value all your input - whatever thoughts or ideas you have, please keep them coming!

I hope your day is being nice to you, be well!
 
Imho there is no "cheap"ready-for-use solution available. My personal approach would be:
- avoid any (expensive) switching on high current 12V side
- Buy an (inexpensive) inverter with remote control which fits the power needs
- Build the desired logic (i.e. temp sensors...) with some programmable module, i.e. cheap Arduino-like clone. That's just a few bucks.

If you want to go that route but are not familiar in programming, I'm here to help ?

Edit: Oh, there's a potential cheap and simple plan B: Open the Inverter and solder some "remote leads" over the power switch. Just attach a relay contact to that "bypassed" power switch. At least on mine cheap-o Inverter, that switch is rated with 5...10amps, thus any standard relay will survive such switching.
 
Last edited:
As before, I really value all your input - whatever thoughts or ideas you have, please keep them coming!

Be cautious. There has been posts re inverter failures by heavy startup loads from the inverter idle.
The solution was keeping the inverter out of idle with a small load running continuously.....totally against the reason for your post....and to complicate things further.... I would give a turned off, ebay hf inverter time to 'boot' before hitting it with the fridges startup load.
 
A quality inverter will have provision for remote switching,
Your temperature control unit will close contacts, connect two wires from these to the inverter control input. Job done.

Any high current relay on the DC input will waste power and eventually fail.

Mike
 
FYI, there is no need for a 3000W inverter to power a freezer compressor.
1000W hf works fine for every one I've tried.
My renogy 1000W cheapo has a remote switch as do all my little ones.
Easy two wire.
No capacitor loading.
I'd get a samlex/Aims/you pick... or similar lf 1200W and use the remote switch terminals for thermostat control.
It has massive steel transformer to handle 3600W startup loads easily...
 
Hey all, thanks for yet more input!

Mmame:

I hear you (and others) re all that...

I might potentially take you up on that "cheap arduino" offer, thank you so much for offering! I believe I would be a fairly straight forward candidate to help get this set up.

And actually the "switch extension" is also pretty neat, I will definitely be looking into that, and am asking myself why I didn't think of that!

Although what Sunshine wrote is reason for some concern... especially since I am aiming for a system that I will not need to baby for many years to come, I want something I can rely on, if possible with my budget. More on that in a bit.
 
Sunshine:

"Be cautious. There has been posts re inverter failures by heavy startup loads from the inverter idle." Thank you for sharing this, very important information. I assume you are referring here to inverters with a "sleep" function yes?

I am also still a bit concerned about the surge load that powering up the inverter would mean; unaddressed, it would be hard on the batteries as well as potentially on the DC relay.

See next post.
 
Let's see if we can beat all of those issues with this SOLAR FRIDGE MASTER PLAN (TM):

1. Relay 1 for inverter power: High current relay between the batteries and the inverter. Leave the inverter switch in the "on" setting.

2. Relay 2 for inverter soft start: High current relay, also between the batteries and the inverters, parallel to and similar to relay 1; but connect this relay to the inverter via a suitable resistor.

3. Relay 3: Relay between the inverter and the fridge.

4. Throw a temperature sensor into the fridge. Set the thermostat of the fridge itself to its lowest possible temperature.


5. Arduino the whole thing together so that it does the following every time the fridge gets above its maximum allowed temperature:

5.1 Relay 2 connects DC power to the inverter via the "slow start" resistor

5.2 30 seconds after that, relay 1 connects "non-resistored" DC power to the inverter

5.3 30 seconds after that, relay 3 connects the fridge to the AC side of the inverter - so the inverter has enough time to fully start up before the load of the fridge start up hits it.


6. Rig up some remote override switch option to the Arduino so I can manually power up the inverter on the odd occasion where I need it to be on for other reasons than powering the fridges.

7. If I add a second fridge, I need to add another relay 4 similar to relay 3, and another temperature sensor in that fridge, but at that point it's just copy-paste really.
 
How does the above sound? Does anyone see any issues with this plan of attack? Weaknesses? Likely points of failure? Ease of repair / replacement?

I don't mind if I might need to switch out a relay every few years, relays are easy and cheap to come by and trivial to swap out. But I don't want to need to replace relays every year.

I also want to avoid fires (obviously) as well as early failure of expensive or harder to replace components.

And finally I don't want to have to spend lots of time babysitting the system year round, so all this is only an option if this can be set up in a reliable way.

The above system to me looks like it might do those things... however I was really hoping for a slightly more plug and playish solution than this! Rather than in essence engineering an entire new device, which kind of is what this is turning into now.

It also looks like this will add a couple hundred dollars to the project once all is said and done... but if it means my inverter will live a long healthy life then that's worth it.
 
Do you think the above system would make sense, or would I be better off to just throw one or two more 160W panels on the roof and accept the inverter standby waste being around 30% of the total production of my system, and keep the inverter on 24/7?

It might be a more reliable solution with less that can go wrong I'm guessing.

Also, with the currently planned 3 x 160 watt panels, I'm right at the edge of what my charge controller can do at 12V (an EPEVER Tracer-BN 40A), so I would need to go to 24V just to add that one or two more panels for the inverter losses.

Then I need a 24V to 12V converter for my 12V devices like laptop, multi USB charge hubs, some lighting etc.

I have deliberately not bought the batteries yet so the 24V option is still open to me at this point.

However I am not sure that I can even find space for more panels at the location where this system is going.

And I loathe the idea of the inverter standby using up 30% of my harvest, that's just insane to me.
 
At this point in time I think both the above "Solar Fridge Masterplan (TM)" and the "just go to 24V and add one or two more panels" option are viable. Can't quite make up my mind between these two yet.

I don't want to go with the "just add more panels" option just because that's the common way of dealing with inverter losses. I want to make this decision based on judging the pros and cons of both options as objectively as is possible.

As always, I super appreciate all your thoughts, ideas, and pointers...!
 
Mikefitz:

I didn't see your post until now, sorry!

"A quality inverter will have provision for remote switching, Your temperature control unit will close contacts, connect two wires from these to the inverter control input. Job done."

I guess what you are suggesting here could be applied in my above "Solare Fridge Masterplan (TM)", in essence doing what you suggest instead of the two relays between the batteries and the inverter.

Would that make sense to you?

"Any high current relay on the DC input will waste power and eventually fail."

Why would it fail if it is designed for those currents? I mean eventually all relays will fail, they are mechanical devices after all, but why would it fail prematurely? I would think that one can buy quality relays with a rated number of switch cycles under the load they are designed for, and schedule swapping them out accordingly?

Also really not sure how a relay would waste power to any noteworthy degree, other than then little bit of power it needs for its electro magnet?

Please know that I am not trying to dismiss what you are saying at all, just really not sure I understand.
 
Simplest solution is to just connect the fridge thermostat to a relay that replaces the inverter on/off switch. Add a 10 second delay relay for the fridge compressor. This gives the inverter time to boot up and level out. Both can shut off together.
 
Supervstech:

I will need to power higher loads on occasion, devices with around 1500W startup current or potentially more.

Also I may run more than one fridge, and if both start their compressors simultaneously by coincidence now and then, it could trip a too weak inverter - if that happens while I am away, I will lose lots of food.

"1000W hf works fine for every one I've tried."

Useful to know, thank you! Do you mean 1000W constant here right, not peak? How much peak did the ones you tried usually have, roughly?

"My renogy 1000W cheapo has a remote switch as do all my little ones. Easy two wire. No capacitor loading."

Why does this one not have a capacitor? How does it ensure it can provide its peak / spike power when needed, or is it just lower than in inverters with capacitor?

"I'd get a samlex/Aims/you pick... or similar lf 1200W and use the remote switch terminals for thermostat control. It has massive steel transformer to handle 3600W startup loads easily..."

That sounds feasible.

The only issue I see with that long term is that when that inverter dies its natural death and I need to replace it, I then need to find a similar inverter that also has a big steel transformer, which limits my options, potentially making it difficult to source inverters. I would like to set up the system so it will work with nearly any inverter.

Also, what is the efficiency of this inverter?

Not critcizing what you said, just digging deeper into it to understand better!
 
Timeselectric:

"Simplest solution is to just connect the fridge thermostat to a relay that replaces the inverter on/off switch. Add a 10 second delay relay for the fridge compressor. This gives the inverter time to boot up and level out. Both can shut off together."

Hm I'll think this through as well, it does sound simple.

Brain fried a bit right now, I'll get back here later again though!

Thank you :)
 
Mmame, Timselectric:

"Open the Inverter and solder some "remote leads" over the power switch. Just attach a relay contact to that "bypassed" power switch. At least on mine cheap-o Inverter, that switch is rated with 5...10amps, thus any standard relay will survive such switching."

"Connect the fridge thermostat to a relay that replaces the inverter on/off switch. Add a 10 second delay relay for the fridge compressor. This gives the inverter time to boot up and level out. Both can shut off together."

As much as I would like to power the inverter up only when the fridge needs power, and as much as I like the way to achieve that that you are suggesting, the more I think about it, the more I am increasingly getting worried what switching an inverter on and off as needed by the fridge, which is at least once an hour, 24/7/365 might do to the inverter mid term, no matter how this is achieved.

In a very roundabout way I am thinking that most inverters will likely not have been designed with this frequency of being switched on and off in mind.

And if this for example halves the lifetime of the inverter which is a fairly expensive piece of equipment and thus doubles its capital cost per year across its lifetime, I might as well have invested the difference in more panels and potentially going 24V, and not have to worry about my inverter. Hm.
 
My power needs are estimates only at this point, but I think a bit over 1 kWh harvest per average winter day will be what I need.

Where I live, the 3x160W panels will give me approximately 1.2kWh/day on average in winter. At 12V, they also max out what the EPEVER Tracer-BN 40A can do without over paneling.

I wonder if the best plan would be doing this in phases:

Phase 1: Build the system as I had initially intended - 3x160W panels, 12V, EPEVER Tracer-BN 40A, find an inverter that is powerful enough for the fridge startup but has a reasonably low power consumption, leave the inverter running 24/7 for maximum longevity and system simplicity, observe and measure and see if this might actually be sufficient. I might also glue some thick styrofoam sheets to the outside of the fridge as well - where it is I don't care at all what it looks like, it's visually scratched up anyway, and this might well save me a couple hundred Wh a day already.

Phase 2: If this turns out not to be quite enough power but only a little more is needed, add 1 or 2 panels, leave the rest of the system as it is, and just over panel the charge controller. Yes, I'd be throwing away the potential harvest from those 1 or 2 extra panels in full sunshine, but during less than ideal times throughout the day I'd be generating more than before, which might already be enough to make up for the little bit extra that's needed, and I won't need to change anything else about the system, just parallel wire one or two more panels in and be done with it.

Phase 3: If more than just a little bit of power is needed, buy a second battery and go 24V so the charge controller can actually make use of the full potential of the 1 or 2 extra panels when the sun is shining fully, and add a 24V/12V converter for my 12V devices. This will then absolutely and definitely cover all my power needs. Although I would prefer not to have to do this.

This seems like a reasonably feasible way to incrementally expand my system until it fully does what I want it to do; without investing in things that I'll throw away along the way.

Thoughts?
 
Back
Top