diy solar

diy solar

Inverters with remote wire

No circuit diagram, but I could sketch something up. There's really not a lot involved though.

Going into the project I used a standard multimeter to check voltages and that's when I decided to use opto-couplers to keep the arduino away from the Giandel as the voltages are all over the place and certainly not TTL 3.3/5v compatible.

The optos that monitor the status LEDs just have their LEDs in parallel with the status LEDs, there is enough current available to drive them as is. I set the Arduino pins I use to monitor them to have their pullups enabled in input mode which let me get away without an external resistor to pull them up and just rely on the opto coupler to pull them down to ground when the status LEDs are on.

The Arduino itself is run via an ebay buck converter direct from the battery. The draw is so low that there's no need, IMO, for any sort of undervoltage lockout on the buck converter as the Arduino etc has such low draw that it'd never be responsible for taking the battery low enough for the BMS to setu in anyway.

I ran a meter over the button pair just now and for my 24VDC input inverter, off shows battery voltage across the button, on shows 14.4v across the button.

Awesome, thank you!!

Time to go source some optocouplers. :)
 
I'm late to the party but someone might find this useful anyway.

I have a couple of Giandel inverters, both 24VDC > 230VAC (single phase no 115VAC), one is 2kW, the other 3kW. Why two? In case one fails the other can be put to use and life continues. So far they have worked completely within expectations. They are HF inverters and exhibit the usual sag / slow rise all of these cheapies do when hit with a high current draw but beyond that they run everything I want them to. The 3kW is the primary and holds up my desktop computer, standard domestic 250l inverter fridge, (microwave oven (1.2kW from the outlet) or 2kW electric kettle), tv etc without complaint (as it should considering the power rating!). Overnight it runs my bedroom Fujitsu reverse cycle inverter (USA will call this a heatpump) split aircon.

I know highly inductive high draw loads can give cheap HF inverters grief letting the smoke out but I haven't tried to blow up my Giandels in this way, so I can't comment how they would cope with it.

Internally both inverters aren't a tin of spaghetti when it comes to wiring but as expected use the cheapest electro capacitors they can get their hands on so I expect that will be the point of failure in time, perhaps taking out transistors in the process.

They have remote soft-on support via a simple 4 wire connection. 1 wire for the button, two for status LEDs, one for ground. I have a $5 arduino board monitoring the battery voltage since the LV cut off for the inverter is below the LV cut off for the BMS and I prefer to not have the BMS potentially switching high currents if it can be avoided. The control logic is quite simple. If the desire is to turn the inverter on, check power LED status, if off, pull button line to ground until the power LED turns on and then release. To turn off, check if power LED on, if so, pull button line to ground for 250ms and release. If the fault LED is lit the arduino will not attempt to turn the inverter on. The arduino is isolated from the inverter with 3 opto-couplers.
I have a 12v 240v 2800w Giandel inverter with the remote switch, had it for a couple of years, never run anything over 2000w. I have ordered an Electrodacus SBMS0. I would like to mainly turn off inverter via the SBMS0 when inverter is inadvertantly left on, and still have plenty of power left in batteries to run essentials, lights, diesel heater. I have never played with arduino before, but willing to learn
 
It's a lot like electronic leggo these days. Programming an arduino is fairly easy. If you ever played with BASIC you have the skills to learn the simple C needed to write an arduino program. There's buckets of demo code to look at, and endless projects for free on the internet that you can download and look at too. A point of sale guy I know that had zero programming experience managed to google search his way through setting up a simple program to run an LED display board. There are endless 'modules' for them too. You name it, you can get it. Motor driver boards, relay boards, all prebuilt just wire up to the arduino board and start tinkering.

You may well find that controlling the Giandel is already within the SBMS' grasp. All that has to be done is monitor the power LED and pull the button line low to toggle it to suit, timings as per my previous post. I'm sure Dacian would be happy to advise on settings to make in the SBMS if it can do it without an arduino to intervene.
 
I bought the 4kw WZRELB. $460 on ebay shipped to me.

1) I realized my victron multiplus compact 24/2000 is a low frequency unit. Even though it is rated at 2kw and 4kw surge. My opinion is that Victron underrated it. Neither unit will start my air conditioner. I started it from home power, then turned on the WZRELB. I now have 2 transfer switches in series. The second thinks my WZRELB is a generator so it waits many seconds before it switches it in, so I am sure it was up to 120v. As soon as the transfer switched in the WZRELB, the AC died. The WZRELB showed 70 volts on its AC meter. I have not tried that test with the Victron. I might. This is annoying because the locked amps of the 2 motors in my AC should require 7kw and the WZRELB should surge to 8kw, so you'd think that it is very close to starting the AC. I expected it to be able to take over a running AC.

I am torn up about this. On the one hand, I feel like getting a 4kw low frequency unit. They are about $650. On the other hand, why not get a $300 soft start thing for the AC? This would reduce the surge amps needed and allow faster blow fuses/breakers in the wiring, which would feel safer. The bigger unit would be a lot heavier and that would be a problem in my water bay of the RV where I am putting the batteries and inverter.

I compared my Victron to the WZRELB at starting my small table saw. Both spun it up slower than home power. It seems like the Victron started it a bit faster.

I turned on my water heater which should be 1440 watts, and a 1500watt space heater. The WZRELB showed 118 volts and powered those for several minutes before I got bored. I measured 150amps from the 24v battery which agrees with 3kw of power plus some waste.

2) I had to pick out some hot glue that holds the power switch in the WZRELB and was then able to push it out the front. I had no trouble soldering in series some wires. However, others have reported that the elctrodacus SBMS0 opto isolators (relays) fail to provide enough current to power them properly. Dacien says that the LED volt meter on the inverter is wired in series with the switch and recommends cutting the wire to it. Because I am not sure I will keep the WZRELB, I bought a 1amp relay that is used to trip the WZRELB. I measured a fluctuating 80mAmps through the switch, so I suspect it peaks over 150mAmps which is the SBMS0' opto isolator's max.

(Note: Dacien's instructions say to wire the switch in parallel, which seems idiotic to me. If the switch is turned on, then the SBMS0 cannot turn the unit off. So a parallel switch is useless, but could be mistakenly put on and kill the batteries. If in series, all makes sense. His argument was that it was easier to solder 2 leads onto the switch. So what, do it right.)

I made a little board that has screw connectors for all the signal wires. I added the relay to it. This board is basically an easy connector spot for: 2 remote switches (for the DC/DC converter and the inverter), 4 signals for the 2 loads and 1 charger and PV from the SBMS0, and 4 signals going to same loads.

The SBMS0 is very annoying in that the EXT IO3/4 are little push in connectors, but EXT IO5/6 require you to solder in a connector to the main board (included) but it uses a ribbon cable. Dacien needs to clean this up and make all outputs available via the same port. Unless I put 3 more relays on my little circuit board, I need all 4 EXT IOs because all 4 charge/loads are different electrical characteristics.

3) I measured 1amp at idle for the inverter (24watts), which seems reasonable to me.

4) The "smart" fan is not the smartest. It is either on or off. When on, it is somewhat loud. It was going on and off with a period of many seconds. I mounted it in the water bay, which is below the main floor of the RV, so I think it will be reasonably quiet there. The Victron had essentially the same sized fans and thus was equally noisy when under load. I don't know the efficiency comparison between the two, or between an LF vs HF. So I assume that a LF unit, with the bigger case could have a bigger fan that could run slower and if the LF unit is more efficient, then it does not have to dissipate as much heat.

In summary, I think I will stick with the WXRELB. I don't relish the idea of taking a chance on a bigger unit. The struggle of fitting it is not worth it and the odds of making a mistake with respect to the remote power switch capability is not worth it.

I could have kept the Victron, but I suspect I will be able to sell it for at least what I paid for the WZRELB. I feel like the WZRELB should handle more resistive loads.
For inverter ac unit starting with minimal extra headroom for surge -hard start devices are almost always needed.

Do you have any type of hard start device on your primary ac unit? The best and more expensive systems include delay logic to have a lighter electrical load at start. This is one of the most effective delay start systems, but also more expensive.
https://www.microair.net/

Simpler, less cost hard start capacitor kit here:
https://www.supco.com/web/supco_live/products/SPP6.html
 
Late to the party as well, but this is a good thread. I'm also buying an Electrodacus and looking at controlling inverters with it. Much of the thread is on LF vs. HF. But back to the original question--did anyone ever summarize which of the options you researched could potentially be controlled by an Electrodacus BMS? I.e. by soldering in a control line or small relay, but without resorting to an Arduino for example. And with ability to separately disable charger and inverter, for inverter/charger options (only one I know of so far is Victron Multiplus).
Considering using electrodacus bms. However the inverter I planned to used giandel 24v 2000 wat has a momentary switch to it. The one electrodacus recommends is "slightly higher in price" at 3 times the price lol

What options do some of you guys have
I am going to use the electrodacus bms too.
...
I don't see the need for the fancier inverter/charger features where the shore power is seamlessly blended with the inverter. Since I will have LiFePo, I like the idea of getting a simpler charger that has no lead acid charging logic. I also see no need for any display or control (other than power on/off via BMS) on this lump of metal.

For the inverter, I also want simplicity. I see no reason to have anything displayed or controlled (except remote on/off via the BMS).
I have similar goals as above, I'd prefer to have the BMS making charging decisions based on its unique knowledge of the state of the battery including individual cells. With charger and inverter being fairly dumb remote controlled devices.

And then on the Arduino option:
It's a lot like electronic leggo these days. Programming an arduino is fairly easy. If you ever played with BASIC you have the skills to learn the simple C needed to write an arduino program. There's buckets of demo code to look at, and endless projects for free on the internet that you can download and look at too.
Could you provide a bit more info on what specific Arduino you used? I'm interested in trying it out, I have some basic C programming experience. I can google and pick something but why not leverage your experience? :)
 
I used a UNO because I have a few of them already but any of these sorts of things is more than enough. An attiny would be just as capable of it.

Code:
//
// read through this carefully to make sure it really works
//
#include <avr/wdt.h>

const int SBMS = 12;
const int POWER_LED = 11;
const int POWER_BUTTON = 10;

void setup() {
  wdt_enable(WDTO_8S); // enable hardware watchdog, 8 sec timeout

  pinMode (SBMS, INPUT); // connected to the SBMS, assume low = on, high = off
  pinMode (POWER_LED, INPUT); // giandel on LED, low = on (optocoupler pulls to ground), high = off

  digitalWrite (POWER_BUTTON, HIGH ); // giandel button, low = press button
                           // might seem odd to write before setting output mode, but
                           // we can set the data out register at any time.  we set it before
                           // configuring pin for output to avoid the pin pulling
                           // in a direction we don't want when we put it into output mode
                        
  pinMode (POWER_BUTTON, OUTPUT);
}

void loop() {
  if ( digitalRead ( SBMS ) ) { // high state, inverter desired off
    if ( ! digitalRead ( POWER_LED ) ) { // inverter is on, so try to turn it off
      digitalWrite ( POWER_BUTTON, LOW); // press button
      delay ( 1000 );
      digitalWrite ( POWER_BUTTON, HIGH);
    }
  } else { // low state, inverter desired on
    if ( digitalRead ( POWER_LED) ) { // inverter is off, so try to turn it on
      digitalWrite ( POWER_BUTTON, LOW );
      delay ( 2000 );
      digitalWrite ( POWER_BUTTON, HIGH );
    }
  }
  wdt_reset(); // reset the watchdog timer
  delay ( 1000 );
}
 
Last edited:
Considering using electrodacus bms. However the inverter I planned to used giandel 24v 2000 wat has a momentary switch to it.
I have come across an infinite cycle relay with a "once of" feature that may trigger the momentary switch. I think that it could be used to turn off the inverter, though it should not restart it. There are 5v, 12v and 24v versions, surely one should work.


My thinking is it is only required when the inverter has been accidently left on and battery is accidently depleted, not as a way to turn it back on. That could be done manually.

This relay could also be used to turn on/off heat pads when temperature falls below the safe charging temperature i.e. 5minutes on, 10 minutes off, to let temperature slowly warm batteries. Lots of heat pads about, probably need a seperate thread on these though.

What are peoples thoughts, hope the link above works
 
Infinite cycle says that it sits there pulling in for a period, then release for a period, repeat until heat death of universe or contacts wear out, which ever comes first.

To turn on from trigger being applied you really want a oneshot style operation with a timeout of say 2 seconds. Apply trigger to setup, relay pulls in and 'pushes' button, 2 seconds later, relay releases and inverter should be on.
 
The momentary switch needs 2-3 seconds to turn it on, but only a fraction of a second to turn it off. If the timer was set to around .5 second, then off indefinitely it should work. Below is the once mode feature

Once Mode(Jumper OP3):
1. The module Power-on, When triggered the relay is switched on, delay T1, and the relay disconnected.
2. When triggered on the delay process, again triggered, it will immediately interrupt the delay.

I am thinking that an activation of the trigger, the relay could send a current to the "once mode" relay, that would give a .5 sec current to turn off the momentary switch. I have no idea what power consumption would be with the activation of the trigger once inverter was turned off. I am just thinking out loud

I did see a Youtube video by a Danny Jensen, he may be a member on here, that highlighted the momentary switch function at the 1m 45s mark. I did email him and asked him to keep working on a solution, but no reply

 
I did see a Youtube video by a Danny Jensen, he may be a member on here, that highlighted the momentary switch function at the 1m 45s mark. I did email him and asked him to keep working on a solution, but no reply
You already have a cheap solution, an Arduino board and an optocoupler. $5. Stick a simple program on it, I even wrote one (untested) as a guide and posted it.
 
  • Like
Reactions: Dzl
I bought the 4kw WZRELB. $460 on ebay shipped to me.

1) I realized my victron multiplus compact 24/2000 is a low frequency unit. Even though it is rated at 2kw and 4kw surge. My opinion is that Victron underrated it. Neither unit will start my air conditioner. I started it from home power, then turned on the WZRELB. I now have 2 transfer switches in series. The second thinks my WZRELB is a generator so it waits many seconds before it switches it in, so I am sure it was up to 120v. As soon as the transfer switched in the WZRELB, the AC died. The WZRELB showed 70 volts on its AC meter. I have not tried that test with the Victron. I might. This is annoying because the locked amps of the 2 motors in my AC should require 7kw and the WZRELB should surge to 8kw, so you'd think that it is very close to starting the AC. I expected it to be able to take over a running AC.

I am torn up about this. On the one hand, I feel like getting a 4kw low frequency unit. They are about $650. On the other hand, why not get a $300 soft start thing for the AC? This would reduce the surge amps needed and allow faster blow fuses/breakers in the wiring, which would feel safer. The bigger unit would be a lot heavier and that would be a problem in my water bay of the RV where I am putting the batteries and inverter.

I compared my Victron to the WZRELB at starting my small table saw. Both spun it up slower than home power. It seems like the Victron started it a bit faster.

I turned on my water heater which should be 1440 watts, and a 1500watt space heater. The WZRELB showed 118 volts and powered those for several minutes before I got bored. I measured 150amps from the 24v battery which agrees with 3kw of power plus some waste.

2) I had to pick out some hot glue that holds the power switch in the WZRELB and was then able to push it out the front. I had no trouble soldering in series some wires. However, others have reported that the elctrodacus SBMS0 opto isolators (relays) fail to provide enough current to power them properly. Dacien says that the LED volt meter on the inverter is wired in series with the switch and recommends cutting the wire to it. Because I am not sure I will keep the WZRELB, I bought a 1amp relay that is used to trip the WZRELB. I measured a fluctuating 80mAmps through the switch, so I suspect it peaks over 150mAmps which is the SBMS0' opto isolator's max.

(Note: Dacien's instructions say to wire the switch in parallel, which seems idiotic to me. If the switch is turned on, then the SBMS0 cannot turn the unit off. So a parallel switch is useless, but could be mistakenly put on and kill the batteries. If in series, all makes sense. His argument was that it was easier to solder 2 leads onto the switch. So what, do it right.)

I made a little board that has screw connectors for all the signal wires. I added the relay to it. This board is basically an easy connector spot for: 2 remote switches (for the DC/DC converter and the inverter), 4 signals for the 2 loads and 1 charger and PV from the SBMS0, and 4 signals going to same loads.

The SBMS0 is very annoying in that the EXT IO3/4 are little push in connectors, but EXT IO5/6 require you to solder in a connector to the main board (included) but it uses a ribbon cable. Dacien needs to clean this up and make all outputs available via the same port. Unless I put 3 more relays on my little circuit board, I need all 4 EXT IOs because all 4 charge/loads are different electrical characteristics.

3) I measured 1amp at idle for the inverter (24watts), which seems reasonable to me.

4) The "smart" fan is not the smartest. It is either on or off. When on, it is somewhat loud. It was going on and off with a period of many seconds. I mounted it in the water bay, which is below the main floor of the RV, so I think it will be reasonably quiet there. The Victron had essentially the same sized fans and thus was equally noisy when under load. I don't know the efficiency comparison between the two, or between an LF vs HF. So I assume that a LF unit, with the bigger case could have a bigger fan that could run slower and if the LF unit is more efficient, then it does not have to dissipate as much heat.

In summary, I think I will stick with the WXRELB. I don't relish the idea of taking a chance on a bigger unit. The struggle of fitting it is not worth it and the odds of making a mistake with respect to the remote power switch capability is not worth it.

I could have kept the Victron, but I suspect I will be able to sell it for at least what I paid for the WZRELB. I feel like the WZRELB should handle more resistive loads.
I would be interested in buying your Victron if you still have it.
 
The multiplus is different from the compact. I can't remember what detail the compact is missing, but it won't work.
The Multiplus Compact does not have the Auxiliary inputs. However if you take a look at the resource I pointed to, there is a (untested) design for a circuit that might work. The trick uses the remote IO pins to control the discharge and the assistant that repossesses the temp sensor pins to control the charge.
 
  • Like
Reactions: Dzl
The Multiplus Compact does not have the Auxiliary inputs. However if you take a look at the resource I pointed to, there is a (untested) design for a circuit that might work. The trick uses the remote IO pins to control the discharge and the assistant that repossesses the temp sensor pins to control the charge.

Just wanted to check in and see if anyone ever ended up testing this.
 
I have not. Sorry.

All good, just thought I'd ask. The Multiplus Compact recently jumped onto my radar as it has a very low no-load idle draw for such a capable transformer based inverter. ~10W, compared to ~20W for its big brother the 3kw multiplus, and 25-60W for most other brands of small to medium sized LF inverter.
 
BTW: I have played with the 2-signal BMS assistant on the Multiplus Compact. I can confirm that the assistant will let you use the Temp Sensor input to either control the charge on-off or the discharge on-off. It just won't do both like you can on the regular MultiPlus.

I have not set up the relay to try to control the charge using the remote control inputs, but I see no reason it would not work.
 
  • Like
Reactions: Dzl
What about the SUNGOLDPOWER 3000W Peak 9000W Pure Sine Wave Power Inverter DC 24V AC 120V with Battery AC Charger Solar Wind Power Inverters LCD Display Low Frequency Solar Converter? It is a low frequency converter and seems to be a good price. The problem that I'm having with inverter/charger is that Electrodacus SBMS0 needs EX IOx wires run both to inverter and charge. Sun Gold Power only has one remote port that is for inverter and nothing for charge. SBMS0 needs to be able to turn off charger once cell hits max charge. Have not figure a way to do that
 
I referring to being connected to shore power so that inverter/charger can top batteries off
 
Back
Top