diy solar

diy solar

Ammeters and diodes

Yeah right. You try it then. Show me a load meter that doesn't go backwards when charging without a diode

Good grief. It has already been stated many times that there will be no reverse current if you put the meters on the separate load and charge lines.
 
So all I have to do is to get another charge controller and run separate wires to the battery.
Or use a diode. I think I'll try the diode first. I did a few tests on a parallel shunt... it doesn't look good.
 
So all I have to do is to get another charge controller and run separate wires to the battery.
Or use a diode.

Or just do what everyone has been saying for the past 40 posts or so. No matter what type of system you have, there has to be a point where the load line separates from the charge line
 
Currents on charger branch and load branch have only one sense, it no need center zero ammeter, on battery branch current is bidirectional (charge discharge) it need a center zero ammeter, if SOC/voltage of battery is well defined it need only a voltmeter to asses SOC, but for analog, the voltage variation is small to be noticed, but you can use zener diodes and resistors to define the scale bottom as empty and scale top as full.
At zero SOC assigned voltage the needle sits at zero and at Full Soc assigned voltage the needle sits at top, this is good as backup of digital SOC meter that is proned to fail more frequently.
 
Construction hit a snag, so I had a few free hours. I did a quick test.
The diode works perfectly. It does get a bit warm, with a 4A load, I'll just find something with the least possible forward drop and put a heat sink on it - when I move the batteries to the house. It's worth a few watts - to me. :·)
 
I'm gonna call bs without seeing the whole setup or a schematic of your exact wiring.
 
Do I really have to draw you a picture?
arrabbiato1.gif
wiring diagrams are useful when asking electrical questions. So, Yes, a picture. sorry i posted before reading on.
 
Last edited:
The diode works perfectly. It does get a bit warm, with a 4A load,
I really want to see just where this diode is. Since you said it was getting warm, sounds like it is in series with one of the lines and not just the meter. If so, you just blocked charging. If not, it was not needed in the first place like people have been telling you.
 
Your diagram. The charge meter will read the amperes going from the solar panels to the system. the discharge meter will read the amperes going from the battery and the solar charger to the loads. if the 'loads' include a charger/inverter the diode will block the charging current from going to the battery if the diode is hefty enough. so the discharge meter will have no back current to read. what effect blocking the charge current will do to the charger/inverter i don't know, except it won't charge the battery.
 
Yes, yes, you're (all) quite right. I wasn't thinking clearly.
It does block the charge. The thing is, it was late afternoon, the charge was little to start with, the PV ammeter was registering a few amps.
But they were going to the loads, not the battery, it was sort of a "bypass" mode :·)
I'll just get a separate charge controller and be done with it.
 
At all in one he has access only to charge current, (panels, generator, input) and battery current (battery port) he has no access to load current, some variants transfer power directly from generator to load via a 400vdc bus, and battery has a bidirectional insulated converter as PIP5048 for example, and others as well, he can not measure any current al lo voltage on load branch.

In such inverters battery acts as hi voltage virtual battery, he can measure generator current ,and load current on 400vdc bus at modulator`s input.
The battery node is on this 400vdc bus.

Using diodes on 24 and 48 v bus result in significant losses due to hi current, it needs large diodes with appropriate heat sinks.
 
Ok, excuse my notepad scribbles. This is the first drawing, simply by separating the load and charge as suggested is very easy and works without diodes.
 

Attachments

  • IMG_20211013_054702992.jpg
    IMG_20211013_054702992.jpg
    49.6 KB · Views: 5
If you cannot separate the load and charge, and need to use diodes it gets much more complicated. As stated before, you need an amplifier before the diode to overcome the forward voltage drop of the diode, then after the diode you need a resistor network to calibrate the meter for the increased voltage from the amplifier. Building the amplifier and selecting the resistors is beyond the scope of this drawing, as neither will be simple to do.
 

Attachments

  • IMG_20211013_054859285.jpg
    IMG_20211013_054859285.jpg
    320.9 KB · Views: 7
I like those panel meters. Could you show an updated circuit diagram?
 
You know, years ago, if you sent an email to not@thistimeonite.com... I got it :·)
It's late enough here (CET)... I just put the charge wire to the battery before (from the battery) the shunt to the load ammeter.
So the meter reads everything (inverter included) after that.

The charge ammeter is between the panels and the controller (which has diodes - mosfets - on its output to start with).
So no communication between the two.
 
Back
Top