TheBackShed.com

Forum Index : Solar : DIY Synchronous PV Charge Controller

Author

Message

Solar Mike
Guru

Joined: 08/02/2015
Location: New Zealand
Posts: 1123

Posted: 10:35am 19 Dec 2023

Copy link to clipboard

Print this post

I'm currently building\installing a pumping control system using 2 x 12v series lead acid batteries that power a 300W 24v DC water pump. The setup has 3 x 180W PV panels paralleled, have had these panels for many years unused sitting under the house, was going to use a simple PWM controller, however the panel spec's have a high mppt voltage of 36V @5 amps, so a lot of power will be wasted, a fully synchronous controller will be more efficient.

Have made a few of these in the past, but its always nice to improve upon a previous design, so this is V1 of the idea. The design automatically senses the charge current and when it gets very low, it changes mode to asynchronous operation, turning off the synchronous mosfet and using its body diode as the buck rectifier. This prevents any possibility of the active rectifier mosfet acting as a boost converter at lower duty cycles and blowing up the buck components. This is transparent to the CPU, so I can use slower CPU software - Picaxe Basic- 20X2 running at 64 Mhz; PWM is 10bit 64Khz. Have most of the software written and it seems to be working pretty well.

Am using 2 mosfets for the PV-Battery isolation switch, Buck stage and active rectifier, so 6 in all; the mosfet specs - bulk electro's determine the max input voltage and current etc. As it stands 40 amps charge current would be the limit using 1oz pcb. The PV setup for the pumps could generate 20 odd amps, have being doing testing with 30 or more amps for 1/2 an hour and the small heatsink gets mildly warm, haven't measured the actual efficiency yet.

Have gone away from building very high current controllers, its makes more sense to split the PV up into smaller arrays each with their own controller.

I want to use this board for some other projects using 200V PV and 100V Lifepo4 batteries, so track spacing 1.6mm would handle the higher voltages.

The mosfet driver circuitry has been placed on a small sub pcb that connects using a 10pin IDC ribbon connector, the pcb sits atop the mosfet leg pins, soldered to the legs protruding through the main board. This arrangement allows the high mosfet currents be isolated away from drive signals that now go direct to the mosfet pins, seems to work very well.

Here are some photo's, will add more as testing progresses.

Main PCB 162 x 173mm:

Driver:

Assembled:

Cheers
Mike

nickskethisniks
Guru

Joined: 17/10/2017
Location: Belgium
Posts: 411

Posted: 12:20pm 20 Dec 2023

Copy link to clipboard

Print this post

That's a real beauty Mike, great job!

Solar Mike
Guru

Joined: 08/02/2015
Location: New Zealand
Posts: 1123

Posted: 01:47am 29 Dec 2023

Copy link to clipboard

Print this post

There is a wiring error on the pcb, PWM signal was going to a pin, that on the X2 version of the chip doesn't have a PWM output; had used the 20M2 in the layout rather than the 20X2. So had to cut a few tracks and run some jumpers. - What happens when you do all this in your head with no circuit dia.

Other than that issue, the main PCB and driver work well; I don't have a lower voltage power supply with high enough current rating to bench test at 40 amps and as I only fitted 80 volt mosfets for the initial test, cannot test it with higher voltages.

However Vers.2 should fix those issues, have fixed the wiring issue, changed the 0 volt earth plane so has more area and replaced the 50 amp ACS770 current sensor with a smaller 40 amp ACS723-40U. Would rather use multiple smaller controllers with their own PV array than a larger unit; much easier to make chokes and less stress on main electro's etc with lower currents. Also altered the driver pcb by removing redundant mosfet connections.
40 amps should be no problem using 1oz copper.

The PCB's arrived today, very quick turn around from JLCPCB considering the Xmas rush. Will build this up with some decent mosfets and see how it goes.

Main PCB V2:

Some of the white overlay lines looks squiggly, its an artifact of the 3D view only.

Driver V2:

Need an in-line current probe for my scope so designed this small pcb, has two boards, will fit with a 50A or 100A sensors, fit inside a small plastic box with 9v battery.

Cheers
Mike
Edited 2023-12-29 11:49 by Solar Mike

Print this page

To reply to this topic, you need to log in.