Thanks for that explanantion Peter, I'm slowly learning how it all works.

I do now have three arrays and three mppt's connected. I will add more panels and the fourth mppt soon.

I certainly wasn't suspecting anything wrong with this third mppt  ...  I was just a little disappointed in the output from the 3kW of panels I put up on the weekend. I did try shutting off the first two to see what this would do on its own  ...  and I also swapped one of the chokes over as on this third one I fitted a couple extra turns than the earlier ones.

Anyway I PULLED A CLANGER  ...  just on dusk tonight I thought I'd check my combiner box to see if I'd made some sort of mistake there. Pulled all the fuses down  ...  and there were TWO FUSES MISSING!!! I hadn't even checked. I knew I was a couple short  ...  but I thought they were in the last box that hasn't been connected yet.

So if my output was only about 75% of what I was expecting  ...  THAT'S BECAUSE ONLY 75% WAS CONNECTED!!! Duh!!!

Man, and I had gone through a hundred scenarios as to what I should do  ...  buy a heap of better panels, pull down the ones I just put up and sell them  ...  haha, what a dill!

And to add insult to injury, here's a picture of the output from the first two mppt's going strong at one stage today, while the third was lagging behind. Had the camera at the wrong angle, so the third display didn't show. Not sure I should run it this hard  ...  but it was cloudy and only intermittant.

What's fascinating is that the first two lots of panels are old used 275 watt panels, two lots of 3.3kW  ...  and for short bursts they were pumping out just over that. I'm impressed. Great design, thanks man.  

Looks very nice!

What version of the MPPT board are you using? It looks a bit different from the one I referenced / built - you've got the Nano(s) on board!

Cheers,
R

Thanks very much. It was one of Wiseguy's boards that (with his blessing and help) I added extra room on the bottom to fit both nano's and 12v and 5v supplies.

It was what I was working on way back here  ... in this thread.

Like most things I do it took me forever to get it done and built but finally there now. I have all four up and pumping well and charging a 48v 980Ah forklift battery. And that is running my Warpverter.  

Oh right, that one! Yes I remember that discussion. I'm glad it works nicely, good job!

R

Possibly spotted a very minor error in (i think) the latest version of the code.ZIP (serial LCD version):

TID=12027&P=47#212592latest version serial LCD

it seemed to do an extra MPPT scan every couple of mins at the 55sec mark ,but i spotted this in the code:



But as the Comment says "should be 60 for an hour" so is easy enough to change.

  mab1 said  

Possibly spotted a very minor error in (i think) the latest version of the code.ZIP (serial LCD version):

TID=12027&P=47#212592latest version serial LCD

it seemed to do an extra MPPT scan every couple of mins at the 55sec mark ,but i spotted this in the code:

if ( timeout_check(8,1))
   {
     cal_sensor_count ++;
     if (cal_sensor_count > 1)           // should be 60 for an hour. testing will have it at 2 or something
       {
         cal_sensor_count = 0;           // one hour is up time to recal current sensor zero points
         if (track_mode != NIGHT)
           {
             //stop it

But as the Comment says "should be 60 for an hour" so is easy enough to change.

it's an easy fix alright. Thanks for reading through the code and finding this.

latest version with the fix:

mpptv5_BV_tempco_day_totals_current_sense_recal.zip

Hi Peter is there a reason you choose for the software serial instead of the already present hardware serial interface on the arduino board?

I ask because I want to add a can interface to the mppt controllers, I need pins 10 and 11 free for the spi bus. The idea was to add it on the nano that does the uart/I²C conversion for the display. The hardware should not conflict the spi bus, I'm not sure about the serial library.

Do you think this would work, or am I missing something?

Thanks.
Edited 2024-05-12 06:42 by nickskethisniks

Is there any update on this for mpptv5_highside_ntc version/serial lcd?

I know this thread hasn't been updated for a while, but thought I might upload a picture of my set of MPPTs and to thank Poida for an amazing project  ...  as well as Nicks and Wiseguy.

If you add it up you can see that combined they have generated 21,862kWh of power since being installed in March last year  ...  so 21 months.

Normally the outputs of all four are connected together to charge a couple of 48v forklift batteries  ...  but just at present I'm refurbing a 3rd battery and the 4th mppt is charging that on its own.

The top two are each fed by 3.3kW of panels  ...  two lots of 12 x 275W panels 3s4p

...  and bottom two by 3kW each ...  two lots of 12 x 250W 3s4p

As you can see I flog them pretty hard and have set them all to max 60A output and have never had any problem doing that.

The only time I have had issues was early in the piece when I turned off the outputs before I turned off the input from the panels. Doesn't seem to like that.

When it happens it blows the little batt volts - 12 volt converter and also once an optocoupler. If I switch them off correctly, no problems.

I added a tvr protection device to the inputs and a 13v zener on the outputs which seems to limit any damage.

Anyway, thanks guys for an amazing project  ...  and anyone considering building one, it's well worth it.

Roger,
good to see. I had to respond when reading you run them up to 60A.
no doubt they can do it.
how are the chokes and heatsink temps when doing this? are they reasonable?

In my experience, these MPPT just do the job.
here are my 2, it's 7pm, cloudy in Melbourne.

Peter, wow yours have clocked up some miles too.

Surprisingly the heatsinks never seem to get warm  ...  maybe a few degrees above ambient.

The chokes can warm up fairly quickly but I have a wide, single barrel fan across the bottom which is set to come on at 40C  ...  and shut off at 35C. The fan seems to keep them under control and doesn't get much above 40. You can hear it cut on and off during the day and doesn't seem to run long. It can start fairly early in the charge cycle too  ...  but again, doesn't have to run for super long periods.

It probably helps being all open but it's in a corrugated iron shed which can get pretty hot at times.

About the only thing I would like to have is the ability to make minor adjustments without having to break out the laptop and try and remember how to talk to it. I'm not all that familiar with Arduinos. Thanks again.

I have used a samsung phone, with usb cable and special USB plug/thing called an OTG adaptor.



find and install "serial usb terminal"
config it to be
9600 baud, 8 bits, 1 stop, no parity
when sending, just "LF" as the newline terminator
choose to clear input on send

connect, send "?" and get the menu.


Edited 2025-11-30 15:08 by poida

Thanks Peter, I had no idea that was possible.

I have a Samsung phone so I downloaded "serial usb terminal" and managed to find an OTG connector in the drawer.

You wrote "when sending, just "LF" as the newline terminator" which I didn't understand  ...  but found when I clicked the funny little symbol at the top  ...  a bit like a waveform with a gap in the middle  ...  it connected and recognised a legit target.

I was able to see the menu and make changes easily  ...  awesome.

Now that I'm rereading your comment I understand why you suggest "choose to clear input on send"  ...  otherwise you have to manually delete the previous command.

Well that's going to make any changes so much easier, thank you very much.