question

kali-the-miller avatar image

Strong A&V Fluctuation of MPPT for Hydro

The "4min" pictures show the effect. In this case we came from Float. The battery is full, no current is flowing. Everything is fine. Then at the load output of the MPPT a light bulb is switched on - about 2 A the bulb wants to pull at 12.75 V = 25.5 W. Makes sense. But the load from the generator starts to swing now from zero to 150 W.4min Solar (W) and Load (A) 20190730-093526.jpeg

The charge controller seemingly can't cope with controlling that, because when it tries to take the voltage back up, it brakes the water wheel, voltage drops and current goes down simultaneously.4min Solar (V) and Batt (V) 20190730-093613.jpeg 4min Solar (A) and Batt (A) 20190730-093635.jpeg

The pictures "42min" show the course over a longer period. Sometimes it goes smoothly for a few minutes, but then everything oscillates again. 42min Solar (W) and Load (A) 20190730-112209.jpeg42min Solar (V) and Batt (V) 20190730-112228.jpeg42min Solar (A) and Batt (A) 20190730-112249.jpeg

The currents jump up to 10A and down to 0A. You can also hear it in the running noise of the system, how the generator grabs and brakes and then lets go again. Definitely not good electrically for the battery and mechanically?

Is there a trick for hydropower units? From my point of view, the charge controller is too fast? If it were a PID controller, you could tune it. But I didn't find any parameters that could be tuned.

We are at a water wheel with a weight of approximately 1 ton (3.5 m diameter, 1.5 m wide); with the water flow and fall height, theoretical power 100 W, rotating at ~ 10 rpm IMG_3530.jpg-> 3 stage gear system to increase rpm IMG_20180323_181324.jpg-> 3-phase AC generator (ZXA 50500) -> bridge rectifier -> BlueSolar MPPT 75/10 -> 100 Ah 12 V battery

Hydro and Wind Power
2 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

3 Answers
tilo avatar image
tilo answered ·

You are probably right with the "resonance" issue. I think that might be answered only by someone (from Victron) who knows how the MPPT algorithm is implemented. But as Markus said, it is a solar controller. and if it does not work well in this non-solar application it is not a Victron "support case". It would be nice if Victron could add a hydro mode, as it is just a software issue.

Just guessing, but I think adding a capacitor on the MPPT input will not improve the situation, have you tried it?

You could try the other approach often used in small hydro installations, which is running the generator always at the maximum power, and burn the excess energy after the output, in your case after the MPPT. If you are familiar with PID regulators, you might implement a variable dump load, and use the battery voltage as the control input variable. That means you reduce the functionality of the MPPT to just optimizing the power conversion, and pass the task of maintaining the battery voltage (float, and sometimes absorption) to your dump load controller. For this you have to raise the float voltage of the MPPT higher, so that it will stay always in MPPT mode. In case you don´t want to implement a dump load controller yourself, you can buy one, for example a Phocos MPS 45 or 80, and add a 12V dump load.

Share
2 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

kali-the-miller avatar image
kali-the-miller answered ·

or do I have the wrong type charge controller?

BTW: Victron - the chart heading says "Spannungn (V)" - note the typo for next update

1 comment Share
2 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Victron MPPT chargers are solar chargers, designed to be used with solar panels only. But under certain conditions - out of support - they can work with a hydro water wheel. I use Victron MPPT chargers with my hydro turbine for 2 years time now, and it perfectly works for me.

I did some more thinking, and it is probably not easy to stall a wheel with a ton of weight. ;o)

Maybe it is a belt slipping?!

But it is hard, to diagnose such a system remotely.

wkirby avatar image
wkirby answered ·

The trick is to limit the maximum MPPT current down from 10A to something lower. This will help to prevent the MPPT from overloading the wheel.

If your wheel has a maximum power of 100W, then set the "Max charge current" setting to 7A and watch how the wheel responds. Tune up or down until it works how you want it.

Do you have provision to prevent over Voltage (more than 75V) going into the MPPT when the generator is unloaded?

1 comment Share
2 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Hey - thanks guys for responding!

I tried the limitation of the current to battery already as this was the only lever to pull - no avail - still oscillating like crazy, just a little lower amplitude. I don't want to limit it down to something so that the generator capability is not used. I send a log chart of that too when I have repeated this.

I think it is actually not overloading the wheel. The wheel is not stalling of course. There is a ton of material rotating. You can not stop that so easy. The issue is that the system is somehow in resonance. Whenever the MPPT makes the generator grip the gearbox it reduces the power output and voltage which in turn makes the MPPT loosen the grip and that in circles.

This can not be that way. We are talking a simple set-up. A battery to charge and a light bulb as load.

I was thinking to install a capacitor in the system. Would that help? Can't be that way that the current swings form zero to max all the time within seconds. The charts show 16 peaks in 4v minutes - means every about 15 seconds the current goes from zero to max amperes. Any electrical wizard out there who can tell what additional element is needed is the system to get the Oszillation out or how to tune the system?

There will be no over-voltage. The voltage at no load is like 24V - when there is a lot of water may be 27V - if it is a flood may be 30V - but that's it. You can see from the 4min chart with the (V)'es - no load is on the left.

As for Markus answer: I will measure a few voltages with different loads only. Will report back.