Author Topic: feeding a grid tie inverter with rectified single phase?  (Read 6651 times)

0 Members and 1 Guest are viewing this topic.

mab

  • Sr. Member
  • ****
  • Posts: 428
  • Country: wales
feeding a grid tie inverter with rectified single phase?
« on: April 18, 2016, 05:05:49 PM »
I'm planning on using a GTI to 'ac couple' my hydro-generator (to my powerjack based 'grid'), and the plan is to convert the generator output to 3 phase and rectify, but that means my current single phase to battery system (and dump loads) won't work. Rather than convert it all to 3ph only to find the GTI doesn't work well, then having to convert it all back, I was wondering if the GTI would play nice with rectified 1ph as a way of testing the GTI setup?

the GTI i've got is a sunny boy 1200  - it's actually a solar GTI but in theory a solar MPPT algorithm should find the MPP of the hydro OK - but I'm not sure without trying. Also the generator output is ~33Hz which probably wont help.

Possibly a daft idea but though it worth asking rather than dismissing outright.

oztules

  • Hero Member
  • *****
  • Posts: 1477
  • Country: aq
  • Village idiot
Re: feeding a grid tie inverter with rectified single phase?
« Reply #1 on: April 18, 2016, 05:42:45 PM »
The GTI should take any source that has a reasonably high impedance.

For example, I rewired one of mine with 50uf capacitor and a bridge rectifier in the front end for the local power station.

This enabled them to just plug it in, and it would try to backfeed@ 300 watts..... if in an 8hr period they could drive 2.4kwh into their grid, then the freq was in spec, if less then there would be a freq error or two... or 50.... This is a galvanically isolated one... not a hf style

This proved that the unit will happily play with a source it can exhaust or modify with it's mppt. ... yes they had a nett loss in the system, but it proved a few things for them regarding their frequency stability in real time.... versus their claims that they had sorted it. All the grid ties on the island would drop off for long periods........

So yes, it should be quite happy with the rectified input from my experiences with this sort of thing.


..................oztules



Flinders Island Australia

mab

  • Sr. Member
  • ****
  • Posts: 428
  • Country: wales
Re: feeding a grid tie inverter with rectified single phase?
« Reply #2 on: April 18, 2016, 05:59:56 PM »
Thanks Oz,

the generator ATM is a capacitor exited induction motor and whilst it may not be very high impedance, pulling too much current does make the voltage fall off, which should look like impedance to the GTI - in theory.

based on your setup maybe I should add 50-100uF as a starting point just to reduce the ripple a bit.

oztules

  • Hero Member
  • *****
  • Posts: 1477
  • Country: aq
  • Village idiot
Re: feeding a grid tie inverter with rectified single phase?
« Reply #3 on: April 18, 2016, 10:30:12 PM »
in that case... probably don't need the caps... they would have been series.

The front end of the grid tie will have ample smoothing caps I suspect.


..........oztules
Flinders Island Australia

mab

  • Sr. Member
  • ****
  • Posts: 428
  • Country: wales
Re: feeding a grid tie inverter with rectified single phase?
« Reply #4 on: April 19, 2016, 01:02:10 PM »
Oh I see: the caps were to limit it to 300w.

Well I'm trying it now  (without adding caps) and am pleasantly surprised at how well the powerjack, sunny boy and hydro are playing together  :) , especially as I'd forgotten that I've ground-referenced the hydro output and the first thing the sb did was tell me that there's a ground fault. Nonetheless, after the interminable wait for it to decide the grid is stable, it cut in and is feeding the hydropower into my grid  :D .

It's interesting to see how the mppt function works with the hydro; it's quite tentative in making changes, which is good as i had feared it would crank up the amps too fast and shut down the generator. seems to wander about a point at slightly higher amps and lower volts than i would have chosen, but perhaps that's because it's expecting the sharper knee of a solar panel.

Thanks again Oztules, I'm another step closer to getting my system sorted (although, I may have to do a transformer upgrade in the power Jack once I'm backfeeding 24/7).

oztules

  • Hero Member
  • *****
  • Posts: 1477
  • Country: aq
  • Village idiot
Re: feeding a grid tie inverter with rectified single phase?
« Reply #5 on: April 19, 2016, 05:46:48 PM »
Looking at your system.... you have a grid tie inverter there before you introduce the SMA.

The induction motor is more than adequate to drive the PJ directly and co-generate.... and you won't even need the capacitors at the motor.

Remember it is an asynchronous motor, so not rpm dependent for the frequency. If you direct connect it to the pj, ( with cap start or run  caps) the pj will make it a motor, if you turn on the water, as the rpm goes through 1500 or 3000 rpm ( pole dependent ), it will start to generate back into your grid. It will increase power until the water can no longer support any more rpm increase.... auto mppt.

This is good for a petrol motor as backup too, this will allow the pj to act as charger, without the use of the pj trying to couple on the input end.

Should be more efficient that way, and you can sell the SMA and get more panels or battery.

This setup is how all the industrial windmills were before they went to permanent magnet designs too.

............oztules
Flinders Island Australia

mab

  • Sr. Member
  • ****
  • Posts: 428
  • Country: wales
Re: feeding a grid tie inverter with rectified single phase?
« Reply #6 on: April 19, 2016, 06:17:54 PM »
well that's an interesting idea - I'd need a motor with more poles (4 ATM - that's why it's only running 33Hz/130v would probably work best with 8 poles which would get the turbine to approx the right RPM), and presumably a single phase motor too. But definitely an interesting idea...

oztules

  • Hero Member
  • *****
  • Posts: 1477
  • Country: aq
  • Village idiot
Re: feeding a grid tie inverter with rectified single phase?
« Reply #7 on: April 19, 2016, 08:41:36 PM »
industrial 3ph motor would be my choice..... use one phase only.

...........oztules
Flinders Island Australia

Eraser3000

  • Newbie
  • *
  • Posts: 36
  • Country: us
Re: feeding a grid tie inverter with rectified single phase?
« Reply #8 on: April 26, 2016, 10:55:34 PM »
You can program the SMA to not wait as long before kicking in.  Down to 5 or 10 seconds I believe.

Mine are set to 10 seconds.

mab

  • Sr. Member
  • ****
  • Posts: 428
  • Country: wales
Re: feeding a grid tie inverter with rectified single phase?
« Reply #9 on: April 27, 2016, 03:01:25 PM »
Thanks for that idea.

I've been looking into that - trouble is I don't have a means of communicating with the SMA - there's no RS232 or bluetooth card that I can see - and the add-on cards I've seen for sale cost more than I paid for the inverter, so I'm looking out for a 2nd hand one...

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: feeding a grid tie inverter with rectified single phase?
« Reply #10 on: April 28, 2016, 01:16:10 AM »
I don't see a point in raising the pole count because the higher pole count induction motors have poorer/lower power factor and will be harder to self excite. (more copper losses)

If you can get a 3 phase motor working properly then it should be able to self excite and a  3 phase diode bridge would give you 3 times the ripple frequency after the rectifier.
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

mab

  • Sr. Member
  • ****
  • Posts: 428
  • Country: wales
Re: feeding a grid tie inverter with rectified single phase?
« Reply #11 on: April 28, 2016, 07:27:35 AM »
Hi Joestue , I was thinking about doing what oztules was suggesting where I think you're essentially running the motor off the inverter and pushing it faster - so not exciting with caps - unless I've misunderstood.

Actually, I'd forgotten I'm  using a smaller runner now so it might work with the existing 4 pole - although it may run a bit fast for efficiency. But now I'm wondering what would happen if the inverter shut down whilst being backfed by the motor.

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: feeding a grid tie inverter with rectified single phase?
« Reply #12 on: April 28, 2016, 08:45:35 PM »
you could do it either way, its just a general rule that high pole count induction motors are lower power factor.

if the inverter shut down, initially there might be some current continue to flow into the system, particularly if other capacitive loads are hooked up, but basically nothing will happen. the inverter can't be damaged because even when the inverter is off, any current sent into the transformer will just charge the batteries, and you have to supply the transformer more than 120/240vac to do that.

i have a very cheap bench grinder, it will self excite through the run capacitor and the lightbulb, the 15watt cfl stays on for about half a second then dims quickly after i turn the grinder off. i suspect something similar would happen in your case.
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

mab

  • Sr. Member
  • ****
  • Posts: 428
  • Country: wales
Re: feeding a grid tie inverter with rectified single phase?
« Reply #13 on: April 30, 2016, 06:10:14 PM »
OK, that's reassuring, I might give it a try then. In theory, if I'm aiming for 1540ish rpm (1420 at full load as a motor) my runner is ~10% too large - so I'm not sure how much that will affect my efficiency.

I still have questions for the experts on this sort of thing:

As it's currently 3ph delta with C-2C capacitor excitation (50/100uF), I'm wondering how much capacitance I'd need to make it spin-up as a motor off single phase before turning on the water?

Or should I lose all the caps, spin it up with the water then connect to 230v?

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: feeding a grid tie inverter with rectified single phase?
« Reply #14 on: April 30, 2016, 06:34:02 PM »
you can spin the motor up, then connect it to the inverter. either connecting the inverter to a stopped motor, or connecting the inverter to an over sped hydro, will result in close to locked rotor amps in either case.

Hot plugging the hydro motor in with capacitors attached is a problem. the peak currents can reach insane levels and will either trip the inverter or potentially cause a high voltage spike when the inverter trips off.

Capacitors at the motor can cancel out the inductive reactance of the motor, reducing the line amps between the motor and the inverter, thus reducing copper losses in your line. how long is it?

Adding capacitors to the other phases can reduce the reactance of the motor as well and offset internally the current needed to excite the motor.. this will also benefit your system. if they are too big it will increase the copper losses internal to the motor.

What I would suggest is you buy a watt meter, either a digital one or analog, and select the capacitors to push maximum watts into your inverter after you get it running.

Then put the disconnect to the motor after the capacitors.. leave the capacitors in circuit at all times, connected to the inverter.

Another option is an NTC resistor, you can find one in every computer power supply. install one of them for each capacitor. then you can hot plug the capacitor.

of course if the line between your inverter and your hydro is long enough, then you can hot plug the capacitors, because the line resistance and inductance will limit the peak currents involved.

but i would not attempt to just plug in a 100 uf cap right onto your 240vac inverter. might work.. but it might trip the inverter.
« Last Edit: April 30, 2016, 06:42:48 PM by joestue »
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

mab

  • Sr. Member
  • ****
  • Posts: 428
  • Country: wales
Re: feeding a grid tie inverter with rectified single phase?
« Reply #15 on: April 30, 2016, 07:24:24 PM »
hmm, well the line from the hydro is about 110m and fairly thin cable so resistance is significant.

The inverter can handle the start current for the motor (particularly, as you say, with the long line) but if it is to start the 3ph motor on 1ph, there needs to be a bit of capacitance or the motor will just sit there humming. I wonder what the minimum I can get away with is.

I do have an analogue wattmeter, although as the wattmeter needs to be at the inverter end and the caps at the hydro end it could mean a lot of running up and down to make adjustments.

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: feeding a grid tie inverter with rectified single phase?
« Reply #16 on: May 01, 2016, 07:42:21 PM »
How is the single phase motor starting now?

With that long run of cable I wouldn't worry about hot plugging capacitors, even if they are relatively large.

usually the starting uf for 3 phase motors is 100uf per hp at 230v nominal mains iirc.
100uf at 230volts is about 10 amps of current.

Typically half hp single phase motors have a 150-200 uf start cap, which would be equivalent to 37-50uf if it were 240vac.

My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

mab

  • Sr. Member
  • ****
  • Posts: 428
  • Country: wales
Re: feeding a grid tie inverter with rectified single phase?
« Reply #17 on: May 02, 2016, 06:54:21 PM »
I've not actually tried it yet - not enough time to play, and some reluctance to take the hydro off-line whilst I rearrange the wiring.
It's a 2hp motor wired delta - so I guess I could see if it starts with 50 or 100uf and take it from there.

What I could really do with is a power factor / watt meter so I could adjust the capacitance until the PF of the power going up the wire to the inverter approaches 1 . Aha - found one online so I might wait 'til I've got one here.