[EE] Wind turbine power regulator

[Brent Brown]

Hi all,

Home project on the go, thoughts appreciated from the fine minds here...

I purchased a low-cost 300W wind turbine to play with. I have a 5kW grid-connected
solar PV system in place, with 20 x 250W PV panels & micro-inverters. What I'd like
to try is effectively connecting the wind turbine in parallel with 1 of the PV panels.
Reasoning is it makes use of inverters already there which are not being fully
utilised each day or even utilised at all each night ;-) And it puts the power
somewhere I can use it, and will use energy monitoring system already in place. If it
works out ok I could potentially add more wind turbine(s) in the future, one to each
inverter.

It's important to state that this PV system uses AC micro-inverters, and there is no
high voltage DC accessible. This not the case with other PV systems, which require
considerable additional electrical safety requirements. What is discussed here is not
safe for, nor applicable to, string connected PV systems.

If I rectify the 3-phase low voltage AC from the wind generator to DC, and if I
connect this and a solar panel to an inverter, and if I use a (Schottly) diode to avoid
reverse power flow into the PV panel, and if the voltage is compatible enough, then
it might just work. A few questions come up, but on the surface most seem to solve
themselves or be not too difficult to overcome...

- Solar inverter MPP regulation is not expected to be optimal when power source is
wind turbine. As long as it plays nicely and doesn't do anything bad I'm ok with that,
as I'm not aiming for maximum efficiency or anything. Am I hoping for too much
here?

- Inverter has the smarts to not exceed it's own power rating, i.e. it is not
un-common for a 300W or more PV panel to be connected to an inverter of lesser
power.

- Inverter has ground fault current detection which I need to be wary of, may be
longish cabling to wind turbine so perhaps some potential for nuiscance triggering.

- Inverter input mustn't exceed a certain maximum DC input voltage. No problem
from the PV panel as it's obviously designed for that, but wind turbine has no built in
voltage regulation. This is where I have to construct something.

I'm thinking a shunt regulator is the way to go with a wind turbine. Set at a voltage
above what the solar PV produces (30.8V @ Vmp), but safely below Vmax allowed
for inverter (45V operating, 55V Abs max), capable of dissipating 300W or so worst
case. Could dump power into a power resistor or incandescent lamp or heater
element, with say a PIC PWM'ing a MOSFET. Or even use multiple power
MOSFET's in linear mode... the heat has to go somewhere and could possibly work
out cheaper by not requiring expensive power resistor(s). Perhaps a fan on the
heatsink kicking in as required. Want to keep it simple enough that it remains worth
doing, but done well enough that I don't damage existing PV system or make it
unsafe.

The wind turbine is rated at 300W, 24VAC 3ph. With a 3-ph bridge rectifier plus
filter capacitor I'm expecting in the order of (24 x 1.41) - (2 x 1V) = 32VDC, current
around 300W/32V = 9.4A.

Sound reasonable, or not? Experience? Thanks~!

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[RussellMc]

Brent

I can add to this if needed but

- If you value WT survival and if it is in a site where it can be
potentially "overpowered" then be sure it has shutdown / loading / braking
/ feathering / stalling / off-winding mechanisms to protect it. SAs you no
doubt know, energy into the WT disk increases with velocity cubed. So a WT
rate to deliver 300W at say 10 m/s is notionally being delivered a
deliverable 2^3 x as much = 2400 W at 20 m/s. This will be modified by
stalling and other mechanisms but disk pressure rises with velocity squared
so you are liable to get significant loading increases even if power into
the alternator does not rise at Vcubed rate.
- In some designs hard shorting of the outputs can provide enough braking
to slow or stop an overspeeded WT but in some/many cases high enough wind
speeds will not be able to be overcome by electrical braking.
- A common method of power reduction is to have the WT head tilt off wind
or sometimes tilt-up under wind pressure. Some Chinese designs have
mounting holes designed to allow the head to pivot accordingly. Some have
these in the wrong place :-). The intention is that the weight forms a
torque to hold the head into the wind and pressure on the disk overcomes
this causing slewing. An alternative is to spring or gravitationally ramp
the "tail" so the WT is not held into the wind and as before turns cross
wind.
- In large commercial units blade pitch control is the norm but this is
liable to be unavailable on your unit.

Several loanable items listed below. I want to use them (again) longer term
but they are available for loan for some to quite a few months:

- I have a currently unused small (250W?) grid tie 24VDC-230VAC inverter
intended for (according to label) wind turbine use. It works but quality
and true capability is unknown and lack of legal certification for
connection to the grid is guaranteed :-). You are welcome to borrow it for
a while to play if of use. More details on request. (I have used it
successfully for trialling PV input. Long term it may end up islanding on a
caravan).

- Controllers often offer "load dump" outputs where excess energy is
shunted to. How and how well is TBD in each case. A good load for WTs are
available low voltage water heater "spears" - these may have two or 3
separate elements which can be either switched sequentially or PWMd. Don't
let them run dry for too too long :-). I have a friend who may have some
for sale and have one unit which may or may not still be usable if you wish
to play.

- Surplustronics in Ak used to be an OK source of large wattage resistors
suitable as loads. I may have some (used by me as exercise bike loads that
MAY suit your use. Resistances and wattages and findability tbd and may not
suit.

- You can also buy large diameter high current rating nichrome wire suited
for low voltage elements. I have a coil rated at 500W + air cooled - free
form self supported spiral wound (also used as exercise bike load - 500
Watt or more at ??? V* and easily tapped. (* Vin the 10's of V range
normally but would rise with user testosterone level and enthusiasm).

- More ...

Russell
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Brent Brown]

Hello Russell,

A little info on this particular wind turbine here, no deep hole - unless you buy one
;-)

https://www.1-day.co.nz/products/ecopower-outdoor-wind-turbine-3

Speed regulation: Automatically adjust windward direction
Control system: Electromagnet

Not externally obvious how it acheives adjusting of the windward direction, but
probably one of the ways you mention. I notice pushing directly backwards on the
shaft it seems to push against a spring, or could be magnetic force.

I'm on a hill, 2 story house, predominant wind westerly but have put pole up on the
north side to keep out of sight from the neighbours and to shield them from any
potential noise issues. Not sure if sound will be much of a problem, but could
relocate later if noise ok and more wind desired. It will be several metres away from
the side of the house, using balcony to help secure the pole, pole goes all the way
to the ground (un-official lightning rod). Turbine height almost the same as roof
peak.

Many thanks for the offer of bits and pieces, will see how we go~!

Brent

On 3 Dec 2018 at 0:10, RussellMc wrote:

> Brent
>
> I can add to this if needed but
>
> - If you value WT survival and if it is in a site where it can be
> potentially "overpowered" then be sure it has shutdown / loading /
> braking
> / feathering / stalling / off-winding mechanisms to protect it. SAs
> you no
> doubt know, energy into the WT disk increases with velocity cubed.
> So a WT
> rate to deliver 300W at say 10 m/s is notionally being delivered
> a
> deliverable 2^3 x as much = 2400 W at 20 m/s. This will be modified
> by
> stalling and other mechanisms but disk pressure rises with velocity
> squared
> so you are liable to get significant loading increases even if power
> into
> the alternator does not rise at Vcubed rate.
> - In some designs hard shorting of the outputs can provide enough
> braking
> to slow or stop an overspeeded WT but in some/many cases high enough
> wind
> speeds will not be able to be overcome by electrical braking.
> - A common method of power reduction is to have the WT head tilt off
> wind
> or sometimes tilt-up under wind pressure. Some Chinese designs
> have
> mounting holes designed to allow the head to pivot accordingly. Some
> have
> these in the wrong place :-). The intention is that the weight forms
> a
> torque to hold the head into the wind and pressure on the disk
> overcomes
> this causing slewing. An alternative is to spring or gravitationally
> ramp
> the "tail" so the WT is not held into the wind and as before turns
> cross
> wind.
> - In large commercial units blade pitch control is the norm but this
> is
> liable to be unavailable on your unit.
>
> Several loanable items listed below. I want to use them (again)
> longer term
> but they are available for loan for some to quite a few months:
>
> - I have a currently unused small (250W?) grid tie 24VDC-230VAC
> inverter
> intended for (according to label) wind turbine use. It works but
> quality
> and true capability is unknown and lack of legal certification for
> connection to the grid is guaranteed :-). You are welcome to borrow
> it for
> a while to play if of use. More details on request. (I have used
> it
> successfully for trialling PV input. Long term it may end up
> islanding on a
> caravan).
>
> - Controllers often offer "load dump" outputs where excess energy
> is
> shunted to. How and how well is TBD in each case. A good load for
> WTs are
> available low voltage water heater "spears" - these may have two or
> 3
> separate elements which can be either switched sequentially or PWMd.
> Don't
> let them run dry for too too long :-). I have a friend who may have
> some
> for sale and have one unit which may or may not still be usable if
> you wish
> to play.
>
> - Surplustronics in Ak used to be an OK source of large wattage
> resistors
> suitable as loads. I may have some (used by me as exercise bike
> loads that
> MAY suit your use. Resistances and wattages and findability tbd and
> may not
> suit.
>
> - You can also buy large diameter high current rating nichrome wire
> suited
> for low voltage elements. I have a coil rated at 500W + air cooled -
> free
> form self supported spiral wound (also used as exercise bike load -
> 500
> Watt or more at ??? V* and easily tapped. (* Vin the 10's of V
> range
> normally but would rise with user testosterone level and
> enthusiasm).
>
> - More ...
>
> Russell
> --
> http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist


--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[RussellMc]

On Mon, 3 Dec 2018 at 11:18, Brent Brown <***@eds.co.nz> wrote:

> Hello Russell,
>
> A little info on this particular wind turbine here, no deep hole - unless
> you buy one
> ;-)
>
> https://www.1-day.co.nz/products/ecopower-outdoor-wind-turbine-3
>
> Speed regulation: Automatically adjust windward direction
> Control system: Electromagnet
>
> Not externally obvious how it acheives adjusting of the windward
> direction, but
> probably one of the ways you mention. I notice pushing directly backwards
> on the
> shaft it seems to push against a spring, or could be magnetic force.
>


This appears to be identical


https://www.amazon.ca/Popsport-Wind-Generator-12V-Powerful/dp/B073LKZ555

WT head on base here.


https://images-na.ssl-images-amazon.com/images/I/61DOq4Ns77L._SL1500_.jpg

The vertical support shaft is probably offcentred in the head such that the
vertical axis of the support is downwind of the vertical axis of the
support rod entering the turbine head proper. Pressing very slightly off
centre against the upwind end of the head will slew it on this offset pivot
and an internal spring resists this.

Probably :-)


Russell
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[AB Pearce - UKRI STFC]

> I'm thinking a shunt regulator is the way to go with a wind turbine. Set at a voltage above what the solar PV produces
> (30.8V @ Vmp), but safely below Vmax allowed for inverter (45V operating, 55V Abs max), capable of dissipating 300W
> or so worst case. Could dump power into a power resistor or incandescent lamp or heater element, with say a PIC
> PWM'ing a MOSFET. Or even use multiple power MOSFET's in linear mode... the heat has to go somewhere and could
> possibly work out cheaper by not requiring expensive power resistor(s). Perhaps a fan on the heatsink kicking in as required.

Huh? If the wind is blowing to the extent that you are having to waste energy won't the wind itself be enough 'blow' across the heat sink, especially if using one with long fine fins that have a lot of surface area? Especially with the Southerlies you are likely to get (or are you in a very sheltered spot.

For a shunt regulator I suspect that a 2N3055 type jelly bean power transistor with a bit of smarts driving its base through a zener diode to set the shunt voltage will be about what you need. May require several transistors, and may want a resistor in each collector to limit the maximum current draw in event of semiconductor failure, or prevent such failure, and even out the dissipation between transistors.

> The wind turbine is rated at 300W, 24VAC 3ph. With a 3-ph bridge rectifier plus filter capacitor I'm expecting
> in the order of (24 x 1.41) - (2 x 1V) = 32VDC, current around 300W/32V = 9.4A.

I would be tempted to use Tungsten car headlamps as the load controlled by transistors as detailed above. However you would need truck ones for 24V, so I suspect aluminium bodied resistors mounted on heat sinks will be the way to go - you can select your load resistance more easily and accurately, and the resistors are available at reasonable price in wattage ratings that will allow you to run them well below rating.



--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Harold Hallikainen]

This now reminds me of a shunt regulator I did to equalize a series string
of lead acid batteries in an electric car. There I used a PIC to PWM a
transistor that switched a power resistor on and off. The PIC also drove
an opto isolated open collector bus with the battery voltage, battery
current (sensed the voltage across the negative lead going to the next
lower battery), temperature, bypass current, etc.

I suggest using power resistors as the load, as they are cheaper per watt
than power transistors. I also don't really like the variable resistance
of using an incandescent lamp as the lad. I did, however, use an
incandescent lamp as a current regulator to charge the batteries. The
charger consisted of three SCRs acting as switched rectifiers (to turn off
the charger at the end of charge) from a 3 phase wye service. These drove
the incandescent lamp to the positive side of the battery string. The
incandescent lamp was bright at the start of charge and dim at the end of
charge.

Good luck!

Harold




--
FCC Rules Updated Daily at http://www.hallikainen.com
Not sent from an iPhone.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Brent Brown]

On 2 Dec 2018 at 9:04, Harold Hallikainen wrote:

> This now reminds me of a shunt regulator I did to equalize a series
> string of lead acid batteries in an electric car. There I used a PIC
> to PWM a transistor that switched a power resistor on and off. The PIC
> also drove an opto isolated open collector bus with the battery
> voltage, battery current (sensed the voltage across the negative lead
> going to the next lower battery), temperature, bypass current, etc.
>
> I suggest using power resistors as the load, as they are cheaper per
> watt than power transistors. I also don't really like the variable
> resistance of using an incandescent lamp as the lad. I did, however,
> use an incandescent lamp as a current regulator to charge the
> batteries. The charger consisted of three SCRs acting as switched
> rectifiers (to turn off the charger at the end of charge) from a 3
> phase wye service. These drove the incandescent lamp to the positive
> side of the battery string. The incandescent lamp was bright at the
> start of charge and dim at the end of charge.

Thanks Harold. I'm not sure power resistors are so cheap, just looking around at
some 100W and 200W parts. Will do some better costing comparisons when I have
narrowed a couple of circuit options. A lamp (or a number of lamps) is certainly
somewhat attractive... with a risk of a blown lamp causing loss of voltage regulation,
but on the other hand high power dissipation in a small device, little or no heatsink
required, and inherent power indication~!

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Brent Brown]

On 2 Dec 2018 at 11:21, AB Pearce - UKRI STFC wrote:

> > I'm thinking a shunt regulator is the way to go with a wind
> turbine. Set at a voltage above what the solar PV produces
> > (30.8V @ Vmp), but safely below Vmax allowed for inverter (45V
> operating, 55V Abs max), capable of dissipating 300W
> > or so worst case. Could dump power into a power resistor or
> incandescent lamp or heater element, with say a PIC
> > PWM'ing a MOSFET. Or even use multiple power MOSFET's in linear
> mode... the heat has to go somewhere and could
> > possibly work out cheaper by not requiring expensive power
> resistor(s). Perhaps a fan on the heatsink kicking in as required.
>
> Huh? If the wind is blowing to the extent that you are having to
> waste energy won't the wind itself be enough 'blow' across the heat
> sink, especially if using one with long fine fins that have a lot of
> surface area? Especially with the Southerlies you are likely to get
> (or are you in a very sheltered spot.

Thanks Alan, I'm in Hamilton, most weather comes from the West. Probably one of
the highest houses in town but that's not saying much, still difficult to get turbine up
high enough into direct wind.

https://www.wunderground.com/personal-weather-station/dashboard?ID=IWAIKAT
O20

Was planning to put the electronics out of the weather, under a balcony, so wind
might help cool a little but fan forced I could make everything smaller.

> For a shunt regulator I suspect that a 2N3055 type jelly bean power
> transistor with a bit of smarts driving its base through a zener
> diode to set the shunt voltage will be about what you need. May
> require several transistors, and may want a resistor in each
> collector to limit the maximum current draw in event of
> semiconductor failure, or prevent such failure, and even out the
> dissipation between transistors.

Yip, I want to keep it fairly simple. By 2N3055 you mean the good old TO-3 steel
package? Can still get them apprently.

> > The wind turbine is rated at 300W, 24VAC 3ph. With a 3-ph bridge
> rectifier plus filter capacitor I'm expecting
> > in the order of (24 x 1.41) - (2 x 1V) = 32VDC, current around
> 300W/32V = 9.4A.
>
> I would be tempted to use Tungsten car headlamps as the load
> controlled by transistors as detailed above. However you would need
> truck ones for 24V, so I suspect aluminium bodied resistors mounted
> on heat sinks will be the way to go - you can select your load
> resistance more easily and accurately, and the resistors are
> available at reasonable price in wattage ratings that will allow you
> to run them well below rating.

Cool. I make potentially add a series regulator as well with decent high voltage
transistor, to protect the shunt regulator... if I have say a 200km/hr wind event, and
the wind turbine doesn't deploy its "face away" tricks then volts and amps might get
out of hand, I'd rather the generator fry than the regulator and then inverter.



--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[AB Pearce - UKRI STFC]

> Thanks Alan, I'm in Hamilton, most weather comes from the West.

Oh, OK, somehow I had got it into my head you were in Otago ...

> Yip, I want to keep it fairly simple. By 2N3055 you mean the good old TO-3 steel package? Can still get them apparently.

Yeah, that is what I was thinking of initially, but you can get equally good devices in a plastic package which would make it easier to mount on a heat sink as you only need tapped holes on the flat side and don't need to mill off fins to mount the body. Any suitably jelly bean power transistor with current and voltage ratings would do.

> Cool. I make potentially add a series regulator as well with decent high voltage transistor, to protect the shunt
> regulator... if I have say a 200km/hr wind event, and the wind turbine doesn't deploy its "face away" tricks then
> volts and amps might get out of hand, I'd rather the generator fry than the regulator and then inverter.

I wouldn't worry with a series regulator, you should be able to find a transistor with a suitable voltage rating to handle anything the generator puts out. As for dealing with over speed problems I would be tempted to shunt each winding with an Triac before your rectifier. Use optos to drive the gate to turn them on when in 'safety' mode in a manner very similar to the old drill speed controller or lamp dimmer. You could then use some very beefy triacs that don't need to worry about switching speed in normal operation, and you should be able to use 400V ones that would normally be used for mains control. Your biggest problem would be they will dissipate around a watt per amp of shunt current, so will probably need to be on a suitable heat sink. Three hefty current stud mount triacs into one HS would probably be acceptable (I am assuming the three phases are delta connected, it will need to be a bit more complicated if not).

Sounds like a fun project.

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist