3 Phase Shunt Regulator

plasmahunt3r
plasmahunt3r Registered Users Posts: 18 ✭✭
I got a shunt regulator to work.  It is different from other's I have seen, because it reads the voltage on the DC side and then Shunts the voltage on the AC side.    It is a technique used on motorcycles.  The regulated voltage is stable and the three dump resistors voltage reading is the difference from the input voltage and the regulated voltage. 

Normally, dump resistors have to be massive, because they are dumping the rectified DC, which is a combination of the 3 AC phases.  For instance, If you had 20A DC that had to be dumped, then Watts = R * I * I, or .73 Ohm * 20A * 20A or 292 Watts.  A lot of heat.  That 20A is combined of less than 7A for each of the three phases of AC.  If you shunt the AC side, then lower power resistors can be used:  Watts = R * I * I; or 1 Ohm * 7A * 7a; or 49 Watts (I used a 100 Wat for each of the three AC phases.
Tagged:

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    I am not sure it makes much difference--You can just parallel the three resistors on the DC side and get the same results.

    There would be a small advantage because you are not drawing full current through the rectifier (diodes run cooler--always a good thing).

    How did the dump controller control current to the AC dump resistors?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • plasmahunt3r
    plasmahunt3r Registered Users Posts: 18 ✭✭
    I built my own controller.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Looks good... Thank you.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • plasmahunt3r
    plasmahunt3r Registered Users Posts: 18 ✭✭
    I have been doing some experimenting with this circuit.  First of all, the 1 Ohm 100 Watt resistors get quite hot while the SCR's are only warm.  I replaced the shunt resistors with .73 Ohm 300 Watt resistors, and the resistors get mildly warm and the SCR's are just warm.  Overall, with 300 Watt Resistors, both Resistors and SCR's ran the coolest, but take up a lot of space.

    I then removed the dump resistors and went with a direct connection to the SCR's.  The SCR's are hot but the small heatsink seems adequate.  I researched "Motorcycle Shunt Regulator Circuits" on the interweb, and they use "Only the SCR's" as the shunt, without Shunt Resistors.  The SCR's i Am using is 2n6509 (800v 25A SCR's). 

    I am going to do some more testing, but I think I can eliminate the Shunt Resistor's in the circuit, and use only SCR's as the shunt.

    I am going to mount the wind generator on my truck and simulate wind conditions at 10, 15, 20, 25, 30, 35, & 40 MPH to complete my testing.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Monitor the stator wiring temperature (the windings in the wind turbine's alternator). "Heat" has to go some where--Using the dump resistors will reduce current in the alternator and can keep the windings cooler (some alternators have overheated their windings when shunted with a dead short in high winds).

    A dead short would, hopefully put so much torque on the alternator so that the blades stall... Adding dump resistors may lower current, and therefore torque, and the alternator will not stall/shut down, but continue generating power.

    In general, if you are expecting high winds, you should shut down the turbine before the high winds it. Or, design a turbine that can furl, feather blades, or mechanical brake (etc.) so that it will not self destruct in a wind storm (good idea to have two or more independent methods to safely shut down a turbine).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • plasmahunt3r
    plasmahunt3r Registered Users Posts: 18 ✭✭
    Thanks BB.  I have a question on Dump Resistors you may know.

    First, my Wind Generator for my sailboat is an Apollo 550W 12v and can do approximately 2 Amp at 15V; and as speed increases, it can to about 20 Amp at 36V.  That is less than 1 Amp per phase at 15v and less than 7 Amp per phase at 36v.

    Since each dump resistor is dedicated to each of 3 phases, I am having trouble seeing how a 1 Ohm Dump Resistor will relieve any stress on the windings

    On the 1 Ohm 100W resistor, Ohms law: 15v / 1 Ohm = 15A.  If the resistor is allowing 15A to pass, is it really doing anything to protect the alternator wiring?  I know it must be doing something, because it gets hot. 

    On the .73 Ohm 300W resistor, Ohms law:  15v / .73 Ohm = 20A.  This resistor runs cooler.

    No Dump Resistor, and the wires to the SCR are cool.

    I tried a 100 Ohm 200W resistor, and the Shunt Regulator wouldn't regulate the voltage. Shunting requires a little or no resistance.
     


  • plasmahunt3r
    plasmahunt3r Registered Users Posts: 18 ✭✭
    I think it is clear from my testing results, that resistance produces heat.  Just like Nichrome Resistance Wire used in a toaster produces heat. 

    Measuring the resistance on the wind generator, the resistance between any two phases on the wind generator is 4 Ohms.  This will produce a centain amount of heat whether or not there is a Shunt Resistor. 

    I know some people have a brake switch on the wind generators, which closes the wind gen circuit on the DC side (and disconnects from battery).  Shunting is similar, except that it is automatic, closes only on the positive half cycle, and SCR turns off on the negative half cycle. So shunting is not a complete brake, but an oscillating on/off switch when voltage on the DC side reaches 14.5v.

    When I mount the generator on the truck for testing, I expect the Wind Generator to slow down at high speeds, but not stop, due to the ocsillating nature of the SCR on positive half cycles.

    The shunting technique is used on Triumph, BSA, and Ducati motorcycles.  Harley has Three Phase PMA Alternators, but use series regulators.  I have ordered a Harley Softail Series Regulator for testing, but no braking would occur on series regulators so it will not meet my goal.

    My goal with the Shunt Regulator is an automatic system that will regulate voltage and protect the Wind Generator from over speed conditions.  But with all projects, there is theory and reality.  Only testing will tell.  I will post my results and final tweaks on the circuit.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Just the lower the dump load resistance, the higher the peak current will be through the windings.

    Some folks like a two stage shut down... First a dump resistor load to slow down the turbine, and then short the turbine once is has slowed down some (keep peak current lower).

    In the end, measure the temperature in the stator of the alternator. If you have large diameter blades vs alternator size--Then you run the risk of overheating. If the blades are small diameter, they will stall before you get a damaging amount of current.

    One possible problem--If your SCRs are "rectifying the current" in the alternator--I wonder if that may case overheating problems on its own... Alternators (and transformers) rely on the current changing direction every cycle... When rectified current flows--I am unsure if that will cause higher than normal eddy currents in the windings/steel pole pieces (not sure--just have not thought/seen anything about that).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • plasmahunt3r
    plasmahunt3r Registered Users Posts: 18 ✭✭
    Here is the revised circuit.  I am going with the 300 Watt resistors because the 100 Watt were too hot.  Also, I moved the Shunt Resistors to the negative side of the SCR's and paralleling them for 600 Watts.

  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    First, power is power.  AC watts = DC watts.   A 500W turbine needs a dump load that can manage 750W for some length of time, if the wind is blowing and the brakes don't hold, you have to dissipate the power somewhere, and you don't want a marginal load starting a fire.

    SCR's on a AC circuit, will only pass half the waveform (positive half or negative half, depending on the wired polarity) A TRIAC will pass both + & - sides.   So maybe you think your 150W of heatsink on the SCR is doing the same as 300W of resistors - but with AC, it is NOT.
    A TRIAC wires up just like an SCR, it's like a dual SCR in one package.

    The windings in the turbine alternator are able to dissipate some heat, but it's best to go with what the Mfg recommends unless you know it's overbuilt enough to short the output leads before the diodes, and brake the turbine that way.
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • plasmahunt3r
    plasmahunt3r Registered Users Posts: 18 ✭✭
    This is a picture of the Shunt Controller using dual 300W Dump Resistors in parallel.  I will run a full system test as soon as I can figure out how to mount the Wind Gen on my truck; without drilling holes.  I have some ideas I am currently working on, like clamping a mount on the truck bed rim. 

    I am using the Island Cutter method of producing circuits.  I use diamond hole saws to cut islands in the copper PCB to isolate components.  This is a quick prototype method.  The rest of the PCB is a negative ground plane.
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    This picture with the two paralleled resistors looks like it is connected up quite different than your earlier schematic.  This new picture looks like the resistors are using the negative output diodes to combine them so you only need one resistor instead of three.  Is that right ?
  • plasmahunt3r
    plasmahunt3r Registered Users Posts: 18 ✭✭
    That's right boB.  This is the revised circuit I did on December 23 (above).  I am running the two parallel to give 600 watts @ .365 Ohms.  The 100 Watt resistors ran too hot and the 300 watt resistors tested cooler.
  • plasmahunt3r
    plasmahunt3r Registered Users Posts: 18 ✭✭
    I have been doing some more testing.  Ideally, I should get 14.6V (a 14V zener plus .6V PNP transistor).  There is a lot of variation in Zener's, with my 14V Zener actually being 13.8V.  So 14.4V should be the result.

    When I start up the Shunt Regulator, it starts at 13.6V and slowly goes up as the battery reaches full charge.  However, when I add a 3000 uF 35V capacitor at the DC side of the Bridge, the voltage jumps to 14.4V right at the start.  The capacitor smooths the voltage at the top voltage.  Without the capacitor, I think I am getting more of an average voltage.

    I think the smoothing capacitor is essential and I have added it to the circuit.

  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Zener diodes have a "knee" where they slowly start to conduct and as voltage increases slightly, they "turn on" a bit more.  Usually, they are used to trigger a better device, as most zeners are fairly low power.
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • plasmahunt3r
    plasmahunt3r Registered Users Posts: 18 ✭✭
    I got the Wind Gen mounted on my truck and got some performance stats on the Apollo 550W 12V Wind Generator.

    I first tried hooking up my meter to two of three phases, with no load, and I got 31v at 15 Mph.  I think this was high and because the meter was connected between two phases, and the reading was cumulative (double).

    I then hooked up the controler and started DC test.

    Based on the performance stats for my Wind Turbine, If I want the Shunt Regulator to kick in at 30 Mph and start slowing the WindTurbine, I think I should adjust the Shunt Regulator to use a 13v Zener and kick in at 30 Mph 13.6v.

    Apollo 550W Performance 10 Mph 12.7v 15 Mph 12.9v 20 Mph 13.1v 25 Mph 13.3v 30 Mph 13.6v 35 Mph 14.1v 40 Mph 14.4v Shunt Regulator kicks in 45 Mph 14.4v
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    The reason we use a (at least) two stage regulator (with some delay timing). More or less, 13.6 volts is not really charging a lead acid battery... It is "float" charging for long term storage/standby power (and/or drawing DC power and want the turbine to "make up" the energy used).

    Nominally, if you are cycling the battery bank (typically to and below ~75% state of charge), you want the battery charging voltage to be around 14.5 to 14.8 volts (when the battery "hits" 14.x volts, the regulator kicks in and holds that voltage, while a timer of 2-4 hours, longer if the battery is more deeply cycled) and then after the time limit, the regulator drops back to float charging.

    For a "simple" wind turbine regulator, nominally set to 14.x volts and don't bother with float--The wind typically only lasts for minutes to a few hours, not for days on end. You could do something like set for 15.0 volts (to pick a voltage) and when the set point is hit, then put the dump/brake load on the turbine. And keep the brake load on until it falls to below 12.7 volts--I.e., only when the battery is being discharged. Let the solar panels (and other charging sources) do the "fine control" of charging.

    If you do have high winds that last for days+==Then you do run the risk of overcharging the battery (and possibly battery fire/explosion). From a code (NEC/NFPA) point of view, dump controllers should be installed in pairs (redundant/backup controllers can be different modes/models/etc.) so that if one fails, there is a second one that can still regulate the battery bank voltage to within safe limits.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • plasmahunt3r
    plasmahunt3r Registered Users Posts: 18 ✭✭
    Hey BB.  Per your statement:  13.6 volts is not really charging a lead acid battery... It is "float" charging for long term storage/standby power (and/or drawing DC power and want the turbine to "make up" the energy used).

    The purpose for this wind generator is to run a refrigerator while I am out sailing.  So I will be "Drawing DC Power" while sailing and using dockside power while at the marina.  The gen's purpose is to make sure the battery is charged and ready while dockside and to supply power during the trip.  I am thinking the Float Charge / Standby is what I need for dockside.

    Also, Deep Cycle Gel Cell batteries have a max range of 13.6-13.8v charging.  I have been looking at some Mastervolt AGM battries that can handle 14.5V, but not lead acid on a boat.

    I guess it all depends on your requirements.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    That sounds good then. AGMs typically can accept upwards of 14.4 volts when charging. But if you start out charged, and plan on the turbine supplying most of the power until you return, then why not.

    Suggest you stay away from GEL batteries. In the US, at least, most can only accept about 5% rate of charge (high charge rates cause gassing and form permanent bubbles in the GEL). AGMs (deep cycle specifically) are much better. Flooded cell are much cheaper than GEL or AGM (but have their own issues a sea).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
     I have been looking at some Mastervolt AGM batteries that can handle 14.5V, but not lead acid on a boat.
    I prefer to have two feet on land, but I've seen lead acid batteries on boats.  What is your alternative to lead acid?
    BB. said:
    Flooded cell are much cheaper than GEL or AGM (but have their own issues a sea).
    What are some of the issues for flooded lead acid?  I would think that the rocking motion would be a great way to break up stratification... you might even be able to use lower charging voltages if you don't have a need for vigorous gassing.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • plasmahunt3r
    plasmahunt3r Registered Users Posts: 18 ✭✭
    Gel cell and AGM battries will work under water.  If you are sinking, you want your bildge pumps to keep working. 
  • plasmahunt3r
    plasmahunt3r Registered Users Posts: 18 ✭✭
    I ran another test and I think the results warranted a final update.  In my first test, the 550W Apollo Wind Turbine didn't produce 14.4v until 40Mph.  Well, the Apollo Wind Turbine comes with some shims that alter the angle of the blades.  These shims cause the blades to produce less torque and more speed.  I put in the shims, and now the Wind Turbine produces 14.4V at 20Mph.  A significant improvement.  With less Torque, the Turbine will not spin at 5-10Mph, but once it gets spinning, the voltage is significantly increased. My 3 phase shunt regulator still caps the voltage at 14.4v; just triggers at a lower wind speed.

    I am done with this project and I am happy with my results.