Author Topic: Step-Down XFMR & Flickering Lights  (Read 21411 times)

listerboy

  • Jr. Member
  • **
  • Posts: 97
  • I love the smell of diesel in the morning....
    • View Profile
Step-Down XFMR & Flickering Lights
« on: February 22, 2010, 07:00:58 PM »
For the electrical experts out there........

I have an opportunity to snatch a GE 10kv 208V to 120/240V transformer and I'm wondering if it would help to reduce the flickering light syndrome and boost my voltage a bit. My generator is wired for 120V only (not using any 240 loads in emergencies) and at 60HZ only puts out 114V. So if I rewire the generator for 240 and run it through this transformer and reduce the rpm to get the required 208 for the transformer what would the results be? I'm sure it would increase the voltage but what would it do for the flickering and what would the efficiency loss be?

Thanks,
Dave

ronmar

  • Hero Member
  • *****
  • Posts: 1227
    • View Profile
Re: Step-Down XFMR & Flickering Lights
« Reply #1 on: February 23, 2010, 01:21:19 AM »
RPM = frequency.  You can't reduce the rpm or you will loose your 60 HZ.  114V at 60HZ isn't that bad, but it probably droops a little more with load application.  The transformer isn't going to help with flicker.  A harmomically excited generator output voltage, as you have discovered, is dependent on RPM.  A single cylnder 4 stroke applys power for 180 degrees out of 720 degrees of rotation.  Because of this, the RPM surges slightly.  This surge results in a slight voltage variation in a harmonic head.  Applying a varying voltage to a transformer input, will result in a varying voltage on the output.  There are types of transformers that can compensate for varying voltage, but they have a large efficiency penalty.  The best remedy for flicker is an AVR. 

As for your low voltage, may I suggest adding some electrolytic capacitance to your field.  The st field waveform after rectification is really loaded with ripple(real ugly waveform), so is not very efficient at powering the field(field expands and collapses with varying input).  The electrolytic capacitance acts like the flter circuit of a traditional DC power supply.  The cap charges with peak inputs and discharges to help smooth the output during lower inputs.  What this results in is a more stable input to the field winding, which results in an increase in output voltage.  It also cleans up the output AC waveform a little.   

Here is what the rectifyer output looks like without filter caps.
 

With filter capacitance added, it becomes a smooth DC voltage and feeds the field winding much more efficiently.  The ones I use are rated at 70 Volts and are around 1500 Ufarad(microfarad).  They are about the size of a "D" cell battery and have a positive and negative screw terminal side by side on one end.  Like a battery, Electrolytic caps are polarity conscious. You hook the + terminal of the cap to the + output of the field rectifyer and the negative terminal to the negative rectifyer output. Two 1500uF caps would probably get you an output voltage closer to 120VAC...  I will look around and see if I can find you a source of supply.  Radio shack dosn't have them that large anymore.  A stereo supply wharehouse might have them as they are used to help feed surge loads on  large automotive base audio amps...
« Last Edit: February 23, 2010, 01:54:24 AM by ronmar »
PS 6/1 - ST-5.

listerboy

  • Jr. Member
  • **
  • Posts: 97
  • I love the smell of diesel in the morning....
    • View Profile
Re: Step-Down XFMR & Flickering Lights
« Reply #2 on: February 23, 2010, 03:16:49 AM »
Thanks for the info ronmar,

I should have given more information on what I'm working with. It's a 2cyl 1800rpm diesel with a direst drive 8000 watt Mecc alte C2 alternator. The Mecc alte specs say it should be putting out 120V @ 1800rpm but I don't get close to that until the killawatt registers 64 HZ and the governor is maxed out. I have two killawatt meters and they are only about one volt off from each other so I'm pretty confident that the readings are accurate enough. Is it normal for the specs to be that far out?

Thanks,
Dave

mike90045

  • Mendocino Metro
  • Hero Member
  • *****
  • Posts: 1594
  • Mmmm BBQ
    • View Profile
    • Mikes Solar PV page
Re: Step-Down XFMR & Flickering Lights
« Reply #3 on: February 23, 2010, 04:15:37 PM »
64 Hz?   Realisticly, that is so close to 60Hz, that I don't understand how it's affecting the voltage with just 4 hz.  Try the capacitor after the rectifier.

ronmar

  • Hero Member
  • *****
  • Posts: 1227
    • View Profile
Re: Step-Down XFMR & Flickering Lights
« Reply #4 on: February 23, 2010, 04:28:10 PM »
Thanks for the info ronmar,

I should have given more information on what I'm working with. It's a 2cyl 1800rpm diesel with a direst drive 8000 watt Mecc alte C2 alternator. The Mecc alte specs say it should be putting out 120V @ 1800rpm but I don't get close to that until the killawatt registers 64 HZ and the governor is maxed out. I have two killawatt meters and they are only about one volt off from each other so I'm pretty confident that the readings are accurate enough. Is it normal for the specs to be that far out?

Thanks,
Dave

Sorry, got ST generator on the brain:)  Is there an electrical wiring diagram for the generator?  Perhaps inside the junction box?  That generator may have an adjustment for final output voltage, at least some do... 
PS 6/1 - ST-5.

listerboy

  • Jr. Member
  • **
  • Posts: 97
  • I love the smell of diesel in the morning....
    • View Profile
Re: Step-Down XFMR & Flickering Lights
« Reply #5 on: February 23, 2010, 06:54:04 PM »
ronmar,

I have the spec sheet on the alternator and there are no adjustments, just some diodes and a capicator. In the trouble shooting section is says "all machines (referring to the alternator in semi broken english from Italian) are regulated during factory testing". If one or more of the diodes was bad would that cause the drop?

ronmar

  • Hero Member
  • *****
  • Posts: 1227
    • View Profile
Re: Step-Down XFMR & Flickering Lights
« Reply #6 on: February 24, 2010, 02:55:53 AM »
Yes, a bad diode would not pass any current, or only pass limited current which would reduce excitation and output voltage.  Where is the capacitor in the circuit?
PS 6/1 - ST-5.

listerboy

  • Jr. Member
  • **
  • Posts: 97
  • I love the smell of diesel in the morning....
    • View Profile
Re: Step-Down XFMR & Flickering Lights
« Reply #7 on: February 24, 2010, 03:35:42 AM »
Thanks for the assistance on this ronmar,

If you click this link;  http://www.meccalte.com/index.php?s=57  and open #9 (the C2 manual in pdf) in the download section, pages 9 to 11 have the wiring diagrams. It's mostly Greek to me except on how to get 120 or 220 at 60 HZ out of it. There is no terminal box or cover from the factory as shown in the diagram, just a junction box for wiring that was attached to a side cover by the company that assembled the genset. Behind that cover are the capacitor and the wires from the alternator. The diodes are attached to the back of the rotor.

D

ronmar

  • Hero Member
  • *****
  • Posts: 1227
    • View Profile
Re: Step-Down XFMR & Flickering Lights
« Reply #8 on: February 24, 2010, 06:13:27 AM »
That is a brushless capacitively regulated generator.  The uF rateing of the capacitor determines the output voltage.  It basically charges from it's coil based on load, then discharges back thru the coil and induces a voltage in the rotor input windings.  This voltage is rectified by diodes located on the rotor and fed to the rotating field windings which induces the output voltage in the main stator windings.

Carefully check the electrical connections at the capacitor.  Also carefully clean the cap to make sure there is no path for current to bypass the cap which would reduce it's feedback to the rotor.  The diodes that rectify the current to be fed into the field are located in on the rotor shaft.  You can probably access them by removing the generator end plate/cover.  You will need a volt/ohm meter to check these, and it is best done with at least one of the two leads of the diode disconnected to get a good reading.  A diode can be thought of as an electrical check valve.  When measured with an ohm meter, a diode should have infinite resistance(open, no current can flow) measured in one direction and near a short(full current flow) when the test leads are reversed(measured backwards).  If this genset has the 50HZ optional connections, make sure the output windings and the capacitor connections are all configured for 60HZ operation.  The manual shows what should be connected where.  If the diodes check out, and all connections are good and capacitor is clean, recheck the voltage at 60HZ on the killowatt.  If it is still low, try applying a little load.  The capacitor is charged based on load current.  That is how it can compensate for load increases, so a little load may increase your output voltage.  Some loaded runtime may improve it's output also since it is, i assume a new generator.  If it remains low, with good diodes and connections, the capacitor is most likley at fault.  That is the most likley point of failure in this type of generator setup.       
PS 6/1 - ST-5.

listerboy

  • Jr. Member
  • **
  • Posts: 97
  • I love the smell of diesel in the morning....
    • View Profile
Re: Step-Down XFMR & Flickering Lights
« Reply #9 on: February 24, 2010, 01:40:37 PM »
Thanks!

Now I understand whats going on and a possible solution. I'll work on it this weekend and check all the connections and wiring.

You guys are the best. I visit this site every day (for the last 4 years) and have learned quite a bit. Probably enough to get myself fried if I'm not careful! ;D

Thanks again,

Dave

listerboy

  • Jr. Member
  • **
  • Posts: 97
  • I love the smell of diesel in the morning....
    • View Profile
Update on Low Voltage
« Reply #10 on: March 12, 2010, 08:33:49 PM »
Loaded the generator up with 7600 watts (95% rated capacity) and after about 30 minutes the voltage was 118 @58.5 HZ. It does just what ronmar said it would - as the load increases the voltage rises. All is well.

Now I need to find a governor that's regulated by frequency. I think I remember someone experimenting with one here a while back...


Thanks guys!
DC