STC-10 AVR and capacitors

[oiler]

Hi all

Now my genset is running....at last, it took me two years to get this far  

I have a question. Where do i find an AVR for my STC-10 three phase? All i can find is Single phase ones  
Also the humming of some of our appliances is annoing SWMBO, as does the flicker. I've read somewhere it possible to add a capacitor, but where to put it and how about the size?

The exitation is 100vdc@4.8 amps. Output is 400-410 VAC 50Hz.

Torsten

[NoSpark]

I've asked these questions, seems a capacitor will help clean up the wave form but does nothing for flicker. Georgia Generator sells 1 and 3 phase regulators on Ebay, I'm still up in the air about trying one. Appliances humming?

[Henry W]

Hello oiler,

Contact Tom, He should be able to answer your questions.

Henry

Tom Osborne
akatosborne@hotmail.com
www.centralgagenerator.com
478-457-5524

[dubbleUJay]

I've asked these questions, seems a capacitor will help clean up the wave form but does nothing for flicker.

Maybe you guys know this, but here goes:

A capacitor is like a short circuit to AC, but you have to match the capacitance to the specific frequency you want to filter/short out.
So if you know what the frequency of the undesired waveform is, you can use a capacitor for that waveform to "short out" this frequency, but it would not affect the 50hz/60hz.
There is a formula to work this out, but I just don't have it on hand right now.

Flicker is caused by the voltage dropping when the engine slows down after the power stroke and going higher again during the power stroke.

Or that's more or less how I understand it in layman's terms 

dubbleUJay

[BruceM]

I'm surprised to hear that the 3 phase ST heads with harmonic excitation have such a bad AC waveform that they cause humming in appliances.  It seems they have the same problem as the single phase heads.

In the single phase ST heads, the waveform distortion is caused by the harmonic winding, and if excitation is taken from the mains (such as by an AVR), then the waveform distortion is gone, and you get a decent sinusoid.  You can prove this to yourself by connecting a variable transformer for excitation. Bill Rogers did some outstanding research on this topic and published waveforms on the old forum. At his advice, I also tried the variable transformer on my ST3, and sure enough, the distortion that is characteristic of the harmonic excitation, known as the "harmonic hump" is GONE.  With my AVR for excitation, I get no "moaning" from my clothes washer timer.  With Harmonic excitation, it's bad.

Other common waveform issues with SOME STs is a "ratcheting" of the waveform seen most prominently near the peaks.  In some cases, there are some high frequency spikes near the peak associated with the ratcheting of the waveform.  Adding some capacitance on the DC side of the bridge can sometimes help with this. On my ST3, it was not effective.  

To reduce high frequency EMI on the AC, which you can pick up using an AM radio held near any connected AC cable, you can put 4 AC rated metal film type capacitors (0.1 uF) around the bridge diode.  There is a piggy back style spade connector that you can use for this, if you have spade type bridge diode. Diodes squeal with a burst of high frequency noise whenever they start or stop conducting. The 0.1 uF caps will absorb much of that EMI, the difference on the AC output is amazing.  

Flicker will be substantially reduced with a good AVR; the AVR is able to compensate for some of the power stroke induced speed variation.  The AVR is limited in this by the huge inductance of the rotor windings, which slows the corrective response time.  Therefore, the smaller the head, the better the flicker reduction.  It's one case where the ST-3 is superior to the ST5.

Bruce M

[NoSpark]

I learn more every day   What about the AVRs that are wired inline with the harmonic winding, anyone with experience with this type? Do they clean up the AC from the harmonic winding?

[dubbleUJay]

Thanks BruceM, do you by any chance have a cct. diagram for a reasonably decent AVR in you huge archives?

dubbleUJay

[BruceM]

Nospark-
For and ST head, I would avoid an AVR that used the harmonic winding for it's power source, if I wanted a good looking waveform and decent flicker reduction.

The waveform artifact known as the "harmonic hump" is caused directly by the harmonic windings position on the stator. If you use the harmonic winding, you'll see the "harmonic hump".

For good flicker reduction, you need a high voltage source to briefly drive the excitation harder on the compression stroke.  You can't do that using the harmonic, it's just barely enough to do the steady state job.

The harmonic voltage does spike big with big loads on power, on my ST3 it was over 300 volts (near 400 I think) peak.

WJ- I sent you schematics for both of my AVR designs.

The simple one for the ST5 is kind on the brushes- it uses rectified line voltage for the excitation, and for normal operation it is only pulsed on briefly near the zero cross time when voltage is low.  For current surges, the pulse stretches into the much higher voltages, and this is also needed for the flicker reduction.

[oiler]

Hello oiler,

Contact Tom, He should be able to answer your questions.

Henry

Tom Osborne
akatosborne@hotmail.com
www.centralgagenerator.com
478-457-5524

Thanks.

I've been allover his site, but i can't find anythig that is set up for 400 volts three phase 50Hz.

When using the line voltage for exitation, can the harmonic winding be used for battery charging?

[BruceM]

Any load on the harmonic will cause distortion of the AC waveform, loads should be on the AC mains.  You'll have to get a step down transformer, the you can drive a conventional AVR off of that.

[NoSpark]

I was assuming that the AVRs that Tom sell are wired in series with the harmonic winding, but maybe they can be fed from line voltage. Bruce, would it be possible to post your AVR schematics.

[BruceM]

Here is the simplest AVR schematic.  Note that this one requires a push button manual start.  This eliminates the need for a 12V battery, but wouldn't be suitable as is for an automated system.  I have a PCB layout for this that Cliff Myers created.  It's simple and very small.  The IGBT transistor isn't available anymore but there are high voltage MOSFETs now that will do the job better and with the same pinout.

[mobile_bob]

Bruce:

thanks for posting the circuit,
i am pleased to see the 5 amp fuse on the output to the field, as you know there
are those concerned with overpowering the field coil should the engine start to lose
rpm for whatever reason (such as running out of fuel)

my hehr regulators have a 10amp standard mini spade fuse that will darn sure blow
long before either my field coil is damaged or the regulator is damaged.

apparently there are some gensets on the market that have no such protection,
and the field or the regulator can be damaged if the unit is powering a load and the engine runs
out of fuel??

seems like a reasonable thing to have, a 50 cent fuse

bob g

[BruceM]

The fuse is crude but effective. This circuit was intended to be the smallest, simplest possible.  It fits the expresspcb.com 2.5" x 3.8" experimenters board, 3/$51.

It wouldn't be hard to use the rectified excitation voltage reduced by resistor and time averaged via capacitor fed back into the op amp circuit to reduce excitation. This would then reduce the output AC voltage during a heavy load.

I haven't found this to be necessary for my use, though I have to keep the belt tighter on the generator than the harmonic setup to avoid chirping on a surge load.  My prototype is different from this circuit in that it uses harmonic winding for starting and it has time delayed over and undervoltage trips.

I woud not bother maintaining harmonic winding function if I were doing it over.  At the time, harmonic backup availble with a single switch seemed a good idea.  

Important note:  The rectified AC (signal from tiny step down transformer) should have a value of 10uF for maximum flicker reduction, though voltage regulation suffers slightly.  

I'm happy to help someone who is building one.

[12gauge]

As with others, I have desired to reduce flicker with my st-5/6-1, and I never could get more than about 113 volts at full load.  So I built the McCreary simple regulator discussed here and am posting the results with the hope someone may find it usefull.  I've consulted with Bruce McCreary several times in the course of this project and he has been a great help with the success of it. 

I used a project board from Radio Shack instead of the ExpressPCB board.
The pass transister Q1 used is IRGB14C40LPBF as the specified one is unavailable.  I mounted it to a small heat sink I had on hand and it operates warm to the touch.  I would use a bigger heat sink next time but I'm not worried about this one.  The regulator is mounted in a box off the generator.

http://picasaweb.google.com/lh/photo/5-9KPEVOVPmdTIUr9N-ffA?authkey=Gv1sRgCN-YqvmH1oi44QE&feat=directlink

When first hooked up I found my inverter/charger would not accept the power and investigation with an oscilloscope showed some voltage spikes on the waveform.  A 50 uf capacitor across the line fixed that problem. 

The end result is less flicker than before, stiffer voltage regulation, and the ability to actually adjust my voltage up to 120 volts.  The flicker is only noticable with incandecent bulbs and just slightly in a mercury yard light.  Compact flourecents have none.

IMO this is a worthwhile project and for whatever reason can easily be reversed as the Z winding wires are just disconnected and taped off.

73, RH