What capasitor to run on a 150v circut?

[verytricky]

Ok,

You guys solved my last problem: I had a 0-300volt DC wind turbine and needed to log it with a 0-30 Volt logger. Excellent, job done and a few weeks worth of data!

Now logging per second, I see a very spiky voltage input. It goes from say 40 to 70 volts in a second. The average is around say 50 volts, but the voltage is definately spikey.

This is not a problem if charging a battery bank. But I am powering a grid tie inverter, and it has a thing to measure the voltage and adjusts to generate the best power from the voltage. So the thing is just clicking away changing over and over and not actually generating much energy.

I have limited the voltage to 150 Volts, as the grid tie inverter runs between 42 and 150 volts only. The similar grid tie inverter running on my solar pannel is excellent, and changes its input voltage style about 5 times a day, to get the best use of the varying voltages produced. The wind turbine one just spends all day clicking ....


What I want to do is smooth the current produced.

So I need a 150 volt capacitor? What sort of rating should it have? Any bright ideas?

Thanks!

 

[Boncuk]

Don't use a capacitor. Use a TRANSIL-diode instead. ST-Micro offers a wide selection of appropriate diodes, unipolar and bipolar.

To effectively suppress 70V spikes you might select the 1.5KE82CA (bipolar, stand-off voltage 70.1V, breakdown voltage 77.9V, peak pulse current 69A) or the 1.5KE68CA (bipolar, stand-off voltage 58.1V, breakdown voltage 64.6V, peak pulse current 83A).

Obtain an ST datasheet, printed title: "1.5KE6V8A/44A" and make your choice.

Part dimensions: body length 9.5mm, body dia 5mm, wire dia 1mm

Boncuk

 

[Leftyretro]

Well spikes are usually very short duration pulses, millisecs or microseconds. Is the voltage variation due to wind speed changes and gusts? If so I think nothing short of a DC regulator is going to be able to deal properly with this. What kind of current is being generated?

Lefty

 

[verytricky]

I would prefer not to cut off the spikes with a diode or such.... That would waste energy...

The windturbine is a thing called a WINDSAVE. It is quite rubbish. It costs around £2200 new - I got it for free as a friend dumpped it as it used more power than it produced. The grid tied inverter was very very poorly designed and built, and sucked electricity in standby mode. The turbine also needed 12m/s windspeeds to generate the 1.2kw it promiced, and then it shut down at 14m/s windspeeds! My average windspeed is 5m/s which produces around 60 volts. During any single day I go from zero volts to around 120 volts. I have not logged 150 volts yet, but based on projections and my historic wind data I will get periods of 150 volts.

Essentially I replaced the grid tied inverter with a Mastervolt inverter. I added a voltage restrictor to limit the voltage to 150 volts as this is the max the mastervolt unit can take.

I ran some samples of the voltages produced vs windspeed, and they corelated nicely day to day - ie a set wind speed always produced a set voltage give or take 10%. I ran this against my year long wind data and figured out I would only ever produce electricity with the WINDSAVE for about 38 hours a year, as it requires 250 volts to start generating electricity!

However, for about 4600 hours a year it would produce voltages between 42 and 150, which is where the Mastervolt inverter comes in - it is optimised for that range.

I believe there may be a flaw in the wind turbine itself, as the voltage produced seems to occilate in a range of around 30 volts, not in any descernable pattern, but hell of a random regularily. ie it is not cyclical or such - the logger is one reading per second, so I may be missing it, but I can not see any pattern.

I dont want to cut off the high voltage spikes, as the range inside which the occilation occurs changes with windspeed, I want to try remove the spikes and smooth the voltage so that the voltmaster can do its job.

Since replacing the WINDSAVE inverter with the mastervolt, I have generated 5 units of electricity in two days. In the three weeks I had the 'official' windsave inverter plugged in I generated 0.1 unit of electricity. If I can get a smoother voltage I will get more from the mastervolt inverter, as it will stop trying to adapt itself to the changing voltage every few seconds.

I remembered long ago in the mists of time that a AC to DC converter was able to smoothe the sawtooth voltage produced after the current had passed through the diodes using a capacitor. It sortof stored the spikes and released the energy in the troughs. I was hoping to effect a similar circut.

Bottom line is I dont have a huge budget, and I would like to get this thing working better than it is.

 

[verytricky]

For what its worth, at approximately 130 volts it produces 5.5 amps. I have only been able to get this one measurement. I assume this changes as the windspeed changes.

 

[Externet]

Hi.
----> "So I need a 150 volt capacitor? What sort of rating should it have? "...

If you have around any spare computer universal switching power supply, they have high voltage-high capacity electrolytics in there, waiting to be canibalized.
Miguel

 

[Boncuk]

A transil diode only cuts off what is beyond its breakdown voltage. If you use it for higher voltages you can make sure that no high voltage spike will damage your circuit(s). To use spikes (supposedly very short in the µs range) you might employ an electrolytic cap of 100µF/385V. Use an appropriate Schottky-diode to charge the cap. (Using a silicium type diode will almost null the power gaining effect.)

The entire increase of power is only worth thinking about if the spikes appear periodically. If they are sporadic just "kill" them.

Boncuk

 

[verytricky]

Hi,

The spikes are consistent - as in they appear 100% of the time.

Perhaps 'spike' is then the wrong word to use?


The voltage changes say +- 15 volts from the medean at that time. The median is determined by the wind speed and that ranges from say 0to 120 volts. If the median was 50 volts, then I would get readings from 35 to 65 volts.


Plotted graph would look something like this:

65 Volts
               x                          x
              xx           xx            x
     x        xx           x x           xx
    xx       x x   xx     x  x         x  x
--x--x--x x--x-x--x--x----x--x--x----------- 50 Volts 
 x       x  x   xx    x   x      x   x    x  x x
x         x      x      x  x       x x      x    x
                          xx         x               x
                           x 

35 Volts

 

[Hero999]

Your datalogger is probably too slow to see the ripple.

The turbine probably uses an alternator with a rectifier to give DC with a large ripple. Adding a capacitor certainly would help smooth the ripple but it probably isn't worth it as it sounds like your inverter can cope.

What's the maximum frequency the logger can handle?

Perhaps you could try an oscilloscope which would be fast enough to see the waveform.

 

[picbits]

Where abouts are you based ? If you're close enough to one of us there is a possibilty of someone popping over with a proper scope.

 

[picbits]

Also if the output is produced by something similar to a car alternator it will be 3 phase rectified DC so a reasonable amount of ripple on the output.

 

[ablemicky]

150 volts capacitor rating

if i get your question right what u are requesting for is the capacitance value of you capacitor,cos u just gave the working voltage.ur load current and the peak to peak voltage should be known,then use this formula C =I*V/2F I=Load current
v=peak to peak voltage f=frequency of operation (60hz i pressume)

 

[Boncuk]

... then use this formula C =I*V/2F I=Load current
v=peak to peak voltage f=frequency of operation (60hz i pressume)

The frequency varies with wind and generator speed.

 

[verytricky]

Wheeeeee!

I live in Warlingham. I have a tiny farmlet on the border of London, Kent and Surrey ( I am the actual border - If you look where all three join, thats me, I have 38m3 of Kent and 280M3 of London, and two acres in Surrey!)

I have obtained an ocilliscope from a mate - It will arrive in the morining.

The 'best' I can record with the datalogger is 1 second intervals. I never thought of an ociliscope till you mentioned it! The best I can record with the wind meter is one minute interval averages, with peak wind in that minute.


The inverter can cope, but it is definately switching - It clicks like a mad thing would click when a mad thing clicks! When I wanted to test the 'smoothing' theory, I connected 5 x 12v batteries in series ( massive yellow 110 amphour batteries from an old UPS ) and fed the wind turbine into that, then took leads off that to the inverter. It stopped clicking and the info from the inverter ( you can plug in to download info ) gave more power output in that 15 minutes than the previous 30 minutes. ( logging is an exact science? )

So the use of a battery cell as a smoothing device works, but -I blew one of the batteries - the side of it popped! I guess it did not like the spikes of 100 volts? So I stopped that idea. Also ? Power stored in the batteries? Could skew the results....

It would be great to be able to limit the voltage to something within a small range, but I then fear losing the ability to generate the full electricity the turbine would produce.

If we use the formula, and apply it to the 'most common' wind type - which is 4m/s we could come up with a capacitor value that works 50% of the time, and is better than nothing the rest of the time?

 

[verytricky]

New information:

The output from the wind turbine is a 'sortof sinewave'

It was measured with an occiliscope, and it definately has the sinewave shape, but it does change.

It starts off at around 30 volts +- 15volts at 2 hertz, and climbed to 90 volts +-22 volts at 4 hertz during the measurement period.

 

[Hero999]

That makes sense, does the ripple look something like this?

I doubt the spikes killed the batteries. I would assume that the float voltage across the batteries was too high. The voltage should never go about 15V per battery, the odd spike probably won't do any harm but if they were continuously charged above 75V.

The optimum voltage is probably more like 13.8V.

 

[verytricky]

YES!

That is almost exactly the wave being generated, However, you appear to have a constant voltage in your example, whils mine does drift up and down due to the wave being driven by windspeed and the wind is not always constant. But yes, that is what the DC looks like when we plugged the Occiliscope into the +ve wire.

Does this have any meaning?

 

[Hero999]

I just simulated a three phase voltage source driving s three phase bridge rectifier and ploted the DC-side waveform.

There is no special name, it's just called ripple.

Connecting a huge 200V 1000µF capacitor will reduce the ripple dramatically.

 

[verytricky]

Right, so I will be off to get me one of them then.

If I understand correctly, I can add several lower farad rated capasitors - which would do the same job as the one biggie?

 

[picbits]

Yes

If you use 10 x 100uf 200V in parallel it will give you 1000uF at 200V