Info on polarized capcitator

Status
Not open for further replies.

urganb

Member
Location
Upstate N Y
Charging from 110 ac single phase input via full wave bridge rectifier 110 dc out.-- Lets say a 200 volt 72000uf cap.--How would i determine approx. ohms of resistor to use between DC power supply and cap??-- Resistor should go between + and +terminals on rectifier and cap??--How do i determine power(amps)of various DC caps-- I Farred is the same as how many UF??--Could ,say a 600volt cap of same UFs be a better(safer)choice??-- Yes,i use bleed resisters on caps and am VERY careful.Any math needs to be simple,only grade schooled. Thanks,Ben.
 

steve66

Senior Member
Location
Illinois
Occupation
Engineer
You will get about 170 Volts out of a full wave bridge.

1,000,000 micro farads equals 1 microfarad.

Other than that, I think you may be getting over your head, so I am reluctant to provide more details. Perhaps you can tell us what your are trying to accomplish.

Wear safety glasses when dealing with electrolytic caps.

Steve
 
P

Physis 3

Guest
urganb said:
Charging from 110 ac single phase input via full wave bridge rectifier 110 dc out.-- Lets say a 200 volt 72000uf cap.--How would i determine approx. ohms of resistor to use between DC power supply and cap?

First, the capacitor is part of a power supply, not after it.

Second, you don't want a resister in series with the capacitor, unless there's some unusual requirement for one.

Aside from that I'd use a 600 v. just for some extra headroom being you're so close to max with a 200 and electrolytics have horribly wide value tolerances.

urganb said:
How do i determine power(amps)of various DC caps-- I Farred is the same as how many UF?

It doesn't quite work like that. A capacitor can discharge real close to instantaneously, throwing as much current out as the circuit and it's internal resistance will allow.

The only parameters you usually will have to consider is it's voltage rating and value in "farads", not Farred.

uf is microfarads or millionths of a whole farad. Or 1 farad is 1,000,000 uf.

What are you attempting to do and why are you using such a huge capacitance?

Edit: Error A
 
Last edited:

dbuckley

Senior Member
I can see the point of the resistor; a 72000 uF cap will look an awful lot like a short when the power is applied, and given the source is the infinite grid their could be a fair current flow, so the resistor will limit inrush current.

I recall an article from usenet some years ago, where some nutter built a really big strobe for a football stadium, and had loads of capacitors in parallel (hundreds maybe), and when that lot recharged it dipped the mains and caused all the audio system digital electronics to reboot, or something like that.
 
P

Physis 3

Guest
dbuckley said:
I can see the point of the resistor; a 72000 uF cap will look an awful lot like a short when the power is applied, and given the source is the infinite grid their could be a fair current flow, so the resistor will limit inrush current.

The capacitor would be, even at 72000 uf, way smaller than miniscule to the grid. It would be a near short for a microsecond more or less. If you want a victim here it's gonna be the bridge rectifier sending current to the capacitor.

I can also say it like this, if you need a current limiting resister in series, you don't need 72000 uf of capacitance. Seriously, think about that. Gobs of instantanious current and you wont allow current to be delivered to it quickly? What's the point? This is a primary reason I'm asking why so much capacitance.

dbuckley said:
I recall an article from usenet some years ago, where some nutter built a really big strobe for a football stadium, and had loads of capacitors in parallel (hundreds maybe), and when that lot recharged it dipped the mains and caused all the audio system digital electronics to reboot, or something like that.

So do you you still subscribe to Usenet? Maybe I'm wrong, but it sounds like pure BS to me. They didn't mention any breakers snapping did they! Did they?

Edit:

If the audio gear can't run for at least a second or two with no power, let alone a micro second or two, who's fault is it really. The guy who sold the stuff, the guy who bought the stuff or some capacitors?
 
Last edited:

RayS

Senior Member
Location
Cincinnati
you may wish to use a resistor large enough to limit inrush to below peak current handling of rectifier, then derate for reliability.

example: at 110v, say a 10A bridge, 110/10=11ohm minimum (assuming worst case into a fully discharged cap). Now ya gotta look at power dissipation-I(squared)xR for continuous current, and the voltage drop at continuous current. These 2 may be significant, and necessitate a different configuration- some type of inrush current limiter beside brute force resistance.
 

dbuckley

Senior Member
Physis 3 said:
[About strobes] They didn't mention any breakers snapping did they! Did they?

Nah. Just relating a story. Its worth what you paid for it.

But... modern strobes are a pain in the respect of breakers. Old strobes have a tube across a cap, and the cap charges, and the tube is triggered and discharges the cap, the cap recharges through a resistor. All peaceful. You can put several quite powerful strobes on one circuit, as the charging current is spread over a few cycles.

Modern strobes have the tube direct across the line, and just trigger it. In 240V land you can have just one strobe unit on a 10A circuit, and even that is less than the manufacturer recommends; They generally specify a 16A circuit. (example HighEnd dataflash, PDF manual, see page 7) Two of these sorts of strobes going at the same time on one circuit pops the breaker. And that is first hand experience, not some BS from the net.
 

rattus

Senior Member
urganb said:
Charging from 110 ac single phase input via full wave bridge rectifier 110 dc out.-- Lets say a 200 volt 72000uf cap.--How would i determine approx. ohms of resistor to use between DC power supply and cap??-- Resistor should go between + and +terminals on rectifier and cap??--How do i determine power(amps)of various DC caps-- I Farred is the same as how many UF??--Could ,say a 600volt cap of same UFs be a better(safer)choice??-- Yes,i use bleed resisters on caps and am VERY careful.Any math needs to be simple,only grade schooled. Thanks,Ben.

A full wave bridge fed directly from the 120V lines cannot be grounded. Both DC rails will move up and down relative to ground! And depending on the load the output voltage can be as high as 170Vdc. Use with care!
 

tallgirl

Senior Member
Location
Great White North
Occupation
Controls Systems firmware engineer
Physis 3 said:
I can also say it like this, if you need a current limiting resister in series, you don't need 72000 uf of capacitance. Seriously, think about that. Gobs of instantanious current and you wont allow current to be delivered to it quickly? What's the point? This is a primary reason I'm asking why so much capacitance.

Ripple current. Too small of a cap and you can't source or sink the current needed to keep the voltage flat.

Think about it at the boundaries. The infinitely small capacitor is a wire that's cut in two. No current is available to support the load when the rectified voltage wave form drops below the desired voltage. The infinitely large capacitor can absorb the entire rectified output current above the desired voltage, and source that current back to any loads without the voltage dropping (recalling that the "farad" is in units of amp-seconds / volt, after converting coulombs / second into amps).

The correctly sized filter caps can store the required charge until the rising side of the rectified output exceeds the voltage across the caps terminals and the caps begin to recharge. Too much capacitor and you waste money, too little capacitor and V = amp-seconds / farads exceeds the design spec.
 
P

Physis 3

Guest
Acually Julie I can see the point of both a charging resistance and a large capacitance for something like a stobe. Not a typical power supply though.
 

rattus

Senior Member
Hey Sam:

Hey Sam:

Physis 3 said:
Acually Julie I can see the point of both a charging resistance and a large capacitance for something like a stobe. Not a typical power supply though.

Sam, a shunt capacitance alone is undesirable for filtering purposes. You need a series R or better still an L in order to have a cutoff frequency.

Without this series element, you will see large current spikes which are undesirable. Then there is the effective series resistance of electrolytic caps which causes the cap to heat up.

Filter design is a bit more complicated than one might think, and I don't claim to be an expert although I have done a few in the last 100 years.
 

dereckbc

Moderator
Staff member
Location
Plano, TX
Not sure what anyone here is talking about. I have designed and built a lot of DC power supplies for ham radio and audiophiles.
You don?t add any series resistance to the filter caps, the resistance in the windings of the transformer and rectifier provide more than enough resistance to limit charge time. The only place for a resistor is across the caps called a bleeder resistor to bleed the filters off when the power is turned off, but they are not necessary. A good rule of thumb or KISS formula is 10,000 uf for every amp of designed output.
 
P

Physis 3

Guest
I agree with Dereck here Rattus. I've designed and built loads of typical power supplies. The filter capacitor is "always" strait shunted after the rectification. Given that it's a typical application.

And it's not the same thing as a filter (even though people like to call these filter capacitors). The complexity of those is far more involved.

Edit: Error A
 
Last edited:
P

Physis 3

Guest
rattus said:
Then there is the effective series resistance of electrolytic caps which causes the cap to heat up.

It's only microseconds of heavy current. Not enough time for making significant heat.

Edit: The site is goofing with the commands I've entered. Sorry Sean, but it's true.
 
Last edited:

rattus

Senior Member
Physis 3 said:
It's only microseconds of heavy current. Not enough time for making significant heat.

E/QUOTE]

Sam, there is ripple current in the cap, and there is ESR in the cap, and there is heat. It must be considered. And, if you are only charging for a few microseconds, the current spike must be huge, and the power increases 4X for each 2X of current increase.
 

ronaldrc

Senior Member
Location
Tennessee
My two cents, I'm with Rattus on this.

It really all depends on the application rather you need a series resistor or not.


If it is a DC supply for a battery charger, DC coil for a magnetic starter the series filter resistor would be useless and an extra expense.

For audio and video equipment we need a DC power supply with as little AC ripple in it as you can get.


The capacitor is a filter, it in conjunction with a series resistor and a filter capacitor create a RC network to smooth out the ac ripple.

If the impedance or current output of the transformer is greater than the charging capabilities of the capacitor, then the capacitor can not keep up with the current input.We need to match this as closely as possible to keep the ripple down.

We could use an enormously large capacitor as large as the input current but in the real world that would not make good economical sense.

All we need to do is supply the voltage and current or the power needs of the equipment
we are powering, and using the series resistor to lower the output of the source to match the input needs of the filter capacitor and the equipment is the
least expensive way of doing this.
 

steve066

Senior Member
A series resistor is common on smaller power supplies to keep the peak current from getting too high. The high spikes can damage the rectifier diodes, and possibly even the capacitors.

It may also be common to just use the transformer resistance and impedence to limit the current, but that would be a design issue. One should do the math for a paticular power supply and then decide if the transformer limits the current enough.

However, our original poster taked about using the 120V supply directly without a step down transformer. Thats something I would consider very unusual. And probably not something that is very safe for an electrician or engineer to be experimenting with, much less a non professional.

I think our original poster should be using a transformer just to make it safer to work with.


Steve
 
P

Physis 3

Guest
A series resistance is used when your using a zener diode for regulation and that's to limit the current through the diode so it doesn't blow up.
 
Status
Not open for further replies.
Top