Measurment Peak or RMS

Thread Starter

Calcifer

Joined Jun 19, 2013
14
i was looking at some circuit designs and i noticed that some of them contained components whose voltage ratings were more in line with the Peak voltage.

*so my question is.. Are the voltage ratings on a component (Capacitors, Resistors, Diodes, Etc) listed as the Peak Voltage for AC circuits?
*and if i am using a DC circuit, do i just use the highest voltage amount?
*if i am using an AC to DC circuit through a bridge rectifier, how do i choose the right voltage for that?

i know that Transformers are listed as RMS, and that Meters Measure RMS
 

crutschow

Joined Mar 14, 2008
34,201
You want to use peak voltages with a margin for reliability. In a rectifier circuit, the peak voltage across the diodes is the peak-peak voltage of the input waveform.
 

studiot

Joined Nov 9, 2007
4,998
Well that's a good question because it varies.

Take diodes.

Diodes are rated for RMS current. That is a 1 amp diode is rated to run at 1 amp RMS (or less) for its entire service life.

The forward voltage drop across a diode depends upon the current through it. This will be RMS current if alternating. And you'd better have a series current limiting device (eg resistor) otherwise you will burn out the diode.

In the blocking or reverse direction the diode (in theory) passes no current, as voltage increases, until it breaks down.
The safe voltage just below this is called the Peak Inverse Voltage or PIV. This is the voltage rating of the diode and you'd better not exceed it, even for a short period or you can destroy the diode.
So you need to make sure that the diode never has more reverse voltage across it than the PIV.
This may be the peak voltage or the peak to peak voltage or the DC voltage depending upon the circuit.

Most component specs are actually like this.
The current is rated working current and may be DC or the AC equivalent ie RMS.
The voltage is an absolute maximum whether DC, peak or peak to peak AC.

Unfortunately collector current is another absolute. Edit oops! (A 25) A 15 amp transistor will handle 15 amps peak, not RMS.

Sometimes, as with resistors, it is the power rating that dominates. So the max current/voltage will vary with resistance so that the power rating is not exceeded. In this case DC or RMS AC values are appropriate for both current and voltage.

You should always check the power as well as the current and voltage ratings because any device may not be able to reach both simultaneously on account of power rating.
 
Last edited:

wmodavis

Joined Oct 23, 2010
739
Very few meters actually measure RMS. If it doesn't say it is a TRUE RMS meter then it does not measure RMS. Most respond to average and are calibrated to indicate the corresponding RMS value of a pure sine wave. Big difference.

Do you know what your meter measures?
 

bountyhunter

Joined Sep 7, 2009
2,512
Very few meters actually measure RMS. If it doesn't say it is a TRUE RMS meter then it does not measure RMS. Most respond to average and are calibrated to indicate the corresponding RMS value of a pure sine wave. Big difference.

Do you know what your meter measures?
Even fairly expensive meters that say TRUE RMS are very limited by bandwidth and crest factor. The only true way to measure RMS value of any arbitrary AC waveform is with a heating element and the meters that do that are ridiculously expensive. Most modern meters use some kind of mathematical integrator thing.
 

Thread Starter

Calcifer

Joined Jun 19, 2013
14
In a rectifier circuit, the peak voltage across the diodes is the peak-peak voltage of the input waveform.
*So if im using a 120v to 24v step-down Transformer and into a Full Wave Bridge Rectifier,
*would a 50v rectifier be sufficient? or are you saying that i would need to use the Peak from the Transformer (about 67v) so i should use a 100v rectifier?
*Or are you saying that The components Following the Rectifier would need to be greater than the 67v peak


Do you know what your meter measures?
yea mine is not True RMS, i thought it was back when i bought it.. Must have misread it.


Well that's a good question because it varies.
Unfortunately collector current is another absolute. Edit oops! (A 25) A 15 amp transistor will handle 15 amps peak, not RMS.
Lol i was just about to Reply and ask you how you had come to that when i noticed the Edit..
That is alot of Really Good Information.
 

studiot

Joined Nov 9, 2007
4,998
The peak alternating voltage to each diode in a bridge type full wave rectifier is the peak (ie 24√2=33.9 in your case) volts.

Edit: But if the rectifier is followed by a capacitor the peak inverse voltage is doubled every negative half cycle.
Consider the diodes feeding the positive terminal of the capacitor. This is at +33.9 volts.
When the other end of the diode is taken negative by the transformer it goes to -33.9 volts
Thus the total voltage across the diode is 2 x 33.9 volts.
So you will need at least 100 PIV diodes.

It is never a good idea to size component to close to their limits.

The components following the rectifier (a capacitor?) will only see the output of the rectifier so need to be 33.9 plus a safety margin say 45 or 50 volt components.
Electrolytic Capacitors are designed to operate at or near their rated voltage, and do not work well at a much lower voltage.

The current rating (you have not stated the operating current) is important for both the diodes and the capacitor as most of the current into the capacitor is delivered in a very short part of the cycle so the input current to the capacitor from the bridge is much larger (often several times larger) than the average output current.
 
Last edited:

The Electrician

Joined Oct 9, 2007
2,971
Take diodes.

Diodes are rated for RMS current. That is a 1 amp diode is rated to run at 1 amp RMS (or less) for its entire service life.
This appears not to be true. Here are a number of manufacturer's datasheets and they all specify maximum average forward current; not one mentions RMS current.

An RMS rating would be appropriate for a device across which the instantaneous voltage is proportional to the instantaneous current through it, such as a resistor, but the voltage across a rectifier diode is more nearly a constant. For example, look at the forward voltage vs. current curve of a Fairchild 1N4001. When the forward current increases from .2A to 1.0A, a 5 times increase, the forward voltage increases from .8 volts to .93 volts, nowhere near the 5 times increase in voltage drop that would occur for the same increase in current through a resistor. Hence, the heating effect is proportional to the average current, not the RMS current.

Some 1 amp diodes:

http://pdf.datasheetcatalog.com/datasheet/DiotecElektronische/mXrwzsy.pdf

http://pdf.datasheetcatalog.com/datasheet/philips/1N4005.pdf

http://pdf.datasheetcatalog.com/datasheet/fairchild/1N4004.pdf

http://pdf.datasheetcatalog.com/datasheet/motorola/1N4002.pdf

http://pdf.datasheetcatalog.com/datasheet/GeneralSemiconductor/mXuzwsr.pdf

http://pdf.datasheetcatalog.com/datasheet/GeneralSemiconductor/mXuzwsr.pdf

Some 3 amp diodes:

http://pdf.datasheetcatalog.com/datasheet/fairchild/1N5401.pdf

http://pdf.datasheetcatalog.com/datasheet/vishay/1n5400.pdf

http://pdf.datasheetcatalog.com/datasheet/GeneralSemiconductor/mXyztrxt.pdf

And you'd better have a series current limiting device (eg resistor) otherwise you will burn out the diode.
All the diodes in the datasheets above have a surge current rating sufficient to preclude the need for a current limiting resistor in series if used in a rectifier circuit sized for the rating of the diode. What I mean by this is that where a 1 amp diode is used to provide a DC current less than 1 amp, the series impedance of a transformer suitable for that purpose will not be able to provide a surge current sufficient to damage the rectifier diode.

The 1N5400 series of 3 amp diodes, for example, have a 200 amp surge rating. It would be hard to hurt one of these with the current able to be supplied by a suitably rated transformer. This is typical for modern silicon rectifier diodes.

In the blocking or reverse direction the diode (in theory) passes no current, as voltage increases, until it breaks down.
The safe voltage just below this is called the Peak Inverse Voltage or PIV. This is the voltage rating of the diode and you'd better not exceed it, even for a short period or you can destroy the diode.
Certainly the PIV rating should not be repetitively exceeded as would happen if one were to use a 100 PIV rated diode in a rectifier circuit supplied from a 200 VAC transformer secondary. But, if the environment is liable to subject the diode to occasional overvoltage spikes, there are so-called controlled avalanche diodes specifically designed not to be destroyed by such events:

http://www.bing.com/search?q=contro...s=0&ppfsrcig=520a73a349834dd1b0b8a77c8dded09f
 

studiot

Joined Nov 9, 2007
4,998
@ the electrician

Calcifer appears to be trying to actually build a power supply and wishes to obtain the correct components.

Are you trying to quibble or help?

If you are trying to help perhaps you would like to address Calcifer's specific questions and voltages (eg 24 volt secondary, not 200).

I know and you know that there are many more complicated rating parameters about many components, but I do not wish to enter into a squabble about them all.
 
Last edited:

studiot

Joined Nov 9, 2007
4,998
Quote:
Originally Posted by studiot
@ the electrician

Calcifer appears to be trying to actually build a power supply and wishes to obtain the correct components.

Are you trying to quibble or help?

I'm trying to help him avoid making his component decisions based on wrong rating methods.
That is why, when calcifer asked a specific question about what votlage rating diode to choose, you didn't answer?

He didn't actually mention current, I did.
 
That is why, when calcifer asked a specific question about what votlage rating diode to choose, you didn't answer?

He didn't actually mention current, I did.
I wasn't responding to calcifer; I was responding to you.

My post wasn't intended to answer the OP's questions directly. My purpose was to correct incorrect information, and supplement incomplete information you gave.

What you told him about diode voltage ratings wasn't incorrect; you said "This is the voltage rating of the diode and you'd better not exceed it, even for a short period or you can destroy the diode." This is true for ordinary rectifier diodes, but there do exist diodes which don't have that problem.
 

studiot

Joined Nov 9, 2007
4,998
Quote:
Originally Posted by studiot
It is never a good idea to size component to close to their limits.

Quote:
Originally Posted by studiot
Electrolytic Capacitors are designed to operate at or near their rated voltage, and do not work well at a much lower voltage

There seems to be a bit of a contradiction there.
So do you consider this correct or incorrect (apart for my atrocious spelling, for which I apologise)
 
The quote you've posted contains several sentences. About what exactly are you asking me asking me my opinion as to correctness?

Are you asking me if I think there's a contradiction in there? Or, are you asking me if I think your advice about electrolytic capacitors is correct?
 

studiot

Joined Nov 9, 2007
4,998
The quote you've posted contains several sentences. About what exactly are you asking me asking me my opinion as to correctness?
You are the one who took several of my sentences, related to different matters and appended a snide remark.

So I am asking you to explain yourself.

Are you, for instance, suggesting that it is good practice to take a 350 volt filter capacitor and run it in a 35 volt power supply?
 
You are the one who took several of my sentences, related to different matters and appended a snide remark.

The two sentences I referred to are not related to different matters. You said:

It is never a good idea to size component to close to their limits.
Capacitors are components. Then you said:

Electrolytic Capacitors are designed to operate at or near their rated voltage, and do not work well at a much lower voltage.
The contradiction is obvious. You could have said "It is never a good idea to size component to close to their limits, except for capacitors."

I don't know why you consider my pointing out the contradiction to be a snide remark. It was the first thing I noticed when I read your advice about capacitors because it came so soon about the first advice about not running components too close to their ratings.


So I am asking you to explain yourself.
I have explained why I commented about the apparent contradiction.

Are you, for instance, suggesting that it is good practice to take a 350 volt filter capacitor and run it in a 35 volt power supply?
I made no suggestion whatsoever about whether your advice about selecting voltage ratings for electrolytic capacitors is right or wrong. I simply pointed out the contradiction.

But now I will outline my beliefs on the matter. I generally believe what the manufacturers tell me. Here is a good manufacturer's document:

http://www.cde.com/catalogs/AEappGUIDE.pdf

On page 16 they say:

"Aluminum electrolytic capacitors made with formation voltages at least 35% higher than rated voltage and with rated temperatures of 85 ºC or higher, don’t require much voltage derating. In applications operating at less than 45 ºC no derating is needed, and with up to 75 ºC, 10% is sufficient. For higher temperatures and with high ripple current, 15% or 20% is appropriate. Since operating life continues to increase for further derating, military and space applications use 50% voltage derating. "

They don't recommend much derating in general, although the last sentence suggests that 50% derating may be helpful where extreme operating life is desired.

I've seen it recommended on this forum and others that 50% derating is a good general practice, but I don't agree. If one uses a capacitor with double the actually necessary voltage rating, the capacitor will have generally have a higher ESR than one of the necessary rating. Also, the case size will probably be larger.

Usually when this question arises on the forum, the person asking is a hobbyist and will be using the capacitor at an ambient temperature much lower than 85°C. This engineering paper from CDE:

http://www.cde.com/technical-papers/multipliers.pdf

has a graph (Figure 2) showing the increase in life with decreased ambient temperature. The graph only goes down to 55°C, but it's not hard to see that at a normal room temperature ambient the lifetime is increased by a factor of much more than 10 times--probably 100 times at 20°C. Since these capacitors are rated for typically 5000 hours at 85°C, a 100 times increase in life would be 500,000 hours. This is such a long lifetime that it's probably not worth increasing more by derating the applied voltage.

But, let's see just how much increase in lifetime we might expect to get by derating the applied voltage by 50%. Figure 1 gives curves; CDE uses the black curve. According to that curve, we could expect a little less than a 3 times increase in life from the 50% derating.

One should balance the merits and demerits. The life at an ambient where a hobbyist will be operating the capacitor is already so long that the project will go into the garbage before the capacitor wears out, even without derating the voltage. Derating the voltage 50% will give a life increase that will not be noticed, but negative consequences will most likely be increased ESR and case size.

I have seen it said that operating a capacitor at half its rated voltage may, after a long time cause it to reform somehow, perhaps changing its capacitance. I know of no authoritative references concerning this effect.

So, the short answer is that I don't recommend derating voltage ratings of capacitors by more than is necessary to account for actual operating voltage extremes, and the manufacturer's recommendation at high temperatures. For example, line voltage may be high sometimes, and the engineer should research this issue.
 
Top