Review of Electrical Noise on 120/240 volt Power Systems. Dec 12,

Transcription

1 Review of Electrical Noise on 120/240 volt Power Systems Dec 12,

2 The electrical noise this investigation covers are the frequencies other than 60 Hertz that exist on a typical secondary power system. Dec 12, 2006 cforster@phasorlabs.com 2

3 To measure the electrical noise some people are using a high pass filter to suppress the 60 Hertz and low order harmonics, leaving only the higher frequencies to pass through to the filter to the measuring device. Dec 12, 2006 cforster@phasorlabs.com 3

4 The filter is constructed and connected as follows: 0.01 uf 0.01 uf 120 volt ac input 10K 10K 10K 10K Output to meter 0.01 uf 0.01 uf Dec 12,

5 Response of the filter to Sinusoidal and Non-linear Voltage Waveforms (In other words compare what goes into the filter and what comes out.) Dec 12,

6 Looking at a broad range of frequencies... Graham Filter Response - 5,000 to 60,000 Hertz P-P with Sq. Wave Input crms with Sq. Wave Input crms with Sine Wave Input 200.0% 180.0% 160.0% Percent of Input Voltage 140.0% 120.0% 100.0% 80.0% 60.0% 40.0% 20.0% 0.0% 0 10,000 20,000 30,000 40,000 50,000 60,000 Frequency in Hertz Dec 12, 2006 cforster@phasorlabs.com 6

7 Looking at a narrower range of frequencies... Graham Filter Response - 1,000 to 20,000 Hertz P-P with Sq. Wave Input crms with Sq. Wave Input crms with Sine Wave Input 200.0% 180.0% 160.0% Percent of Input Voltage 140.0% 120.0% 100.0% 80.0% 60.0% 40.0% 20.0% 0.0% 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 18,000 20,000 Frequency in Hertz Dec 12, 2006 cforster@phasorlabs.com 7

8 Looking at just above 60 Hertz and common harmonics... Graham Filter Response to 1,000 Hertz P-P with Sq. Wave Input crms with Sq. Wave Input crms with Sine Wave Input 200.0% 180.0% 160.0% Percent of Input Voltage 140.0% 120.0% 100.0% 80.0% 60.0% 40.0% 20.0% 0.0% Frequency in Hertz Dec 12, 2006 cforster@phasorlabs.com 8

9 What do the response charts show???? Using peak to peak output meters with very distorted input voltages results in values that are disproportionately higher than they should be. Using true RMS output meters with very distorted input voltages results in values that are disproportionately lower than they should be, plus the spiking produced by the filter prevents reliable readings by a true RMS meter. Using peak to peak output meters with normal noisy input voltages results in values that are about 2.8 times higher than RMS values. This is expected. Using true RMS output meters with normal noisy input voltages results in values that are acceptable. Dec 12, 2006 cforster@phasorlabs.com 9

10 Lets take a look at what the filter is trying to measure.. Dec 12, 2006 cforster@phasorlabs.com 10

11 This is what the input voltage from my home looks like. The fuzz is the noise on the 120 volt ac voltage. One cycle of 60 Hz is shown. Lets look at a section of the waveform about as long in time as this line is wide. Dec 12, 2006 cforster@phasorlabs.com 11

12 If we look a little closer. This is high frequency noise riding on the 60 Hz waveform. The frequency of this noise burst is 30,000,000 Hz. The magnitude of the noise is about 2,000 times smaller than the 60 Hz. Dec 12, 2006 cforster@phasorlabs.com 12

13 What meter should you use to measure the noise??? Graham Filter Response - 5,000 to 60,000 Hertz P-P with Sq. Wave Input crms with Sq. Wave Input crms with Sine Wave Input 200.0% Percent of Input Voltage 180.0% 160.0% 140.0% 120.0% 100.0% 80.0% 60.0% Section of frequency range measured by typical DVM 40.0% 20.0% 0.0% 0 10,000 20,000 30,000 40,000 50,000 60,000 Frequency in Hertz Dec 12, 2006 cforster@phasorlabs.com 13

14 Now that you can measure the noise voltage on the power wiring. How will that voltage affect you? Dec 12,

15 Since most people do not contact the 120 volt hot wire and the neutral wire intentionally.. Dec 12,

16 And capacitive coupling will not transfer a significant amount of high frequency energy.. Dec 12, 2006 cforster@phasorlabs.com 16

17 About the only way for you to personally experience this noise voltage is by radiation of the noise from the house wiring. Let s break down the energy that could be radiated from the house wiring and look at it a section at a time. Dec 12, 2006 cforster@phasorlabs.com 17

18 For frequencies from 60 to 3,000 Hertz... The filter blocks the voltage to the meter. This must not be a frequency range of concern. This is the frequency range of 60 Hertz harmonics from #1 to #50. Dec 12, 2006 cforster@phasorlabs.com 18

19 For frequencies from 200 to 20,000 Hertz... These are audio frequencies. These frequencies are in the same range as those on your Hi-Fi speaker wires. These frequencies do not radiate off wires very well and have never been considered a concern. Dec 12, 2006 cforster@phasorlabs.com 19

20 For frequencies from 20,000 to 300,000 Hertz... As you get above 60,000 Hertz radiation from a wire will occur. There are several international standards on radiated energy that do not consider this frequency range a range of concern. Dec 12, 2006 cforster@phasorlabs.com 20

21 For frequencies from 300,000 Hertz and above... There are several international standards on radiated energy that do consider this frequency range a range of concern. There are also very good standards for the level of acceptable radiated energy in this frequency range. Dec 12, 2006 cforster@phasorlabs.com 21

22 What range of frequencies exist on the power wiring???? The next slide shows the output of a spectrum analyzer connected to the 120/240 volt power system at my home. The analyzer trace shows magnitude levels as an increase in vertical deflection and the horizontal span shows the frequency range measured. Dec 12, 2006 cforster@phasorlabs.com 22

23 This is a scan from 10,000,000 Hertz to 1,800,000,000 Hertz FM Stations TV Stations Cell tower Dec 12, 2006 cforster@phasorlabs.com 23

24 This is a scan from 1,000,000 Hertz to 10,000,000 Hertz AM radio Short wave radio and electrical equipment noise Dec 12, 2006 cforster@phasorlabs.com 24

25 This is a scan from 100,000 Hertz to 1,000,000 Hertz Electrical equipment noise AM radio Dec 12, 2006 cforster@phasorlabs.com 25

26 This is a scan from 10,000 Hertz to 100,000 Hertz Nearby computer and power supply Electrical equipment noise over entire range Dec 12, 2006 cforster@phasorlabs.com 26

27 So, there you have it. There are lots of frequencies on your power system wiring. If you do not touch the wires, your not going to sense conducted energy. If you stand near the wires you may pickup radiated energy, but how much energy is really there? Dec 12,

28 This meter measures how much. Anything in the air will be recorded and compared to recognized international standards for un-controlled environments. (That s for people like you) Dec 12, 2006 cforster@phasorlabs.com 28

29 But the levels are well below the level of any reasonable concern. Dec 12,

30 Lets review what we discussed.. The filter lets you measure the electrical noise in the frequency range above 60 Hertz. You will not transfer significant energy to yourself by conduction or capacitive coupling. The radiated energy you will absorb is not significant. Dec 12,

31 What is the concern??? What are harmful levels to the body? How do they reach the body? How can they be measured on the body? These are important unanswered questions. Dec 12,

32 New Topic Dec 12,

33 How about adding capacitors to the 120/240 volt ac power system? Adding capacitors across the 120 volt circuits DOES reduce the electrical noise in the immediate area of the capacitors. But not without a price for safety. Dec 12, 2006 cforster@phasorlabs.com 33

34 Some dangers are: 1. If a person purchases an electrolytic capacitor at an electronics shop and applies 120 volt ac power to the unit, it will explode like a small hand grenade. 2. Capacitors not protected by individual fuses stand a risk of violent rupture when they fail internally. Dec 12, 2006 cforster@phasorlabs.com 34

35 Some dangers are: 3. Capacitors not protected by individual discharge resistors will maintain a significant charge for a considerable period of time. This can cause quite a shock for the unsuspecting person that touches the units. 4. Capacitors not in a protective enclosure are dangerous from a shock hazard and from a rupture hazard. Dec 12, 2006 cforster@phasorlabs.com 35

36 Some dangers are: 5. Capacitive filters should be engineered and be listed by UL or other safety testing laboratory. Construction of filters by nonelectrical persons is dangerous. 6. Adding excessive capacitors to the secondary of a home or farm distribution transformer will cause over-voltage on the 120/240 volt system during night time hours that could damage other home electrical equipment. Dec 12, 2006 cforster@phasorlabs.com 36

37 Some dangers are: 7. Adding excessive capacitors to the secondary winding of a distribution transformer may cause resonant circuits that will damage electronic equipment, especially variable speed drives. 8. Inappropriately applied capacitors will rupture. The rupture may cause flying metal from the case or heat adequate to start a fire. It is only a matter of time before this occurs. Dec 12, 2006 cforster@phasorlabs.com 37

38 End Dec 12,