|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
98.217.204.116
How large can the capacitance be in a DIY parallel AC filter?Unfortunately, a person can touch the metal poles on the male AC power plug. That causes safety issue, I think.
Suppose somebody unplugs parallel filter at the moment when the capacitor inside is at peak voltage: 120 X 1.414 volts = approx 170 Volts. (240 X 1.414 = approx 340 Volts in the UK)
If the capacitance is large, the capacitor holds large amount of electrical charge at that not-so-safe voltage. The energy stored in the capacitor is (1/2) * C * V^2.
If somebody touches the metal on the unplugged male power plug of that parallel AC filter, then the large amount of electrical energy is dissipated through his/her finger. Isn't it dangerous?
Is there any circuit which disconnects the capacitors at the moment it detects unplugging of the power cord from the wall outlet?
Edits: 11/28/14 11/28/14Follow Ups:
The magnan cable website. Interesting. There are many who look at these caps as modifying the power factor of the affected line.
I googled, but could not find magnan cable home page.
I guess because caps were used for power factor correction for motors and inductive loads. Caps work for that. But the kind of power factor problems that audio gear presents (anything with a power supply basically) is a different kind of problem - not the relatively simple issue of the current and voltage being just a bit out of phase, but the problem is that the current is drawn in short bursts, rather than uniformly over the whole cycle (non-linear). Caps can't fix this problem.
The line itself doesn't have a power factor, as power factor is a function of a whole circuit, and revolves around how the load consumes power.
"the problem is that the current is drawn in short bursts, rather than uniformly over the whole cycle (non-linear). Caps can't fix this problem. "
I think large caps can reduce the effect from short burst current draw: some of the current is drawn from the cap. But I do not know how a large cap in an external box AC filter can be safe for human beings.
I was talking about power factor correction. And caps can't fix this. In fact the caps that follow the diodes in a power supply are what *cause* the power factor problem; the bigger the caps, the worse the power factor (roughly speaking), as the bursts that the AC is higher voltage than the caps (and hence charging them) is shorter.
The bigger the DC caps, the worse the power factor.AC caps, however, can make the power factor better.
Reducing THD in the AC waveform, caused by current surge by AC to DC converters, is also power factor correction in broad sense.
"The capacitance be in parallel AC filter" in my original post was the capacitance of AC capacitor.
Edits: 12/13/14
I maintain that putting caps across the AC line is not really going to affect the power factor a whole lot. Not without an inductor there as well. Sure, a little bit. But as the calculations above have shown, you can't really have a very large cap directly across the AC, ~100uF max thereabouts. This isn't typically going to affect the PF much.
Actually this is something I'd been thinking about for some time : why does nobody make audio gear with active power factor correction? (Correct me if I'm wrong) Passive correction is not very efficient, even when done properly with inductors and capacitors you only normally get up to 0.75 or thereabouts.
Active correction can get you a PF up to almost 1. But as with most engineering, there are tradeoffs, the most significant of which for us audiophiles is the EMI that such switching devices produce. Look up "duel stage active pfc" for example.
MIT made a z terminator plc. Quite effective though analysis was difficult because boards were covered in black resin. I believe it was a progressive PF corrector.
What is a progressive PF corrector?
OK I admit I'm guessing. Z stabilizer Unit seemed to have three parallel boards. Exact values unknown because of the black resin wash. Worked very well and I still use mine. Plug in anywhere but works best closest to components. Doesn't seem to interact with PLCs. No visible inductors but resisters and assorted . curiously plugging in unit occasionally causes a small spark at the duplex. No large visible caps though not semi conductors
Sound sort of smooths out. Quieter background ,better micro dynamics with better detail. All in all better musical control of system
Of course YMMV
What is a progressive PF corrector?
There are several points of different sizes for different mile-posts.
Step one: stay within UL and other Safety agency requirements for leakage currents, and avoid any issues with insurance being denied, etc.
This would be some where on the order of 0.01 uF (10,000 nF).
This value would also avoid tripping GFI type outlets, and so is a safe bet for universal usage.
Step two: a reasonable amount of capacitance that is not decidedly dangerous, but still violates UL, etc.
This would be from several to 1 uF of capacitance, and should only be used if you know the consequences of doing so.
Step three: so much capacitance that the current draw is enough to trip the breaker when the equipment is turned on.
This is going to vary due to how much current the equipment plugged into that outlet draws, steady state as well as turn on surge.
With no equipment plugged in or turned on, the maximum capacitance is approx. 300 uF, which will draw approx. 15 amps at 120 VAC.
So if the equipment draws 7.5 amps, then the total capacitance can not exceed approx. 150 uF.
Going to the extreme of step three is NOT recommended or any sort of a good idea, step two is risky enough.
Jon Risch
(Jon), I'm told by an electronics engineer that there is no path for a shock with caps across the ac but one should stay at 4700p from neutral to ground and hot to ground. So, it's ok to go .01 as apposed to 4700??? Tweaker
" 0.01 uF (10,000 nF)."
Should read " 0.01 uF (10nF or 10,000pF)."
Always place a high value resistor across large capacitors to bleed off any residual charge.
I think you want to know how small a capacitor should be in a filter. The idea is to get as close to a short to take the noise to ground reference.
Why not try a search for capacitive reactance and see if this better fits your needs.
The extreme in the direction you mentioned would be no extra capacitor at all. Just the parallel capacitance of the power cord remains.A company called AudioPrism makes parallel filters. Its more expensive parallel filter (
ACFX) has larger case than its less expensive parallel filter (Quietline).So I guess the more expensive filter has more capacitors inside, i.e. larger parallel capacitance value.
Of course, large capacitors are slow and thus need "small" bypass capacitors.
Edits: 11/29/14
nt
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: