|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
173.75.242.249
I have been dabbling in main line conditioning for a while now, but very much following a monkey see, monkey do approach. I am somewhat electrically “challenged”
Could somebody explain to me what the difference is between using inductors as to common mode chokes for filtering, and under which conditions one would be preferable. On the vast majority recipes I have come across the recipe has always followed the following format.
(I'm keeping it simple, many recipe's include MOV's and caps across the neutral live and live to ground and neutral to ground)
1.) capacitor(s) – common mode choke – capacitor(s)
or
2.) capacitor(s), -- pair of inductors, – capacitor, capacitor(s)
I have tried both. Both seem to work great and to be honest I have not been able to discern a difference between using a CMC or a pair of inductors.
Is there a benefit in using both?
Thanks
Cleet.
Follow Ups:
A "standard" inductor stops differential noise a common mode inductor stops common mode noise.
Thanks for the links
If I understand it correctly its seems that
1.) inductors will lower differential noise AND common noise given a cap-inductor-cap design, BUT inductors cause problems of their own; they distort the wave forms.
2.) A CMC is kind on the wave form, but it will not reduce differential noise.
Maybe I use both? Get get inductors and a CMC with low DCR and do the following
cap(s) -> inductor -> cap(s) -> CMC -> cap(s)
Thanks
Cleet
RE #1, inductors don't inherently distort the wave form, only if the cores are saturated by any given current draw.
When building audiophile grade equipment, it is relatively easy to utilize inductors with cores substantial enough to not cause such problems.
This is why I specify a specific set of part numbers/manufacturer's for the series inductors in my DIY AC Line Filter.
If saturation is avoided, then having a pair of independent (NOT common mode connected) inductors assures the lowest differential and common mode interference feed through.
RE #2, the big deal with a CMC is that the core can be made much smaller, and thus save on costs. There is absolutely no strictly technical/performance benefit to the use of common mode chokes in an AC line filtering circuit, just a cost reduction aspect.
This is why virtually all commercial/retail filters use CMC's, since the costs are the lowest.
If you use separate independent inductors with a large enough current rating, then you don't need a CMC for any reason.
One thing that CMC's do tend to encourage, is the use of a VERY large inductance in the filter circuit. However, what a lot of people overlook is that placing 1 mH in the AC line actually degrades the peak current delivery due to the sheer AC reactance at the 60 Hz powerline frequency, and the higher harmonics that represent the true current draw waveform.
A 1 mH CMC will have an AC reactance of approx. 0.38 ohms at 60 Hz.
Nevermind the DCR of 0.05 ohms, all for naught as the reactance dominates.
Thus the current surges needed to replenish the power supply caps are being limited by this relatively large amount of AC reactance due to the inductance. With these types of current surges, they are a fraction of the total AC line sine wave, and the fundamental rep rate is 120 Hz, with the fundamental frequency of these current surge blips somewhere above 1,200 Hz. The reactance of a 1 mH CMC at 1,200 Hz is approx. 7.6 ohms.
Now you are really limiting the current surges.
Now I hope it becomes clear that the use of such a large inductance is counter-productive in terms of musical dynamics.
Thus, the use of a much more reasonably sized independent and separate pair of inductors does not require a large and costly core with a lot of windings. You don't really want it to be that large of an inductance to begin with.
Jon Risch
Same as your design but say 100 to 120 watts. See any problem or reason not to?? I just met someone who described a filter that took 120 up to 240 and then down. The opposite ?? If you think this is a good idea( whole system) can you suggest the trannies? Thanks, Tweaker
1st, 100-120W is not enough to adequately run an entire system. There is some losses with the down and up conversion of the voltage, so if you had 100W units, and a filter cap in between the two transformers, you might only end up with a total capacity without significant line droop of 80 w.
Even a HTIB would need more power than that to operate properly on peak output.
2nd, my original design was stepping down the voltage so that a relatively massive amount of capacitance could be placed across the line for filtering, and yet avoid the usual penalty of a lot of leakage current being present due to the cap.
Stepping up the voltage would require a higher voltage rating cap, and provide less safety margin for that cap.
Step down transformers are much more readily available, and much easier to wire up safely, etc.
If one were to try to take this simple quick and dirty technique to it's maximum capability, then a pair of one to one (120vac IN, 120vac OUT) E-I core power transformers with a rating of 15 amps at 120VAC would be the way to go, or perhaps as a close second, a pair if identical 500 to 800 watt rated transformers with a 50-60 VAC secondary would be OK. How OK would depend on the exact details of your system with regard to the power demands and requirements of the total sum of the equipment.
Neither one of these would be a cheap purchase, and with such large units, there would need to be adequate distance between the two transformers so there was no magnetic coupling between the two. I would say with proper orientation, they would need to be at least 6-8" apart.
Jon Risch
The caps are one coil away from the ac source so do these filter at all at the ac source?? Tweaker
The main idea of this type of filter, is to isolate the load from the line, and the line (and everything else on it) from the load.
What you are asking is, does the filter provide any filtering directly to the AC line, therefore providing some benefit to other loads on the AC line.
The answer is yes, sort of. And, it depends.
One reason I use EI core transformers, is that they tend to filter out higher frequencies on the AC line, and this works well towards the original goal of this type of filter. However, it also makes the capacitance that is placed across the secondary less effective, because it is not coupled back onto the AC line as well.
So the more HF losses the transformer has, the less effective the capacitance will be in terms of directly filtering the AC line itself.
Just remember, that once you start placing capacitors directly across the AC line, that once a certain size is reached, you have exceeded the allowable leakage current according to the various Safety Agencies.
This leakage current can trip GFI type outlets, which at the least, is a nuisance and bother.
The capacitance that will cause the leakage current limits to be exceeded is not that great, just a small fraction of a uF.
Jon Risch
Dosn't the step down feature have something to due with it's filtering? Maybe not if you think equal transformers will work. Lots of caps is a very good feature. Can you hear these small transformers buzz from more than a few feet away?? Tweaker
I use a lot of Corcom type modules, EV/ER series, medical grade.
They are made in various current handling modules ,from 3 amps to 20 amps. I like the smaller modules. They offer more noise reduction but you have to be careful to select the right module for your component. Using PS Audio recommendation, I allow 5 times fuse value of each component per PLC. I find divide and conquer works the best sonically.
of courseYMMV n FWIW
"One thing that CMC's do tend to encourage, is the use of a VERY large inductance in the filter circuit. However, what a lot of people overlook is that placing 1 mH in the AC line actually degrades the peak current delivery due to the sheer AC reactance at the 60 Hz."
Jon, I think you got 'er backward. The AC power doesn't "see" the CMC because the current is differential so it can be most anything without affecting power delivery.
Regards, Rick
Nope. The AC line "sees" the full inductance of the CMC, what you are thinking of is the inductor CORE of a CMC, it sees only the differential current, which is only present when there is a fast spike coming in on the H or N line. Normal AC current draw by the load unit "cancels out" in the CMC core, this is where it can "get by" with a much smaller core than if it were an independent inductor.
The inductance does NOT cancel out, but is effectively increased by the common mode wiring of the inductance elements.
Unfortunately, when a differential voltage spike like this does occur, the spike generates a reverse polarity spike in the other winding of the CMC, and this essentially DOUBLES the spike voltage the load unit sees. With independent inductors, this does not occur.
Since I have measured actual AC line operating conditions as a result of my design time with the Discwasher SpikeMaster surge protector and filter, I have seen how this can actually occur fairly often, usually due to less than perfect AC wiring conditions, and lopsided induction of a transient signal onto/into the AC line.
This is exactly the reason I did not use a CMC type inductor in my AC Line Filter and Surge Suppressor design, and one of the reasons it keeps the output so clean and free from extraneous noise and voltage spikes.
Jon Risch
"The inductance does NOT cancel out, but is effectively increased by the common mode wiring of the inductance elements."
Inductance cancels for differential mode current and increases for common mode current passing through the common mode choke. That is why they are useful. Nothing is perfect, however.
Sorry, yes I have swapped the concept of typical common mode wiring for common mode current, which is incorrect. Been working late a lot, shouldn't post when I am tired.
However, the bottom line is the measurements I took indicates that a voltage spike or surge does NOT always flow through the CMC equally, in fact, more often than not, the return flow does not occur due the the load absorbing the voltage spike, so it does not fully get through to the return leg of the CMC, and the end result of the voltage spike doubling I spoke of does occur.
Jon Risch
Thanks for the comments Jon.
I have been using Bourns 5520-R's and I have use one pair per outlet (i.e. per device) in an attempt to isolate outlets from one another. Nothing I have component wise that have I measured exceeds 1A draw. So I believe I am not even close to saturating the cores.
Jon, I am planning to use a combination of your cap, inductor, cap recipe and you two trans recipe line filters.
Which sequence, i.e. the inductors upstream from the transformers, or transformers upstream from the inductors, will yield best results?
I'll be balancing the power on the second trans and I'll be using the filter for a phono amp only which draws < 0.1A
Again thank you for sharing your knowledge.
Cleet.
That subject has been discussed before, please do a search for those discussions.
Jon Risch
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: