How HUMplug-e ground loop solver works ?

No but is sounds as if the HUMplug has a circuit similar to the one you found...except it may have a 15 amp bridge rather than a 35 amp one.

The basic idea is that the ground loop is "busted" with a 10 Ohm resistor. (That may or may not actually cure a hum problem, depends on the sensitivity of the amplifier circuits involved) It does, however, provide a reasonable ground path for static charges and RF filter leakage.

To "short" RF noise to ground a capacitor is placed across the 10 Ohm. This will not take care of frequencies in the "hum" range.

Finally a heavy current bridge is connected in such a way that it passes current both directions. This might seem to be short that is placed across the 10 Ohm resistor, but the truth is that these diodes are non-linear devices. Their incremental impedance in response to small signals depends upon the DC current through the device. Look at the slope of the Diode forward bias curve. That it flattens out near zero current is the key. But if there is a ground fault, and heavy current occurs, the diodes pass the current with minimal voltage drop and allow the tripping of breakers and fuses.

The device is "legal" so long as it's just plugged in to an outlet (like the HUMplug) but as for "safe", that is another question! My personal feeling a 35 Amp bridge is too small and a 15 amp bridge is dangerous. You want the bridge to be large enough to trip breakers and blow fuses. I'd use a 100 amp bridge for circuits fused in the

30-50 amp range (as many are). Most diodes take a pretty healthy pulsed current, and failing open is more rare than shorted, but still you should always remember the rule: SEMICONDUCTOR DEVICES MAKE MUCH BETTER FUSES THAN FUSES! You obviously don't need a heat sink of the kind that might be needed to handle a continuous 100 amps but the higher the current (and voltage...you don't want the diodes failing from a high voltage spike either!) rating the better. You may have problems finding a bridge module over 50 amps, but four 1000 volt 100 amp stud diodes are not big deal.

page

" Simply plug your equipment through this, to cure it of Ground

to

at

Could it simply be a ferrite-based toroidal choke placed in series in the ground lead? It would safely maintain the safety continuity, but the high impedance would reduce the ground currents.

H. R.(Bob) Hofmann

The ferrite choke would be "high impedance" only at high frequencies. It would do nothing for typical ground- loop currents.

Yes, I wonder how "high" an impedance he thinks a small toroid is going to be at 50/60 Hz!

MrT.

With suitably selected big ferrite core and more than few turns it is very well possible to get millenries even at 50 Hz. For example 10 mH inductance at 50 Hz is around 3 ohms impedance. This value will definately reduce ground loop currents on normal situations, but would it be then too much to make the grounding connection not good enough for electrical safety.

So my question is would any impedance value that is well effective for solving round loop problems too high to be added to the safety ground connection (would not be safe in rela life and/or woudl not meet safety regulations) ?

I have made ground loop solving products that have been connected audio and/or video lines. I have made many experiments on different constructions. For signal lines I have tried different ferrite material, toroidal cores (simular as used in toroidal mains transformers), E-cores etc.. Different contructions have their all good and bad properties. With suitable ferrite selection is possible to get inductances that are high enough for solving ground loop problems on signal lines. The downside is that materials with what I can get high inducances with reasonable number of turns tend to saturate quite easily.

--
Tomi Engdahl (http://www.iki.fi/then/)
Take a look at my electronics web links and documents at 
http://www.epanorama.net/

Uh, ignoring core losses and other parasitics, you'll get whatever inductance you get inductance you get at ANY frequency. You don't get "millihenries at 50 Hz", you get millihenries.

Yeah, and to get 10 mH, you'd need a LOT "more than a few turns." And unless you wound it with sufficiently large-gauge wire, you'd have a fair amount of resistance as well.

Why you'd want any appreciable impedance in a safety ground could be a puzzle to some.

In alt.engineering.electrical snipped-for-privacy@cartchunk.org wrote: | On Jun 19, 6:44 am, Tomi Holger Engdahl | wrote: |> "Richard Crowley" writes: |> > wrote ... |> > > Could it simply be a ferrite-based toroidal choke placed in series in |> > > the ground lead? It would safely maintain the safety continuity, but |> > > the high impedance would reduce the ground currents. |>

|> > The ferrite choke would be "high impedance" only at high |> > frequencies. It would do nothing for typical ground- |> > loop currents. |>

|> With suitably selected big ferrite core and more than few turns |> it is very well possible to get millenries even at 50 Hz. | | Uh, ignoring core losses and other parasitics, | you'll get whatever inductance you get inductance | you get at ANY frequency. You don't get "millihenries | at 50 Hz", you get millihenries. | |> For example 10 mH inductance at 50 Hz is around 3 |> ohms impedance. | | Yeah, and to get 10 mH, you'd need a LOT "more | than a few turns." And unless you wound it with sufficiently | large-gauge wire, you'd have a fair amount of resistance | as well. | | Why you'd want any appreciable impedance in a safety | ground could be a puzzle to some.

If they don't know about the ground loop problem. The metallic interconnect between equipment is what makes this so hard. There is the loop being formed.

Maybe more effort needs to be put into why it is that the ground gets so much current induced into it.

--
|---------------------------------------/----------------------------------|
| Phil Howard KA9WGN (ka9wgn.ham.org)  /  Do not send to the address below |
| first name lower case at ipal.net   /  spamtrap-2007-06-19-1238@ipal.net |
|------------------------------------/-------------------------------------|

There are many ways to induce current.

The one thing thats clear to me is, providing some RFI filtering to the ground along with hot and neutral, is good.

Who is going to buy one of these hum gadgets and find out whats inside?

greg

If they're smart, they potted it. Likely the only way to prevent easy reverse-engineering (and subsequent cloning in some cheap far- east factory.)

Only if the core is air, or vacuum.

The permeability of any ferromagnetic core varies with frequency, hence the inductance of a coil wound on such a core will vary with frequency, too.Some ferrites would be totally useless at 50 Hz. Laminated silicon iron stampings would work at 50 Hz and be no good at, say 100KHz.

This is different from "core losses", which are resistive.

The *apparent* inductance will be affected by winding capacitance as well. Above its self-resonant frequency, an "inductor" looks like a capacitor.

There's nothing wrong with Tomi's statement.

--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
                                             (Stephen Leacock)

I am glad I had at least a little support for my inductance comments. The increase in resistance of the safety ground due to the wire itself, and the increase in impedance at 60Hz for the safety ground are not desirable, but unavoidable consequences of adding impedance to take care of the currents at 60, 120, 180 Hz and higher harmonics (assuming we are talking USA standard 60HZ supply voltage).

H. R. (Bob) Hofmann

I wasn't really hazarding any sort of guess as to how the thing "works". My best guess, if pressed would be that it doesn't.

The only way to be sure is for somebody to stump up the ninety bucks or however much, and do some measurements on input and output. Whatever else it is, it's a four-port network, so it shouldn't be too hard to come up with an equivalent circuit from measurements.

I smell snake oil.

--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
                                             (Stephen Leacock)

Garter Snake, or King Cobra?

--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

snip

Likely a common rattler, I guess...

-- "Electricity is of two kinds, positive and negative. The difference is, I presume, that one comes a little more expensive, but is more durable; the other is a cheaper thing, but the moths get into it." (Stephen Leacock)

If a common rattler had any snake oil, it wouldn't rattle! ;-)

--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

That's because the snake oil salesmen took all the oil :-)

As an aside: I've only once run over a rattler, but I made damn sure that I could see it on the road, in the rear view mirror afterward. I don't know whether those stories about rattlers getting stuck up inside fenders are true, or just legend, but I sure as hell wasn't going to find out the hard way, changing a tire.

--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
                                             (Stephen Leacock)