How safe are Ebtech Hum X devices?

Janus Alfador

Member
Messages
723
What is the danger involved in these, if they fail?

Here's a write-up of the Hum X' design by somebody who took one apart: https://groups.google.com/d/msg/alt.guitar.amps/CJyWD3GaHfg/V40Y6kgm6XkJ


To surmise what's said about the device's construction:

A pair of 6A/1kV diodes between the wall ground connection and the adaptor's output socket ground. Connected in parallel, with the anode on one diode connected to the cathode on the other.

The pair of diodes has a 1K 1/2-W rsistor in parallel with it. This resistor seems to serve to produce a "correct" indication when an outlet tester is plugged into the adaptor, since without it, the outlet tester will give an "open ground" indication.



In reading about this design, it's possible that it could fail, and fail in a way that the device doesn't shut down automatically, but continues operation. In which case, what occurs? Is then the ground connection lifted, without the hum returning to an amp setup to indicate that the device has failed?

Would these devices be likely to fail at some point over a few decades?
 

pdf64

Member
Messages
7,895
There's generally only one safety critical circuit in an amp, and that's the one between the ground of the wall outlet and the amp metalwork / chassis.
This device would appear to put components other than a very low resistance metal conductor of a suitable current carrying capacity into that circuit.
My understanding is that's a really bad idea; any component can fail.
The golden rule is not to f**k with safety critical circuits.
These devices only seem to be licensed / approved for use in North America.
If I lived there I'd be querying the regulatory authorities about that.
I hope Trump doesn't insist that the UK accept a dilution of safety standards that products like this would seem to represent, in order to get a trade deal with the USA.
Apologies to Ebtech if I've misunderstood something; it would be great to hear the designer's safety case for the Hum X.
I accept that the most likely diode failure mode resulting from overcurrent is a short, rather than open, circuit.
 
Last edited:

Kyle B

Silver Supporting Member
Messages
5,102
These devices only seem to be licensed / approved for use in North America.
If I lived there I'd be querying the regulatory authorities about that.
In the US, there is no regulatory authority who has jurisdiction specifically over electronic product safety. That job is taken by UL (Underwriters Lab). We do have the CPSC (Consumer Product Safety Commission), but they don't write or enforce electrical laws. UL is a PRIVATE COMPANY. They have some competition, but they're so well entrenched in the public psyche that most non-EE's think a UL marking is indeed issued by a government agency. There are electric codes for buildings, but not for consumer products. UL is "sorta" like VDE in that regard. It's been this way since 1901 or whenever UL opened for business. I guess they do a good enough job that there's never been a push to 'government-ize' electrical product safety.
So don't worry about Trump -- He doesn't have any say in this ;)

Reading the description of 'how it works', I would propose it's not 'licensed/approved for use only in North America', but rather that there's NFW these would pass the CE review. I guarantee with an impedance in the ground leg, these aren't getting a legitimate UL approval. (Or they shouldn't... Don't get me started on UL's practice of hiring newbs right out of college....)
 

Kyle B

Silver Supporting Member
Messages
5,102
My theory of operation...

Assuming the writeup is accurate (two parallel diodes and a paralleled resistor), I don't agree with the writers explanation of 'how it works'. I don't think that resistor is there to fool ground-continuity testers.

IMO - The resistor is what stops the hum. It is presenting an impedance to ground currents. But since ground currents (specifically those which create hum) are measured in milli-or-microamps, the voltage drop is relatively small... enough to break an annoying interference loop, but not completely disconnecting the ground - The voltage on said resistor should never get high enough for the diodes to conduct.

The diodes are present to ensure safety if there IS a fault in the load. Should the load suddenly start drawing massive current, the voltage on the resistor would rise enough to forward bias one of the diodes (then the other, flipping back & forth 60 times a second). Since the diode would instantly clamp to 0.3 or 0.7V, 'ground' would not take on a high voltage.

BUT --- If you had a diode go open-circuit and there was a ground fault in your load, you absolutely would be exposing yourself to potentially lethal voltages. And you would not know this danger exists until it presents itself --- You can't easily check for diode function with a resistor soldered in parallel with a multimeter. You'd have to keep pumping current in until the diode starts conducting and see if it clamps at 0.7V (if my analysis is correct.....)


Some dude on a random forum said:
t's good to hear from you. Most technicians are aware that disconnecting safety grounds will "solve" most serious ground-loop problems. Usually the voltage differences that drive the ground-loop currents are under a few hundred millivolts, so it's an appealing idea to insert diodes (paralleled in opposite directions because it's AC) in series with the safety ground connection. Because a silicon diode requires some 600 mV before significant current flows, this keeps the circuit open at low voltages. But the catch is that, under fault conditions when this path is critically important to avoid electrocution, these diodes must conduct extremely high currents until the circuit breaker on the branch circuit opens. In order to safely handle such high currents, each of the diodes would need to be the size of a hockey puck!

In my most recent seminars, I mention the HumX in particular when I talk about power-line safety. I've attached the slide used in my Sept 09 seminar for CEDIA. The question "Will it survive ..." is a rather rhetorical one because the tiny diodes inside (rated at around 5 A, judging by their size when we tore one of these units down), would certainly become a puff of silicon vapor should an actual fault occur ... and I certainly won't bet my life that they'll fail in a shorted condition! Based on my years of experience as a repair technician, most rectifiers fail by opening like a fuse under serious overcurrent. IMHO, this unit is a poster child for marketing gone wild and commerce without conscience! For the past several years, I've stopped at their booth and asked about the UL listing ... their answer is "We're working on that." I'm a member of UL's advisory panel for professional audio equipment (UL 1419) and feel sure UL would never grant such an approval.

Beyond the safety issues, this "solution" ... just as disconnection of safety grounds ... doesn't solve the real problem anyway.
Found here:
http://forums.prosoundweb.com/index.php?topic=158888.10
 

Newkular

Member
Messages
1
Apologies for the necro-post, but I thought it useful to share what appears to be the source for the idea behind the product (or, at least the best write-up I've seen for this old trick.) Rod Elliot wrote a great article on grounding/earthing some time ago:

http://sound.whsites.net/earthing.htm

Read the section titled Use of Loop Breaker Circuits and note that the schematic presented is very similar to the Hum X design (if the original post is correct). The article further describes the circuit theory:

The loop breaker works by adding a resistance in the earth return circuit. This reduces circulating loop currents to a very small value, and thus 'breaks' the loop. The capacitor in parallel ensures that the electronics are connected to the chassis for radio frequency signals, and helps to prevent radio frequency interference. Finally, the diode bridge provides the path for fault currents.

The bridge rectifier here is performing the same job as the diodes described in the original post. (For the sake of brevity, just take it at face value that most would prefer the 35A bridge for fault insurance reasons ...)
 

Timbre Wolf

Hyperspatial Gravity Surfer
Platinum Supporting Member
Messages
10,319
In the US, there is no regulatory authority who has jurisdiction specifically over electronic product safety.
This is not entirely true. However, it does point toward some truth, and some challenges with the US electrical safety world.

In US workplaces, the Occupational Safety and Health Administration (OSHA) does regulate electrical safety. And they do reference UL and other Nationally Recognized Testing Labs (NRTL) for electrical safety programs:
https://www.osha.gov/dts/otpca/nrtl/nrtllist.html

Now OSHA does not protect consumers, but safety-savvy consumers can at least protect themselves if they limit purchases to devices passing NRTL approval.

- T
 

TimmyP

Member
Messages
2,488
Having an altered safety ground is never a problem - until it is. At that point you might be dead and won't have to worry about it. Break ground loops in the audio cables, not the mains.
 

HotBluePlates

Member
Messages
11,499
Apologies for the necro-post, but I thought it useful to share what appears to be the source for the idea behind the product ...
Very good write-up!

I would offer that Ebtech's actual "idea behind the product" is:
- "Offer a lower-cost alternative to transformer isolation to solve ground loop hum."​

There is real concern about safety involved in the approach, because it messes with the safety ground being a solid conductive bond from chassis to the ground pin of the outlet. The sole function of that ground pin is to flow so much current in the event of power contacting the chassis that the breaker trips back at the service panel.

What's interesting is the tactic used is identical to how a ground lift is incorporated inside some pieces of gear. In this case, the resistor/cap/diode network is placed between signal/power grounds of the amplifier circuit & the actual chassis ground. This breaks the loop; however, the safety ground always stays bonded to the chassis, even in the no-hum ground-lift position.

The Ebtech obviously works, as it uses the same approach as other cases that work. It is less-safe, even if only marginally less-safe, as it relies on interconnecting cables (that created the ground loop in the first place) to carry ground fault currents. It's not "The Right Way" of accomplishing the task, but I suspect a great many guitarists won't care, and likely will never have a bad outcome from using them.






_______________________

That said, I know someone who never wears a seatbelt. 35+ years of driving & riding in cars, and they never put on the seat belt. Day after day, year after year, they drive and get along just fine without it.

Except that one time ~15 years ago when they got into a serious car accident & smashed their head against the windshield... The resulting head injury caused permanent brain damage that has altered their personality & judgment/impulse control.

Out of hundreds of thousands of times "getting away with it" there was one time that they didn't, and their life is forever altered as a result. They still don't wear a seatbelt and survive just fine (for now), but it doesn't mean it's a good idea...
 






Top