Hum eliminators - what drives their price?

[Moroccomoose]

I am looking to reduce hum from my TC Electronics G-System which is connected in 4CM (Some may remember my earlier thread about psudo balanced signal from a few weeks ago.)

Anyway there is a revered guide for setting up the G-System, acknowledging this problem and suggests a Morley Ebtech Hum Eliminator HE-2 as a solution.

My question is what do I get for the additional £71 from the Morley Ebtech over the £18 'Subzero Hum Destroyer' from Gear4Music?

https://www.gear4music.com/Recording-an ... gJdI_D_BwE

https://www.hotroxuk.com/morley-ebtech- ... GoogleBase

Cheers,

Stu.

Edit: I found this thread - https://www.soundonsound.com/forum/view ... hp?t=92346 - but if anyone can comment specifically about hum eliminators, that would be great.

[James Perrett]

Looking at the descriptions, neither of them inspire much confidence. It is obvious that the marketing arm of both of those manufacturers don't have the faintest idea of what they are talking about. I wouldn't be surprised to find that they are identical internally.

Personally I would prefer to deal with a supplier with a bit more of an understanding of the problem. Orchid have a good reputation

https://www.showbitz.co.uk/product/orch ... -isolator/

as do ART

https://www.soundonsound.com/reviews/art-dti

[Hugh Robjohns]

My question is what do I get for the additional £71 from the Morley Ebtech over the £18 'Subzero Hum Destroyer' from Gear4Music?

Assuming there is a real technical difference (as opposed to just brand marketing and profit margins), it will be down to the quality of the transformer(s).

A better quality transformer can handle higher signal levels with lower distortion (especially at low frequencies). A cheap transformer will saturate at moderate levels and introduce significant distortion... although some might see these characteristics as beneficial, of course!

Unfortunately, the law of diminishing returns applies in transformers, in that getting a little bit better is expensive, and getting a lot better is flippin expensive!

...and only the most reputable manufacturers will quote distortion specs at different signal levels and frequencies.

For example, the ART DTI costs about £70 (two channel line isolator with lots of connectivity), and produces 0.5% distortion at 50 Hz with a +20dBu signal (it's below 0.1% at 150Hz and under 0.01% above 1kHz)... which isn't bad at all. I have five of them!

I also have one Canford Audio two channel isolator fitted with Lundahl transformers that reduces the LF distortion to 0.2% at 50Hz and +24dBu, but it costs around £220.

It's only a small difference on paper, but a big difference to the invoice.

Yes, there is a noticeable difference to the sound quality: the Canford box is much more neutral and transparent compared to the DTI, without a doubt. So I use the Canford box in really critical quality situations, but I use the DTI quite happily for most problem solving situations and no one has ever complained.

Whether they would with a £20 box I can't say because I've never tried (or measured) one !

Instrument levels are much lower than pro line levels which helps a lot, of course.

[Wonks]

The Morley one says "Made in the USA" whilst the G4M one is undoubtedly made in China. That will account for most of the price difference, regardless of the internal components used.

I have an ART DTI box and a Behringer hum isolator that is similar to the G4M unit. I can't tell any noticeable difference between the two, though I did have to ground the case of the Behringer to an outgoing (connected to a guitar amplifier) TS jack sleeve connection, otherwise it hummed badly. Once I did that, it was silent.

[BillB]

This - the price/performance ratio of a number of isolators, with distortion stats and based on the findings of experienced ears - would make a great SOS article.

[cashhewn]

For example, the ART DTI costs about £70 (two channel line isolator with lots of connectivity), and produces 0.5% distortion at 50 Hz with a +20dBu signal (it's below 0.1% at 150Hz and under 0.01% above 1kHz)... which isn't bad at all. I have five of them!

The literature that came with my Ebtech hum eliminator gives its specs as:

Frequency response 20Hz to 70kHz plus or minus 0.5dB
• Distortion less than 0.005% THD @ 1kHz
• Crosstalk better than –97dB
• Maximum source impedance 1kOhm. Minimum load impedance 10kOhm

Interesting that the THD figure is listed without giving the value of the signal (dBu or otherwise)...

I have had no issues with the one I use to balance the group busses of my Soundcraft 600 Series analog console (which I think in retrospect is unnecessary but I've been to lazy to figure out the unlacing them from the patch bay).

[ef37a]

For example, the ART DTI costs about £70 (two channel line isolator with lots of connectivity), and produces 0.5% distortion at 50 Hz with a +20dBu signal (it's below 0.1% at 150Hz and under 0.01% above 1kHz)... which isn't bad at all. I have five of them!

The literature that came with my Ebtech hum eliminator gives its specs as:

Frequency response 20Hz to 70kHz plus or minus 0.5dB
• Distortion less than 0.005% THD @ 1kHz
• Crosstalk better than –97dB
• Maximum source impedance 1kOhm. Minimum load impedance 10kOhm

Interesting that the THD figure is listed without giving the value of the signal (dBu or otherwise)...

I have had no issues with the one I use to balance the group busses of my Soundcraft 600 Series analog console (which I think in retrospect is unnecessary but I've been to lazy to figure out the unlacing them from the patch bay).

I would take those number with a fistful, never mind a pinch of salt. As Hugh says, it is at low frequencies that transformer distortion become most prominent so a figure at 1kHz means little and I would still say it is optimistic even for 1kHz.
Source resistance needs to be low to keep LF distortion down, in theory a zero source R means zero distortion but then there will always be the winding resistance. "Pro" signal sources will be low, 100 Ohms or less but not so "neg ten" kit such as pedals and many synths.

A transformer is always a compromise but I for one could live with a tiny bit more THD at 50Hz rather than an onbvious hum!

Dave.

[Wonks]

And of course if your guitar signal is going into an overdriven amp, even 0.1% THD is hardly going to matter!

[Moroccomoose]

Thanks everyone, good information as usual!

Exactly Wonks, this is for a guitar rig so based the following:

The law of diminished returns in transformer quality,
The fact that it is instrument signal rather than line level
And that I am feeding a guitar amp which will be mostly distorted anyway,

I am going to take a punt on the G4M (or similar) offering - and if that doesn't work, I'll get the ART DTI and I'll have still spent less than getting the Morly Ebtech.

I'll let you know how I get on!

While we are discussing, should the short cables be between amp and isolation or pedal and isolation? I'm guessing short cable between the isolation and amp send/return. Long cable to pedal.

Stu.

[ajay_m]

Out of curiosity, why would two cheap toslink convertors not work for ground isolation? You can purchase a stereo line to toslink (adc) and toslink to stereo line (dac) for about £12 each. Wouldn't this provide ground isolation with vanishingly low distortion figures compared to a transformer based solution? Latency I'm guessing is fairly low,they claim to go to 192KHz I think. Probably 16 bit convertors but still, if levels are reasonable....

[Wonks]

They might, but you've got quite a wide range of signal levels, so it's difficult to optimise the input/outputs, and you need power. Remember this is in a guitar context. A transformer (or two) if a 2-channel isolator doesn't. Why over-complicate things? Signal/noise ratio might be a lot worse as well you are probably 20dB down on peak line level signals at best.

The extra latency can certainly matter if you are using digital effects and digital modellers etc. It all adds up.

Once power becomes available, simple opto-isolators will do the job without latency, but you are still then looking with an enclosure, and something more expensive than the Behringer and G4M options.

[ajay_m]

Good point of course. I was really just thinking in general regarding ground isolation as the topic comes up quite often but indeed in a guitar context you're quite right. In a studio environment though it might be a useful way of breaking a ground loop. With usb induced ground loops the usb1.1 isolators are an amazing solution for up to four audio channels but I didn't have much luck getting a usb2 device working. That technology though is still quite new so with luck they'll get the kinks out of it soon.

The transformer based solutions are great for guitar as you say but distortion rises quite rapidly at low frequencies, not an issue I guess for guitar of course.
Incidentally you say a simple optoisolator could be used. These are pretty non linear though unless you had some kind of pwm based solution using a high frequency device, I would have thought.

[ef37a]

And of course if your guitar signal is going into an overdriven amp, even 0.1% THD is hardly going to matter!

Heh! I agree. I often think all the time and trouble spent on discussing the finer points of electric guitar sound are akin to a "Hare&Horse stew".
Those not familiar with the term just ask.
BTW been trying to reply since about 9am off and on. Just whirling blobs. As usual, only SoS affected.

Dave.

[Moroccomoose]

So the G4M subzero hum destroyer arrived. £20 all in, including postage.

I set it up and certainly the hum is attenuated pretty significantly. 'destroyed' or 'eliminated'.... well I wouldn't go that far! It is definitely an improvement. Definitely quieter than it was, but not as quiet as guitar-amp with nothing the the fx loop.

I think the overall signal level was also reduced, so I find myself turning up the amp levels more to get the same volume - but I have also adjusted the gain structure in the G-System unit too. I've a little more tinkering to do with the gain structure in the G-System, but non-the-less a significant improvement. It can stay!

I'm now thinking about a double-sided sticky foam pad, so I can mount it inside cabinet and a pair of decent matching short TS cables to have it permanently hooked up that are pretty enough to satisfy my mild OCD!

Stu.

[ef37a]

Stu, if you have a branch of Hobby Craft near you they sell a self adhesive Velcro. This stuff is of "industrial" strength, both in terms of the adhesive and the "grippiness" of the fuzz.
For best results clean each surface with alcohol.
For a really, really seriously strong double sided pad try Halfords. They do one for fixing mirrors to windscreens but don't do that if you ever want to get the box off, probably take the Tolex with it!

Dave.

[Drew Stephenson]

Also, before you do fix it, check where in the cab, and what orientation, results in the lowest hum. The place for the shortest patch leads, or otherwise neatest arrangement, might be right next to the big old mains transformer...

[Hugh Robjohns]

I set it up and certainly the hum is attenuated pretty significantly. 'destroyed' or 'eliminated'.... well I wouldn't go that far!

Then something is amiss.

If the original problem was a ground-loop then a line-isolating transformer will break that ground-loop and its unwanted noise will go away completely. There's no grey area of 'partial improvement' — it's either fixed or it isn't.

So something else is going on...

The classic error many make when using a line-isolation transformer is to place it near to a large mains transformer.

Mains transformers (especially in guitar amp's!) generally radiate a strong varying magnetic field. If that interacts with a cheap line-isolating transformer (which won't be magnetically screened because that's expensive to implement) that magnetic field will induce mains-frequency hum into the audio transformer. So you break the ground-loop but induce a new hum instead!

Sometimes you can reduce the induction by changing the orientation of the line-isolation transformer, but the best option is to keep it well away from any mains transformers.

The other possibility is that there are still other ground loops within your system contributing unwanted noises...

[Wonks]

I would take it apart to see if the case is grounded. If like my Behringer, it isn't, then grounding it to one of the outgoing sleeve connections should improve matters.

[Moroccomoose]

Sorry to resurrect an old thread, but this is interesting... While googling around I happened across this service note from TC Elctronics RE: 'Hum sound when connected to lower jacks' (Of a G-System. - The pairs of send/ return connections are made of a row of double jack connectors on the rear panel).

IMPORTANT:
Modification should only be carried out by authorized technicians.
Affected products page 3.
Symptoms:
We have found out that all the double-jack connectors have a chassis
connection that shorts AGND, DGND and CGND when a jack is inserted into the
lower receptacle of the double-jack connector you hear a load hum sound.
Modification:
The easiest way to disable the spring contact is to use a small tool, for
example a small screw driver, to bend the spring in the jack connector, please
be careful not damaging the connector. This modification has to be done on J9,
J7, J26, J6, J5, J4, J3, J2. See fig.1
This “spring”-contact has to be disabled on all Double-jack connectors. It
shorts cgnd, dgnd and agnd when a connector is inserted see fig.2

Fig 1 shows the rear panel, highlighting the necessary connectors for correction, Figure 2 shows a series of images of an 'authorized technician' shoving a little screwdriver in the lower connectors to bend the problematic contact spring.

Being 'un-authorized' to perform this delicate procedure, I was interested in your opinions on whether this seems a reasonable fix for the hum problem I described in this thread?

Stu.

[Wonks]

Presumably analogue ground, digital ground and control circuit ground.

They are probably star-connected at one single point, but the cross connection through the jacks allows ground loops within the unit itself.

It would be good to know if TC ever modified the jack connection circuit board after discovering this, as there may be earlier units that need this, and later units where this mod is superfluous.

[Moroccomoose]

I think 'chassis ground'

They absolutely did fix it in subsequent releases of the pedal. There was a subsequent 'special edition' black version and red version which were specifically marketed as improving noise issues from the original.

[ken long]

Looking at the descriptions, neither of them inspire much confidence. It is obvious that the marketing arm of both of those manufacturers don't have the faintest idea of what they are talking about. I wouldn't be surprised to find that they are identical internally.

Personally I would prefer to deal with a supplier with a bit more of an understanding of the problem. Orchid have a good reputation

https://www.showbitz.co.uk/product/orch ... -isolator/

as do ART

https://www.soundonsound.com/reviews/art-dti

I have a couple Orchids and a DTI and can vouch for their effectiveness.