Fixing Hum/Buzz issues with Schiit Jotunheim Headphone Amplifier

[restorer-john]

How exactly was it creating the hum you measured?

Transformer directly below the aluminium top panel- stray fields most likely inducing eddy currents in it. Those currents likely affected the low level analogue circuitry. By touching the top panel, the panel gets a larger 'virtual earth' and the issue disappears.

 

[amirm]

Transformer directly below the aluminium top panel- stray fields most likely inducing eddy currents in it. Those currents likely affected the low level analogue circuitry. By touching the top panel, the panel gets a larger 'virtual earth' and the issue disappears.

This is my guess too as stated before. The top part of the case bends down and becomes the front panel and gets attached to the volume control where the input of the amplifier is. Hence it is the last part you want to convey the magnetic field.

 

[amirm]

Voltage (push) over a certain resistance is required to produce a flow of current.

No, that is not how it works. Any voltage will induce a current flow or ohm's law would be incorrect: I = V/R. You can reduce the V all you want and the formula will spit out a current as it should. Resistance is a continuous function, not one with a step as you imagine.

Inversely if you know the current and resistance you can compute the voltage as V = I * R. The math I explained is exactly what is used to determine what impedance/resistance you need to have for safety of insulated products. If you have 100 amps and resistance of 0.2 ohms, you will have 20 volts develop across that junction. If you don't have 20 volts, then you can't have 100 amps and 0.2 ohms either.

t is entirely possible for 100V over 1 ohm to draw 100A, however the current is produced by the voltage, not the other way around.

But if you don't know the voltage, you compute it from the other two variables per Ohm's law. Everything we know from electronics falls apart if Ohm's law is incorrect.

 

[Canook]

Huh?
You can’t have current (electron flow) without voltage (the potential which creates the “push”). V certainly equals IR (in a DC, and/or purely resistive AC circuit), and you’re free to solve the equation any way you like. Current flow and resistance don’t make a voltage though, the voltage potential over a closed circuit creates the electron flow- I think we both agree... it’s just silly to say that the current and resistance create the voltage. The resistance exists; the voltage creates the flow of current. Eddy currents induced by the transformer are certainly nowhere near any kind of potential to be a safety hazard. The real “hazard” would be something in the power circuit of the Jot somehow touching the ungrounded case, and then an unsuspecting person (who is grounded) making contact with the circuit. The eddy currents may certainly have been creating the hum... it sounds like you had an unlucky Jot with paint that didn’t get scraped.

It is funny that others are posting about getting rid of their Jot and asking what amp in the same price range to replace it with. I chalk this up to people who are being mislead.

 

[amirm]

Eddy currents induced by the transformer are certainly nowhere near any kind of potential to be a safety hazard.

You are confusing two different issues with respect to this flaw. One was audio performance/hum which had to do with eddy currents and magnetic coupling. The other was how low the resistance of the case needs to be with respect to ground to avoid electric shock. It is in this context that I did the math and hence the 100 amp example. Eddy currents are not 100 amps. The impedance to ground must be very low as to keep the voltage down.

It is funny that others are posting about getting rid of their Jot and asking what amp in the same price range to replace it with. I chalk this up to people who are being mislead.

No one is being misled. Equipment safety standards must be complied with especially when a device has CE mark. Clearly ungrounded enclosure does not comply with such a standard and is in direct violation of the mark. If this had been a mass volume product, a safety recall would have been done. That it is a boutique product lets them get away with it but the device is not compliant for sure. Or safe as shipped.

it sounds like you had an unlucky Jot with paint that didn’t get scraped.

Unlucky? No. There are plenty of reports of the hum and hence, lack of grounding of the case:

Hum when touching the volume knob: https://www.head-fi.org/threads/sch...view-head-fi-tv.818323/page-378#post-14348577
Buzzing changes when touching volume and top of enclosure: https://www.head-fi.org/threads/sch...view-head-fi-tv.818323/page-371#post-14315594
Buzzing when touching enclosure: https://www.head-fi.org/threads/sch...view-head-fi-tv.818323/page-371#post-14316187
Buzzing when touching enclosure: https://www.head-fi.org/threads/sch...view-head-fi-tv.818323/page-372#post-14316404
It must be your cellphone: https://www.head-fi.org/threads/sch...view-head-fi-tv.818323/page-174#post-13022772

 

[amirm]

You can’t have current (electron flow) without voltage (the potential which creates the “push”). V certainly equals IR (in a DC, and/or purely resistive AC circuit), and you’re free to solve the equation any way you like. Current flow and resistance don’t make a voltage though, the voltage potential over a closed circuit creates the electron flow- I think we both agree...

Of course there is a "voltage." It is the mains coming into the device.

How much is developed across the enclosure depends on its impedance and current available. At 60 Hz, impedance and resistance are pretty close so if we are reading 1 ohm, that is basically what it is. And if the breaker allows 100 amps for a brief amount of time before it pops, we have 100 * 1 = 100 volts. Mains in US is 120 volts RMS or 170 volts peak to peak. You will have a voltage divider where 100 volts will be across the case and the rest, elsewhere in the line.

Change the resistance to 0.5 volt and the voltage created around the case resistance will drop to 50 volts.

No one is talking about voltage getting created out of thin air without the mains being there.

 

[DonH56]

As an aside, you can have current flow at 0 V (superconductors, virtual ground inputs, etc.) and voltage potential without current (unloaded voltage source).

 

[Dana reed]

In my review of Schiit Jotunheim by chance I noticed that the upper part of the case is not grounded. As such, it allowed the transformer to induce hum into the output of the amplifier. Here are the measurements showing such:

A jumper between the drilled holes that are not insulated and the back RCA jack shield got rid of the noise:

This quick article is about accomplishing the same without said jumper.

Note that this is both a performance issue and potential shock hazard. Metal cases on a three-pronged outlet must be security connected to mains safety ground pin. While my fix will also help with the safety issue, it is insufficient. A proper screw and ground wire is necessary for proper safety grounding.

Start by disconnecting all cables, especially the power one. There are lethal voltages inside and you can easily shock and kill yourself. So check and double check that you have unplugged the mains cable.



The case is tricky to take apart and that actually contributes to the problem. First, take off the large volume control by unscrewing its tiny set screw (be careful to not round it). Behind it are a nut and washer which you need to remove with a pair of pliers Once there, you have this:

View attachment 13922

Now the hard part. Pull the case forward some. It will put up fair amount of resistance at first. Slide it forward with fair amount of force and it should move at least a quarter of an inch. It will then get stuck. The pins are in keyholes and if you slide the case too far, they will latch on the opposite side. I found that putting the case upside down on my lap, sliding it forward and back, and eventually tilting it to clear the volume control shaft works.

Don't be concerned if you can't get it off immediately. Go ahead and curse fair bit as I did every time i tried to open and close it. It is necessary and gets you refocused on getting it open. It will eventually come apart and you see the inside guts:
View attachment 13923

There are four pins in the top lid that slide into those keyholds. As luck would have it, they are not insulated like the rest of the case is. So what you need to do is sand/file off the narrow part of the keyhole where the pins seat when closed. I used a diamond file and kept the unit upside down to make sure nothing fell in there. Alternatively you can use a sharp pick and scrape off the paint.

Here, you need to have an ohmmeter to check that you have sufficiently exposed bare metal. Connect the probes to the different keyholes and you should get connectivity in the 1 ohm or less.

The powder coating or whatever they have used is very tough. It took fair amount of doing to get the paint exposed. A better solution is to take the entire unit apart and use a rotary tool to get in there and really clean up the paint. Better yet you could sand of the paint around the top of the keyholes to expose even more metal for good contact. Be sure to do the same in the lid. The issue here is that these components may rust. Use your own judgement.

Turns out that the above is NOT sufficient because even the bottom of the case is ungrounded! All that stays between you and electric shock is that powder coat. To fix that, you need to take out these screws and like above get rid of the pain that exists on the chamfer in the case:

View attachment 13924

Screw them back in and check with a multimeter that the keyhole area you sanded off now reads a low resistance with either the mains safety pin in the IEC connector or the outside of the RCA connectors. Keep sanding and exposing fresh metal until you get to a near short (1 ohm or less ideally).

Then put the lid back on following the reverse process (yeh right, as if it is that easy). Once you put the case all together and before putting the volume knob and screw back on, run this test to make sure there is good continuity:

Unlike the above picture where I tested before this mod, you should get a short/1 ohm or less resistance. I had to be careful with my loaned unit and I think I got it down to 2 ohms or so.

That's it.

I am hoping that Schiit responds and makes this fix for owners. But if not, you can improve things with this mod which takes 10 to 15 minutes.

Post here if you have this problem and the mod helps things.

As is always the case, you make these mods at your own risks. You can easily damage the unit once inside. So be careful.

Thanks for posting this! Originally when listening to full size headphones with this amp, I was unable to hear any hum or issues with the power supply/case. But recently I got a pair of Shure SE315 IEMs and they are sensitive enough that I could hear this issue. After doing this mod to make sure the connection to the top/bottom case from ground was solid, the hum is gone, even at full volume into the 116 dB sensitive IEMs.

 

[Alcophone]

Schiit fixed my Jotunheim. Volume knob and top are properly grounded. More here.

 

[DonH56]

Great they fixed it, too bad quality control appears to be pretty lax to begin with. I wonder how many bad units are simply due to poor final QC? You've had two components that have needed repairs, yes?

 

[amirm]

Schiit fixed my Jotunheim. Volume knob and top are properly grounded. More here.

Hmmm. You say:

Really? They released a product with both performance and safety issues and you thought to thank them? And not the folks who pointed out the problem?

If you had a car that steering wheel would vibrate because there was a danger of a wheel breaking free, after the hassle of taking it to the dealer to get it fixed, you would post online that you are thankful?

 

[amirm]

Great they fixed it, too bad quality control appears to be pretty lax to begin with. I wonder how many bad units are simply due to poor final QC? You've had two components that have needed repairs, yes?

In my opinion they should have reached out to all of their customers, telling them the unit may be unsafe and that they need to send it to them to verify and fix the problem. Waiting for customers to hear about my findings to know to contact them to repair safety problem is just irresponsible.

 

[Alcophone]

Technically nothing I ordered from Schiit "needed" repairs just to function, but both of my Jotunheims behaved the same in that funny test with no inputs/outputs connected other than power and headphones, volume on max, and touching the volume knob/top. That's gone now on this one after I asked Jason whether he wants to take a look. I sold the other one before Amir's grounding post, but the new owner also checked with his multimeter and his top wasn't grounded, either. He ultimately addressed that himself. I bought both units within one month from each other.

They have added grounding tests to their routine for the Jotunheim and Lyr 3 (same style of case), maybe other units as well.

I was blissfully unaware of the issue until one guy on Head-Fi started having noise issues during regular use, albeit with reduced gain due to software EQ, prompting another person to mention funny results as outlined in the first paragraph. The guy with noise issues during regular use later had basically the same issue with two other amps, the Sennheiser HDVA 600 and an Auralic Taurus MKII, so something seems wonky on his end. The Questyle CMA400i did not have noise issues, but he didn't like how it sounded.

 

[Alcophone]

Thank you, @amirm, for showing how to measure ground continuity! I love you!

 

[amirm]

Thank you, @amirm, for showing how to measure ground continuity! I love you!

OK, now I feel better. Thank you.

 

[JJB70]

I find this deeply worrying, whether a product performs well or performs badly it should be safe to use. And earthing is not advanced electrical engineering, I'm a mechanical engineer and even I have a decent understanding of earthing (for enthusiasts of earthing, the earthing grids under power plants are rather impressive to see during installation). To me this indicates a sloppiness that is deeply worrying not so much because of this particular earthing issue (which is bad enough) but for what it indicates about their design, manufacture and QA processes. This indicates deeper systemic issues with how they design and manufacture equipment as something this basic should never have reached consumers.

The self certification approach is quite a hot potato in some sectors, but that's a whole different bunch of hand grenades.

 

[LuckyLuke575]

Shit was was famous for their amplifiers damaging headphones a few years years ago, although they supposedly fixed that issue.

Also, every time I tell people Shit is bad I am "trolling."

Damaging headphones!? Unbelievable. How can they design and sell a "finished product" that does that? These people are playing fast and loose with the audio listening public. Their brand is toast IMHO.

 

[solderdude]

That was only the case with the first LYR amps (about 9 years ago ?)
Their brand is thriving and seems to have a big fanbase despite a few reports of some amps blowing up precious headphones.

Willing to bet there are more brands without proper DC protection on its output that can fry headphones.

 

[flipflop]

That was only the case with the first LYR amps (about 9 years ago ?)

There was also a problem with the first Asgard. This was 8 years ago.
https://nwavguy.blogspot.com/2011/07/banned-at-head-fi.html

 

[solderdude]

There are still problems here and there ...