Focusrite Scarlett 2i2 (2017 model) dead line out

[ruffrecords]

One of the line outs has died on my 2i2. iO got it cheap back in 2017 and it was always a bit flaky. Headphone out on same side died too. Are these easy/worth fixing?

Cheers

Ian

 

[pucho812]

All surface mount components.
At a price tag of an average of 120 dollars seems to me it would be more costly between time and parts then to replace.
Perhaps you open it and see if you can notice the obvious. Then see if it’s worth the hassle or just replace.

 

[Khron]

If both the line out AND the corresponding side of the headphone out died, that would point to the DAC side of the codec, or whatever opamps are common before the signal splits off.

Which generation is that, though? Gen1 has the plastic silver volume knob (no black line), Gen2 has metallic (?) volume knob (with a black like), Gen3 has all black knobs.

 

[ruffrecords]

It is Gen 2. I only got it because I updated Windows and my Gen 1 was no longer supported which is why I bought a B stock Gen2.

At least the Gen1 still works with Linux.

Cheers

ian

 

[Khron]

For what it's worth, i did a teardown and documenting of the more important silicon in a gen1 2i2, as well as gen1 and gen2 2i4 on my blog, in case it's any help..?

 

[Ricardus]

It is Gen 2. I only got it because I updated Windows and my Gen 1 was no longer supported which is why I bought a B stock Gen2.

At least the Gen1 still works with Linux.

Cheers

ian

My Gen3 works on linux also.

 

[ruffrecords]

For what it's worth, i did a teardown and documenting of the more important silicon in a gen1 2i2, as well as gen1 and gen2 2i4 on my blog, in case it's any help..?

Cool. I will definitely check that out.

Cheers

Ian

 

[user 133392]

My Gen3 works on linux also.

Can also confirm that my 2i2 Gen 3 is plug-and-play under Linux. No additional drivers are required for Ubuntu.

The 2i2 is a very good value. I use the iDefender power injector/ground isolator and an external 5V supply with mine and they become significantly quieter.

It's my daily driver on the bench and desktop - I have two.
I only have to switch to the PCM4222EVM for low-level noise measurements. The 2i2 bottoms out around -96 dBu/-104 dBFS when calibrated for 0 dB signal gain. The PCM4222 is around -123 dBFS.

The 2i2's balanced 300Ω/leg outputs are capable of +15 dBu into a high impedance load.

 

[capacitorless]

Happy with my 2nd gen 6i6 Scarlett. It was cheap & cheerful because one of the two phone outputs had a single channel out. I'm pretty good with fixing circuits, but after opening it up, I decided to pass. Just fwiw, and maybe I'll try later after I get some significant SMT work under my belt

 

[merlin]

Quick ghetto sketch of a Focusrite 2i2 I traced a couple of years ago, in case it is useful to you

 

[Ricardus]

I use the iDefender power injector/ground isolator and an external 5V supply with mine and they become significantly quieter.

Significantly quieter? Would love to see some measurements on that.

 

[rock soderstrom]

Significantly quieter? Would love to see some measurements on that.

I believe that this will improve things. The question is rather, can't we do it ourselves for much less money? Ultimately, only the grounding of the USB line is interrupted and an external power supply is connected. Right?

 

[k brown]

How about neither?

I can run my Scarlett 8i6 off one of these for over 2 hours; don't need to be near an AC outlet, or tax my laptops's battery.
https://www.amazon.com/gp/product/B01M7Z9Z1N/ref=ppx_yo_dt_b_asin_title_o04_s00?ie=UTF8&psc=1

 

[Khron]

How about neither?

I can run my Scarlett 8i6 off one of these for over 2 hours; don't need to be near an AC outlet, or tax my laptops's battery. A 2i2 should last much longer.
https://www.amazon.com/gp/product/B01M7Z9Z1N/ref=ppx_yo_dt_b_asin_title_o04_s00?ie=UTF8&psc=1

Since a 2i2 is bus-powered, you could get away with a powerbank meant for phones (with an appropriately confectioned Y-cable).

I myself shoehorned a 6-7000mAh pair of lithium cells out of a tablet, plus supervision & charging circuitry into a Saffire 6 USB Of course, the form factor was a bit more conducive for this solution, but 1-3 18650's could likely(?) fit inside a 2i2

https://khronscave.blogspot.com/2021/09/78-focusrite-saffire-6-usb-20-modding.htm

 

[Ricardus]

I believe that this will improve things. The question is rather, can't we do it ourselves for much less money? Ultimately, only the grounding of the USB line is interrupted and an external power supply is connected. Right?

Improve what? I bet if you test it there is no real improvement, and certainly no audible improvement.

 

[rock soderstrom]

Improve what? I bet if you test it there is no real improvement, and certainly no audible improvement.

Okay, we seem to be talking about different things. My experience is that USB powered audio interfaces suffer from the interference of computer hardware. This is not always the case, but quite often, depending on the computer. Chirping, hissing, artifacts to the point of audio dropouts, I have seen it all. With my Steinberg interface you could model the noise with the mouse, almost an instrument.

If you disconnect the ground of the USB connection and use an external power supply, everything is quiet again.

For this problem, a separate device genre has been established. Whatever they do, I find this stuff overpriced.

PS: the improvement is clearly audible, no need for measurements

 

[Ricardus]

Okay, we seem to be talking about different things. My experience is that USB powered audio interfaces suffer from the interference of computer hardware. This is not always the case, but quite often, depending on the computer. Chirping, hissing, artifacts to the point of audio dropouts, I have seen it all. With my Steinberg interface you could model the noise with the mouse, almost an instrument.

If you disconnect the ground of the USB connection and use an external power supply, everything is quiet again.

For this problem, a separate device genre has been established. Whatever they do, I find this stuff overpriced.

PS: the improvement is clearly audible, no need for measurements

I do hear people occasionally talk about these artifacts, but I have never had them. It doesn't mean they don't exist. Do you have any recorded examples?

My problem with the claim that it's audible is that every home hifi enthusiast in the world claims an expensive IEC cable makes their rig sound better. They're absolutely convinced of it. So without objective testing I just file it in the circular bin.

To quote one of our mods:

"Audio technology is quite mature, while human nature remains fallible and susceptible to suggestion."

 

[rock soderstrom]

My problem with the claim that it's audible is that every home hifi enthusiast in the world claims an expensive IEC cable makes their rig sound better. They're absolutely convinced of it. So without objective testing I just file it in the circular bin.

To quote one of our mods:

"Audio technology is quite mature, while human nature remains fallible and susceptible to suggestion."

Yes, once again. We are talking about two different topics. I'm not talking about the audiophool nonsense of "sparkling highs" and "abysmal bass" due to exotic power cords and gold-plated fuses. Don't put me in that corner.

I'm talking about real technical problems and audible noises, so that normal work is not possible at all. This happens quite often and is very often caused by "contaminated" USB power supplies or the interaction of different (faulty?) PC components. Check the Internet and you will see what I mean.

I'm out now, this conversation is going nowhere.

 

[Newmarket]

The noise from USB power is definitely a thing. System dependent of course.
The problem with simply running in a 'clean' power supply is that you still need a 0V connection of some sort to ensure reliable USB operation. And this can cause noises. USB data signalling may be differential but it is not isolated. So there's a possible issue of exceeding the common mode range if there is no reference of the USB 0V of the eg the laptop to the 0V of the external DAC. You can get widgets to galvanically isolate USB (or diy one if you have the knowledge and time). But ime they do get significantly more expensive as the USB max data rate increases (I've used them in instrumentation applications) and I don't know about availability for USB-C / Thunderbolt etc.
The other alternative would seem to be adding an impedance / diodes in the 0V connection to minimise interaction whilst keeping the 0V levels close enough to each other. Essentially a 'ground lift' network but for data signals rather than analogue audio (no I haven't tried it).
Anyone know what's inside the iDefender commercial device mentioned above ?

 

[user 133392]

Significantly quieter? Would love to see some measurements on that.

By "quieter" I mean tonal USB "hash" deep in the noise floor.
Someday I may measure it. I could crank the monitors up and record it acoustically with my phone and you could hear it.
Defending the iDefender is not a priority however.

It's not a universal solution to USB noise and doesn't always provide benefit, but its an option in the toolbox.
More often that not however the DAC output idle channel noise sounds more like noise rather than hash when it's used.
Your mileage may vary.

I don't know if the iDefender provides complete galvanic isolation or simply puts impedance in the ground line. I suspect the later.
It claims to chose the best source of power and ground - either bus or external - to provide the lowest noise.
Sometimes, when I'm switching audio connections and lift signal grounds, the iDefender will switch modes which requires the USB link to be re-initialized.
It almost always appears to be using external power which is far cleaner than bus power with it's added I*R drop and inductance from cabling.

The problem with using external USB battery supplies is that they're not batteries.
They do provide galvanic isolation but there's always a switcher, potentially a noisy one, between the battery and USB connector.