Potentiometers and LR balance issue

[IAIxara]

Hi! Long time ASR reader, just freshly registered.

I got a Scarlett 2i2 3rd gen long time ago. I'm used to low headphone levels, thus only using like the first 10% of its headphone pots. Unfortunately, the stereo pot is not very well-matched at that range, resulting in a +-1.5dB difference between left and right channels. I ended up with turning the pot 20dB higher than needed (which is still at 9:00) and trim it down with correction EQ.

Since then, I only buy interfaces with digitally controlled HP outs, but this post points out that it's possible to use a single pot to control both channels' DAC reference level, thus eliminating the LR balance problem at lower volume.

So, long story short:
1. I'm planning to buy some of the following gears: Topping L30 and 4x4, RME 802fs, Presonus 1824c and Studio 192. Do their HP outs have LR balance issues?
2. Is there any way to tell, even vaguely by series / brand or price, whether a pot would cause LR balance issue? According to the link above, I'm assuming MOTU M series is free of this problem.

 

[antcollinet]

Personally I just use digital volume control, and fail to worry about it.

 

[sergeauckland]

Analogue resistive pots always have some imbalance. 1.5dB at the bottom 10% isn't at all bad. Pots can be better matched, but that costs a lot more, and as Ant mentioned above, digital volume controls just don't have that problem and are cheap to implement so are increasingly the norm. My now well over 30 years old Meridian preamp has a digitally controlled analogue volume control that's balanced to 0.1dB.

The benefit of a true mechanical resistive pot is that there's no risk of software turning it to 100%, but apart from that, I can't see any benefit of a physical pot.

If your intended purchases use digitally controlled analogue pots or do volume control in the digital domain, you won't have balance problems.

S

 

[Speedskater]

Use a signal attenuater to get the pot higher in the rotation, where the channel tracking is better.

 

[IAIxara]

Thanks for everyone's reply... Too busy with IRL things.

Personally I just use digital volume control, and fail to worry about it.

Surely digital controls are good, but not very common on lower budget interfaces, while more expensive ones usually comes with too many I/Os or features that I don't really want to pay for.

Analogue resistive pots always have some imbalance. 1.5dB at the bottom 10% isn't at all bad.

I understand, but from that MOTU M4 post I linked above, it seems that there's a way to control both DACs' "reference voltage" with a single potentiometer, thus free of this pairing issue. I wonder why not all manufacturers do it this way.

Use a signal attenuater...

Any recommendations? I know some Topping amps could do a few dB of attenuation, but that's not what they were designed for. And none of them comes with an Aux input.

 

[bennybbbx]

If the output and headphone is digital control you can see mostly if the knob can endless rotate, and have a LED position display. the LR inbalance come due to wear and tear. it happen in the range of the pot you use often more. i test it. when new there is diffrence only ~0,2 db. when move them 10 times per day it can get in 11 to 1 clock ~0,8 db diffrence after 1 month of use. pot developer mostly give durability of their pots and tolerance +/-20% and 10.000 movements .for more details about channel inbalance see here. https://www.audiosciencereview.com/...en-interface-review.58790/page-9#post-2319193 I have see simular results on 7 analog pot devices i buy in the hope

2 * presonus ar8 mixer (2 because i get a free repair)
2 * mackie Big knob (2 because i get a free replacement)
1 * behringer Mini Amp 4 channel headphone
1 * behringer UMC 22
1 * focusrite 8i6o

after the waste Learning money for this devices, i come to conclusion never use analog stereo pots and i buy a apollo solo which is digtital control also on headphones
So all in all can say, that such stereo anolog pots should never use on a good quality device and the most problem is that no tester, notice this or do tests after 1000 pot movements. so the industrie of studio equipment continue to sale such worse low durability devices. when new i am sure all analog pots are good. I have in link https://www.audiosciencereview.com/...n-interface-review.58790/page-10#post-2325384 a hearing test, so you can check if you can hear 0,9 db diffrence between left and right as bad stereo image.

I buy the apollo solo that it can replace also a DSP for speaker correction and there is a big plugin packet attach. for no DSP usage, with digital headphone control the cheapest i see is presonus https://de.presonus.com/products/quantum-es-2-usb-c-audio-interface

 

[IAIxara]

it happen in the range of the pot you use often more.

Thanks for the info, this is indeed concerning. My main question here is that, it seems possible to use a single pot for dual channel volume control, which would eliminate balance issues, but not many manufacturers are using it; this wearing problem makes the case even weirder.

I actually had an ES2 but immediately sold it. Maybe it's my laptop's spec (AMD CPU?), but it just doesn't work, crackles even with 1024spl buffer. Not to mention that 33ohm (or maybe 51) headphone output impedance that always comes with Presonus.

It's sarcastic (and frustrating) that most "pro" audio companies make headphone jacks that are worse than a normal laptop headphone output.

 

[Zek]

it seems possible to use a single pot for dual channel volume control

As far as I know, this is not possible because each channel has its own pot.
Unless you mean a shared dual-pot that is on one shaft and has one knob for regulation.

 

[bennybbbx]

Thanks for the info, this is indeed concerning. My main question here is that, it seems possible to use a single pot for dual channel volume control, which would eliminate balance issues, but not many manufacturers are using it; this wearing problem makes the case even weirder.

I actually had an ES2 but immediately sold it. Maybe it's my laptop's spec (AMD CPU?), but it just doesn't work, crackles even with 1024spl buffer. Not to mention that 33ohm (or maybe 51) headphone output impedance that always comes with Presonus.

It's sarcastic (and frustrating) that most "pro" audio companies make headphone jacks that are worse than a normal laptop headphone output.

the idea to change the reference voltage of the DAC and so all channels are reduce in same and need only a mono pot to control is in theorie possible. Only because there are no precise reference voltage circuits that can create any voltage so it do not work with current stuff. keep in mind the reference voltage need less riple and need the voltage keep in 0.000x% to get low distortions. good Zener Diode only reach 0,07% and are fixed to 1 voltage only. , maybe if the testers report 20 years ago the high wear and tear of pots, we have now such controllable DAC reference voltage. this avoid a dsp and can be maybe cheaper because need only a mono pot to control this.

 

[antcollinet]

the idea to change the reference voltage of the DAC and so all channels are reduce in same and need only a mono pot to control is in theorie possible. Only because there are no precise reference voltage circuits that can create any voltage so it do not work with current stuff. keep in mind the reference voltage need less riple and need the voltage keep in 0.000x% to get low distortions. good Zener Diode only reach 0,07% and are fixed to 1 voltage only. , maybe if the testers report 20 years ago the high wear and tear of pots, we have now such controllable DAC reference voltage. this avoid a dsp and can be maybe cheaper because need only a mono pot to control this.

Much MUCH easier to use a dual volume control IC. Most though seem to be digitally controlled, and they go to low volume on power up.

 

[bennybbbx]

Much MUCH easier to use a dual volume control IC. Most though seem to be digitally controlled, and they go to low volume on power up.

What IC do you mean, so can look at specs and price ?

 

[antcollinet]

What IC do you mean, so can look at specs and price ?

A selection

https://www.analog.com/en/parametricsearch/12881#/

 

[bennybbbx]

A selection

https://www.analog.com/en/parametricsearch/12881#/

some of them have bandwidth only 20 khz . this have 100 khz https://www.analog.com/media/en/technical-documentation/data-sheets/MAX5440.pdf but as see in the specs diffrence of left to right channel they list with +/- 0,5 db (so can be max 1 db) https://www.analog.com/media/en/technical-documentation/data-sheets/MAX5440.pdf

can gues in the series it is mostly less but for a developer of studio hardware this look not good. 1 db. also that this do 2 db steps.
this is the most expensive but also have from the specs +/-0,5 db so overall 1 db allowed.

https://www.analog.com/media/en/technical-documentation/data-sheets/DS1801.pdf

4. Inter-channel matching is used to determine the relative voltage difference in dB between the same
tap position on each potentiometer. The DS1801 is specified for ±0.5 dB inter-channel matching.

My apollo solo have 0,007 db diffrence between left and right measure which is of course more than good enough. maybe the left right diffrence of the chips you list are in series production higher as a new analog stereo pot. if a tester really tests left right diffrence(nothing is imposible) then it can happen that the new analog pot give less left right diffrence. A new stereo analog pot have at 8 o clock to full less than 0,3 db diffrence. mean tolerance +/-0,15 .so what this chip reach in series production maybe nobody test it. advantage of the chips is the get no wear and tear.

 

[AnalogSteph]

Scaling the reference voltage seems to be what CS43131/CS43198 are doing internally. You can't get such a wide dynamic range (approaching 130 dB) on low-power chips without some analog trickery. You can take the notoriously hot-headed ES9038PRO as an example for what happens when playing it straight with 8 channels.

Looking at the ES9312 reference voltage regulator datasheet, its output is set by two external resistors per channel, so scaling certainly seems possible... the crux being, it's a 2-channel affair so you'd be introducing channel imbalance issues right back, plus it looks like ESS DACs (and many others) have a shared DAC reference voltage and analog supply so there may be some unintended side effects and there is likely to be a lower limit, giving a very low usable adjustment range by the standards of volume controls (maybe 12 dB if you're lucky).

Anyway, you are unlikely to find PGAs for output volume control in a modern audio interface. The ones that do get near modern DAC performance in dynamic range and distortion (e.g. PGA2311 for +/-5 V, PGA2320 for +/-15 V) are fairly expensive. (About 5 or 10 USD in quantities of 1000 in this example, and approaching twice that in small numbers.) These days, it is just more economical to implement only a basic constant-gain output stage and do digital volume control instead. Only CS4272 class chips (AK4621, ES9290) still require external analog volume control.

How that digital volume is actually controlled is up the the developer. Using encoders is a classic approach, but leaves you with the problem of having to store the last used volume, or else it'll be reset to (usually) zero next time the device is powered up. On the upside this makes software control easy to implement. Using a pot to generate a DC voltage read out by an ADC eliminates the reset problem, but obviously also means software control is out of the question unless you're going for a motor pot (not the least expensive thing ever). For a home studio with little to no automation, this may be the better approach - ultimately it's going to depend on whether the user is more of a knobtwiddler or prefers pushing sliders on-screen, i.e. it's a matter of workflow.

 

[bennybbbx]

the ES9312 have few steps of voltage possible 1,2, 1,8, 3,3 4.5 Volt. see in datasheet https://www.esstech.com/wp-content/uploads/2024/03/ES9312_DataSheet_v0.2.1.pdf

ESS TECHNOLOGY, INC. 109 Bonaventura Drive, San Jose, CA 95134, USA Tel (408) 643-8800 • WWW.ESSTECH.COM 8
CONFIDENTIAL ADVANCE INFORMATION
ES9312 Product Datasheet
VERSION 0.2.1
Functional Description
The ES9312 is a dual output ultra-low noise regulator for powering noise sensitive applications. The ES9312 features
multiple voltage output modes making it the perfect companion for ESS DACs or ADCs. The dual regulator features
overcurrent, undervoltage, overtemperature warnings as well as overtemperature shutdown. The ES9312 can output up to
180mA per channel at preset settings of 1.2V/1.8V/3.3V and can output up to 100mA at 4.5V.
The ES9312’s dual outputs can operate in any of three modes: Internal Fixed Voltage Mode for using the preset voltage
output, External Fixed Voltage Mode for adjusting the output voltage using a combination of two resistors connected to
RCALx/VSELx and REXTx, and Calibration Mode for using with compatible ESS DACs.
Internal Fixed Voltage Mode
The ES9312 has internal fixed voltage modes that can output 4.5V, 3.3V, 1.8V, and 1.2V per channel, using the voltage
select pin (VSEL), pins 4 and 9. To enable the functionality of the VSEL pin, the corresponding REXT pin must be floating.
The VSEL pin is then set to 4 possible different states to determine the channels output voltage. When using both channels
in internal fixed voltage mode, independent resistors must be used per VSEL pin.
REXT VSEL1/2 Setting VSEL1/2 Resistor (Ω) Vout (V)
Floating
(NC)
Tie High 47 to AVCC 4.52
Pull High 47k to AVCC 3.3
Pull Low 47k to GND 1.8
Tie Low 47 to GND 1.2

EDIT: a closer look there seem another mode which seem allow any voltage, but need change resistors

External Fixed Voltage Mode
The ratio of the resistors connected to pins REXT and RCAL will determine the output voltages between 1.2V and 5V.

EDIT 2. because voltage can only from 1.2 to 5 volt change if it can really control with a pot in realtime. problem is, the diffrence from max to min volume is only ~12 db