Digital Volume Infinite Encoders slow and need many rotations. Why is that?

[DrSpan]

AliExpress shows about $€£ 25 f

What does it mean in the final product though. Say 1 costs 10 and the other dollars. The device will bew more than 10 dollars more expensive i suppose, but what is the ratio?


"Obviously you can’t just jerry rig such a thing in a Flex. You could however probably find a drop in replacement for the Flex encoder with more pulses per revolution.“

Never in life could i modify the device. I was never the soldering and technically super skilled type

 

[voodooless]

What does it mean in the final product though. Say 1 costs 10 and the other dollars. The device will bew more than 10 dollars more expensive i suppose, but what is the ratio?

Hardly. The encoder in the Flex is probably a < $1 part. Something costing $ 10 will probably add like $ 30 to $ 50 to the product price.

 

[DrSpan]

Hardly. The encoder in the Flex is probably a < $1 part. Something costing $ 10 will probably add like $ 30 to $ 50 to the product price.

I would happily pay 50 more for having this encoder tbh .

 

[TurtlePaul]

A Google search "single turn absolute encoder" gives plenty of 12 bit (4096 steps) and even saw 22 bit (way too many steps) encoders. I guess state of the art will be even more bits, and more $$$. Now you only need to find someone that builds you a volume knob with them.

There is a tradeoff. If you want to have a single turn encoder and also have 100 dB range, then each ‘hour’ is 8.25 dB (e.g. noon to 1 o’clock position) and on smaller knob this is far far too sensitive.

 

[voodooless]

There is a tradeoff. If you want to have a single turn encoder and also have 100 dB range, then each ‘hour’ is 8.25 dB (e.g. noon to 1 o’clock position) and on smaller knob this is far far too sensitive.

You don’t need the range to be db linear. The first 30 or 50 dB can probably be compressed into 2 hours. That leaves more for the rest.

 

[TurtlePaul]

You don’t need the range to be db linear. The first 30 or 50 dB can probably be compressed into 2 hours. That leaves more for the rest.

But then -50 to -100 dB is unusable even if it can be reached… may as well make it another turn of a multi-turn encoder… which is exactly how these are implemented.

 

[Berwhale]

You don't need an expensive single turn encoder (absolute encoder), you just need to program your device to understand velocity of an incremental encoder which transmits a set amount of pulses per rotation.

You probably all had optomechanical mice at some point, they had incremental optical encoders in them, it didn't stop you ticking the box in Windows (or you OS of choice) labelled 'mouse acceleration' or similar.

Mouse Rotary Encoder: The Working Principle and How to Connect It With the Arduino

the mouse rotary encoder is the ideal device for the job as the sensor helps you convert the oscillation of a knob to an output signal.

www.ourpcb.com

 

[DrSpan]

You don't need an expensive single turn encoder (absolute encoder), you just need to program your device to understand velocity of an incremental encoder which transmits a set amount of pulses per rotation.

You probably all had optomechanical mice at some point, they had incremental optical encoders in them, it didn't stop you ticking the box in Windows (or you OS of choice) labelled 'mouse acceleration' or similar.

Mouse Rotary Encoder: The Working Principle and How to Connect It With the Arduino

the mouse rotary encoder is the ideal device for the job as the sensor helps you convert the oscillation of a knob to an output signal.

www.ourpcb.com

What i am missing and why i wish for a One rotation encoder is the Mechanical feedback.
I could blindly turn the knob and immediately knew where i am and how loud it should-would sound.
With Infinite knobs i always need to look at numbers and also have zero feedback as to where i am without constanlty looking at numbers.
I simply i don`t like this. I think its a taste thing cause as i said before, no one seems to be feeling the same.

 

[NTK]

Check out Accuphase if you want to see a first-class volume knob mechanism design.

 

[Berwhale]

What i am missing and why i wish for a One rotation encoder is the Mechanical feedback.
I could blindly turn the knob and immediately knew where i am and how loud it should-would sound.
With Infinite knobs i always need to look at numbers and also have zero feedback as to where i am without constanlty looking at numbers.
I simply i don`t like this. I think its a taste thing cause as i said before, no one seems to be feeling the same.

I understand. Personally, I think it's a lot easier to read the big numbers off the display on my EX5 than it would be to read the angular position of a volume knob...

I appreciate that this might not be the case on other devices where the display is not as clear or responsive.

 

[DrSpan]

I understand. Personally, I think it's a lot easier to read the big numbers off the display on my EX5 than it would be to read the angular position of a volume knob...

View attachment 351796

I appreciate that this might not be the case on other devices where the display is not as clear or responsive.

I am sure that for many people it is nice and also looks nice to have numbers and such.
However i am mixing music. I am reading numbers all day long. I want the Listening volume knob to be "out of sight and inside reach“ if that makes sense.
If i want to find out where i am at , when i had the Spl a simple turn to the left and then i knew where the start is and then when i turned to the right again i knew pretty much exactly which position-how much travel should sound at what volume so i know where the mix is at. I think it is simply the way my brain works. I dont want distractions if they are not necessary. I had 90% of my most used listening volume positions from Min to around not even Half rotation. But could also regulate miniscule differences if i wanted to. This is also what i absolutely hate about my iPhone. There are times when you want to simply listen to something at night over headphones and one position is too quiet and the next one just a bit too loud. I am being honest that i never had issues with Volume control until digital came. At this point i have to
also say that i am in no way an „analog vs digital“ or „analog is so much better“ kind of person. Not at all. I love what plugins can do nowadays. But the Volume,,,, Jesus.
To me Numbers for Listening volume are not necessary when working as i only control using this and not print mixes using it. Same way i leave my master fader on 0db when mixing cause it is my reference. If it was only about listening to music i would have settled a long time ago tbh
Hope what i say makes sense

 

[solderdude]

Time to buy a different volume control ?

 

[dweeeeb2]

With any luck it could be modified for a future FW update?

 

[Mnyb]

I think it’s all deliberate, it’s gives finer control to the end user without challenging the end users motor skills .

In the old bad days your amps also had to much gain for modern sources so in day to day use you actually turned the knob like a third of the way and had to be careful.

With multi turns the manufacturers can accommodate users with very bad gain scaling in their system to.

 

[Mnyb]

With any luck it could be modified for a future FW update?

+1 I would like settings for number of turns and max and min volume for the knob to work within .

But it still would not have mechanical end stops ? So you could not adjust by feel alone ?

 

[Galliardist]

How much of that volume range do you use on a regular basis anyway? I've had two integrated amps where I don't think I ever actually turned the volume control by more than five degrees, that range being from too quiet to too loud, and spent a lot of time trying to get the volume right.

 

[KSTR]

Because it is poorly programmed.

This! There is nothing else to say.
Programming a decoder algorithm with dynamic acceleration when turned faster is pretty much trivial but indeed it's not done in many products for reasons we can only speculated about... lazyness? Lack of skills? Lack of even grasping the problem?

 

[Mnyb]

This! There is nothing else to say.
Programming a decoder algorithm with dynamic acceleration when turned faster is pretty much trivial but indeed it's not done in many products for reasons we can only speculated about... lazyness? Lack of skills? Lack of even grasping the problem?

Many cheap encoders are also terrible and jittery so it can be a challenge. So it can be a lazy way to make it smoother averaging over many pulses ?

The old Squeezebox transporter had one with motion feedback , how they used I don’t know ? It’s a nice idea to program tactile feedback in the knob .

 

[voodooless]

Programming a decoder algorithm with dynamic acceleration when turned faster is pretty much trivial but indeed it's not done in many products for reasons we can only speculated about..

For one, it’s very unnatural feeling. I really don’t like encoders that way. Secondly, with only 12 pulses per revolution, there isn’t much to work with. You’ll need more angular resolution to make this work halfway decent.

 

[KSTR]

Many cheap encoders are also terrible and jittery so it can be a challenge. So it can be a lazy way to make it smoother averaging over many pulses ?

The old Squeezebox transporter had one with motion feedback , how they used I don’t know ? It’s a nice idea to program tactile feedback in the knob .

Even with cheap glitchy encoders you can write a decoder that correctly gets what the user meant 99% of the time.
RME for example had a problem with a bad batch of encoders in some products which they could fix with intelligent software.

Haptic feedback is the future... I'm currently working on this in my day job, with endless continuous (analog) rotary encoders... where you have the same problem, you need a robust decoder algorithm.