Quad Potentiometer - Does such a thing exist?

This may be hard to clearly explain, but here goes:

My goal is to take a 4 channel (quad) sound system, and rotate the channels so they "spin" around the room, using a manually controlled knob, similar to a joystick. This device will be placed into the effects loop on a commercial sound system, or between the preamp and power amp.

So, lets give each channel a number 1 thru 4.

Normally, channel 1 - 2 - 3 - and 4 are going to speakers placed in each corner of a room. But when this device is turned 1/4 turn, channel 1 becomes 2, 2 becomes 3, 3 becomes 4, and 4 becomes 1.

A half turn will move the sound one more speaker around the room, and so on.....

If this device is turned rapidly, a full 360 deg. the sound will "spin" around the room.....

Ok, so far, it probably sounds like a simple rotary switch, except for one thing. I want the sound to fade and gain. In other words, the sound will fade as it leaves speaker 1, and gain as it enters speaker 2. At the same time, speaker 4 will fade and gain to speaker 1.

Each channel fades as it leaves the speaker it was using and gains as it enters the next speaker. Thus, the overall sound level (volume) from all speakers combined, will remain constant, while the sound travels from speaker to speaker.

What I'm picturing is having four potentiometers. Each one is acting as a "volume control", but as one is lowering the volume to speaker 1, the next control is raising the level to speaker 2, while at the same time, speaker 1 is increasing in volume from the source which was previously powering speaker 4.

[I said this would be hard to explain]....

To sum up the whole thing, the sound will simply flow from speaker to speaker as it fades from the prior speaker to the next one.

I'm starting to catch on how to wire this (sort of), but I have no idea how to make all 4 potentiometer shafts to turn at the same time. I also believe I need potentiometers that do not have stops. (meaning that they can be rotated 360 deg and continue to rotate in the same direction indefinitely.

This is where my question arises: Do they have such a device already made, where there are 4 potentiometers connected by gears to one central shaft, and they are endless (no stop) type pots?

I have never really studied joysticks, but I did look up some web articles about them, and it appears that some of the early ones were simply potentiometers and were referred to as ANALOG joysticks.

But, if they do exist, where do I get more info about them and can they be purchased?

Reply to
oldschool
Loading thread data ...

On a sunny day (Mon, 29 May 2017 03:51:41 -0400) it happened snipped-for-privacy@tubes.com wrote in :

In a stage light dimmer, I have seen many pots with hollow shaft put on one main shaft.

But seriously, I know you post as 'old school' & 'tubes' but that sort of thing is done digital in a simple way.

An other aternative is use those voltage controlled ampliers on a single pot.,

Dunno about that, rotary encoders, digital...

Touch screen... circle, digital.

Yes, those have 2 sliders inside.

Put listener on rotating podium, or rotate speakers around listener on rails, or ...

I'd choose: rotary encoder, PIC, voltage controlled amplifiers. Would be a real-time enough user response.

Reply to
Jan Panteltje
[snip]

I saw this done in the early '70s, electronically. They were using jfets to control the level to each speaker and driving the jfets with an appropriate signal. It worked well.

Reply to
gray_wolf

y
l

to control

al.

That's pretty much how I'd do it, with gain control ICs or discretes. Your single pot can drive an encoding circuit. You would of course need many of those gain controls, since every input maps to every output.

Quad pots exist, but I've not seen them made in decades. To move input chan nel 1 to output channels 1 then 2 then 3 then 4 you'd want a 16 channel pot . Pots that could be stacked on a common shaft used to be available, don't know if they still are. Pots are out of fashion nowadays.

NT

Reply to
tabbypurr

There are 360 degree potentiometers, used for servos, but they're boutique, rare and expensive. A stack of four would do the job.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
 Click to see the full signature
Reply to
Tim Williams

Yes it is easily done and I have done on XY vector rotation from 2 sources, you just have 2 extra sources.

I don't know any commercial products, but we included this feature back in a SCADA system in 1978.

motors to servo,control Bournes SIN+COS continuously variable pots. Now it could be done with incremental encoder dials and sin/cos firmware mixers.

You would need a Sin+Cos pot with 4 taps to go from 4 inputs to 4 outputs on taps. If they exist... ask suppliers.

formatting link

2932300_1496068563.png. shows typical XY with only 2 taps, you need 4 taps for line IN to Line OUT.

------

I also remember a demo in our EE student lounge at U of M in 1970 where "Ad vance" Electronics in Winnipeg installed a demo of Quad sound with a modifi ed TV CRT to act as a vector scope for Quad sound with each of the 3 bands converted to colours and projecting from the centre at angles of 45 deg, +9

0,+180,+270. The sound was amazing and the display showed wonderful loops of coherent s pectrum shifting in every quadrant and dancing on the screen returning to a centre dot between tracks.

A 2D colour organ if you like, with 4 vectors and 3 colours , one for each decade. Some came out yellow and purple indicating a crossover in bands. A rather novel fourier spectral density display.

Warmest regards, Tony

Reply to
Anthony Stewart

How about reading the value of a single pot, calculating the desired gains and combinations, and use multiplying DACs to effect the fractional sourcing for each output. Hey, you'll only need 16 DACs to cover all combinations!

Haha, that makes it clear why digitizing four channels and computing new outputs in real time makes more sense. Try an Arduino Zero running at 48MHz, easy programming. Or step up to the Teensy 3.6, fast, 180 MHz. Advantage: extensive audio libraries written by Paul J. Stoffregen. Plug Teensy into an adapter board to use Arduino shields.

--
 Thanks, 
    - Win
Reply to
Winfield Hill

On a sunny day (29 May 2017 08:42:54 -0700) it happened Winfield Hill wrote in :

16 DACs? 4 channels? 160 MHz?

In the attic I think I may still have some of those Philips i2c controlled audio potmeters.. something like TDA7318

formatting link
then all you need is 2 (or 4) of these and a 2$ PIC to read a rotary encoder and send the i2c to those chips. Maybe sine cos lookup table, depends. Digitizing causes latency.. why bother. ?

4 MHz internal clock.
Reply to
Jan Panteltje

PS TDA7318D ST SOP-28 2$ a piece with free chipping, eh sjipping, eh shi//

formatting link

Reply to
Jan Panteltje

Good. Dual is nice, but then aren't 8 required? Direct and combine, any part of four inputs to any part of four outputs.

--
 Thanks, 
    - Win
Reply to
Winfield Hill

I would complain if pin 28 is not bent like in the photo.

--sp

Reply to
speff

On a sunny day (29 May 2017 10:29:04 -0700) it happened Winfield Hill wrote in :

I dunno, each chip has 4 left and 4 right inputs, and 4 independent attenuators. It seems to me tha tsomewhere in the 'round' rotation all you need to do is switch inputs? (at 180 degrees?)

Reply to
Jan Panteltje

On a sunny day (Mon, 29 May 2017 10:31:01 -0700 (PDT)) it happened speff wrote in :

Huh, that is nothing, in my picture the pin is broken of completely and fixed with a piece of wire:

formatting link

I changed the soft and hardware and moved that signal to pin 16 (from pin 15) this morning, so I can keep using the damaged PIC.

Poor PIC. Just had fresh 18F14K22 in.

Reply to
Jan Panteltje

I hope the CAD footprint is compatible with the actual package.

Reply to
krw

~ 35 years ago I used a sine/cosine potentiometer . Use 4 of those on one axis, connect the outputs to amplifiers/speakers, and that should work. Feed "sine cosine potmeter" to google picture, and check them out. :)

Reply to
Sjouke Burry

On a sunny day (Mon, 29 May 2017 18:10:17 GMT) it happened Jan Panteltje wrote in :

PS, actually that is interesting, this is what I came up with:

formatting link

the restriction of the chips are that the 2 channels in each chip use the same switch position, both 'stereo' input switches are coupled. The good thing is that we can set both left and right speaker attenuation from 0 -30 dB (leaving common main volume at 100% ). By combining the 2 speaker output on the same chip we can get to 36 dB,

Now if you look at the right of the drawing you have 4 signal sources (quad audio) A, B, C, D. at the top right of the drawing we have speaker front (the one in front of us), it, at his moment in time, has as input source A and B, where A is not attenuated, and B maximum attenuated, so we basically hear source A in front of us. Now we turn our rotary encoder clockwise, and decrease the gain G1 and increase the gain G2 (speaker attenuators remember), so after turning the knob 45 degrees we have zero from source A and 100% from source B, now we hear B in front of us. When we move the knob further to 45.001 degrees? we switch so the inputs B and C are selected, where B is set to 100% and C is zero, nothing changes in the output. Turning further we decrease B and increase C until at 90 degrees knob rotation we hear only C, and the source C is now in front of us. And round and round we go...

Same procedure for the other speakers (outputs) and you will notice that by some miracle the sequence of switch input always appear twice, so those will be in the same chip,

I have only drawn 2 output resistors, but to get to 36 db you need 4 resistors (one from each speaker (attenuator) output). The limitation of only 36 dB range is a bit disappointing, if you need more then you will need to spend 4 x 2$ extra for 8 chips and get 70 dB, there you are right. But I doubt you can hear a 36 dB down positional error over 45 degrees ... How much do we decrease / increase gain? Well sound goes a straight line from on speaker to an other, and halfway must be 50/50, and half of half way should perhaps be half of 50 / 50, so maybe no need for sine lookup tables either but that requires deep mass something I would do once I start coding... else try to avoid, but [sound attunates] square of distance or something? Oh this get really complicated.

Hope this is a more precise answer to your question.

Reply to
Jan Panteltje

On a sunny day (Mon, 29 May 2017 18:10:17 GMT) it happened Jan Panteltje

PS an other aspect of doing it this way is, that unless I am mistaken, all chips (that have identical i2c addresses) can be on the same bus doing the same thing. That leaves about 6 bytes for a 45 degrees change. If one can push i2c speed above 1 MHz then the switch-over may be above human audible frequency. Else one could add a 4016 switch or something at the output (mute a click?). I have run chips at much higher than specified i2c speed. Dunno about the internal switches of that chip.

Reply to
Jan Panteltje

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.