optocouplers to reduce digital noise?

[dorkus]

has anyone tried using optocouplers to reduce noise transmission from, say, a microcontroller to a serial-controlled audio IC like a crystal or TI volume control chip? i'm going to use a Atmel AVR chip to control a multi-channel remote control preamp i want to build using most likely the Wolfson WM8816 chip to control volume, which requires serial data to control it. i will keep the digital circuitry shielded in a separate part of the chassis, but want to minimize noise transmitted through the control lines as well. any thoughts?

thanx,
marc

[mlloyd1]

I had (actually still have; the laser is dead though ) an old Onkyo DX5700 CD player that did this. The white papers from Onkyo on the machine made claims for the benefits, but I had no way to prove or disprove them.
However, the machine performed well, was well designed electronically and well built physically, so I'd be inclined to beleive the claims were valid.

I, for one, would like to see your design when you're done, if you're so inclined
mlloyd1

[dorkus]

cool, i think i'll give it a shot. don't hold your breath waiting for my design though - i'm only just starting to get my act together and order an atmel dev kit, and i have grand plans to make a really cool multi-channel preamp, complete with remote control, menu-driven LCD display, etc... hopefully i'll have something working to show for it before too long though. i'll let you know how it goes.

narc

[dorkus]

ralf klassen was kind enough to send me info about his preamp project, which is very similar to mine:

http://www.die-klaassens.de/Audio_un...reamp_l10.html
you can translate it from german using google if necessary:
http://www.google.com/language_tools?hl=en

awesome!

[guido]

Hi,

Also build my own preamp, actually a kit from a dutch firm.
The remotecontrol part is my own design. I am using an Alps
motorised pot for the control, but that is not my issue.

What i would like to say: use a microcontroller (i used a microchip 16f84) which has a sleepmode. The 16f84 actually stops the oscillator when in sleepmode. Use the remote receivers signal to get it out of its sleep. In this way you have NO problems with interference. That changes if you use a display, those things have their own controller and clockcircuit..

Have a look at my website,

Guido

[capslock]

Hi Guido,

I visited your homepage & saved the RC5 code for possible future use. I have a Kenwood DPF-R4010 CD changer. If you think the code is similar, any hint about where to get the documentation would be much appreciated. Similarly, if anybody has information about the system remote code used in Onkyo tuners built in the late 80s, I'd be glad to hear about it.

Thanks,

Eric

[guido]

Hi,

The code for kenwood is a 2 wire full duplex bus. There are 2 versions, 8 and 16 bit. My tuner is 8, but it is an older model.

Got the info on paper and i don't own a scanner.. Have to get a friend to scan it in. The code itself is quite easy. 5 ms high per bit and also 5ms low for a 0 and 10ms for a 1.

Dont know if i have all the info on 16 bit, got 4 pages in total.
I have code for the pic16f84. It is quite old (written in atom basic, then 6502 assembly and finally pic). Not the best written code so to say, but it works fine.

More info will follow,

Guido

[Spartacus]

I was told by a BB engineer some years ago to avoid opto isolators in jitter sensetive situations - they are really bad in this respect. Isolators such as BB's own ISO150 are capacitive couplers, and much better in this respect.

He also told me that he has tested his own DAC (which uses ISO150s) with and without the grounds connected. Noise was lower without the connection. ie. with full isolation.

[guido]

Capslock and others,

Here is the info on the Kenwood XS bus. It is the green 3.5 mm plug on the back. This is all i have !! The code is using a PIC16F84 microchip microcontroller for sending codes.

Essentially i made a RC5 to Kenwood converter with it. I only have a tuner on the bus, i have no experience with more equipment and more then one busmaster.

Regards,

Guido

http://www.home.zonnet.nl/baltusg/Kenwood_1.bmp
http://www.home.zonnet.nl/baltusg/Kenwood_2.bmp
http://www.home.zonnet.nl/baltusg/Kenwood_3.bmp
http://www.home.zonnet.nl/baltusg/Kenwood_4.bmp
http://www.home.zonnet.nl/baltusg/kenwoodXS.txt

[dorkus]

yeah, i would never use standard optoisolators for digital signals... just for the control logic etc. if i were doing a DAC or similar digital project (i'm not), i would keep the digital lines as clean and straightforward as possible to minimize induced jitter.

for my project, i'm just trying to keep microprocessor noise out of the audio circuits, and i guess optoisos would help here.

Originally posted by Spartacus
I was told by a BB engineer some years ago to avoid opto isolators in jitter sensetive situations - they are really bad in this respect. Isolators such as BB's own ISO150 are capacitive couplers, and much better in this respect.

He also told me that he has tested his own DAC (which uses ISO150s) with and without the grounds connected. Noise was lower without the connection. ie. with full isolation.