Antichambre

MIDIbox 2069

20 posts in this topic

Hello,

Here, it's all about the NJM2069 and its MIDIbox Module.
Here an old topic which is talking about too, I mention it too in the TIA Cartridge topic as a possible option for.
First some words about the chip.
Some of you already know it. It's present in some KORG products from the 80s, In Analog Hybrid Synths like the DW and EX series, DW-6000, DW-8000, and the Poly800...
I've got the smallest one, the EX-800.
You can find it too in Some samplers of the same brand, in the DSS-1 for example.

The NJM2069 is Music Voicing System, like SSM brand used to say. Very close to the SSM2045, same functions but not the same tech.
It's an OTA based LPF but it's not just a filter.
It is composed of 4 VCA and 1 VCF Low Pass 4 Poles. 2 VCA are fixed on the VCF inputs named SIG1 and SIG2. 1 is used for the Resonance. The last VCA if fully independent.

The VCF part has 3 Inputs:
- SIG1 Signal 1 with VCA
- SIG2 Signal 2 with VCA
- VCF IN is a direct input to the VCF.
2 Outputs:
- 2P, 2 poles Output
- 4P, 4 poles Output
and 5 CV Inputs
- SIG1 Level
- SIG2 Level
- Freq Log
- Freq Lin
- Resonance

The Independent VCA Has is own input and Output and 2 CV inputs, first is linear CV the other is labelled Log in some version and seems to be a bias on another and is finally not usable, i think.
Here the function diagram and pinout.
5999ec9e8c645_NJM2069Diagram.thumb.jpg.4

I didn't find too much documentation, no datasheet. And some written information on the internet are wrong.
For example i read somewhere that the chip is not able to self oscillate! It's wrong! Mine on a bread board do it very well ;)
Not too much data then i used the service manual of the Poly-800, DW-8000 and DSS-1 to prototype my board. And after some testing, analyze... Finally design a first PCB version.
DDS-1 Service Manual
DW-8000 Service Manual
Poly-800 Service Manual
...
Enough for today, i will continue later ;) ...


 

Edited by Antichambre

Share this post


Link to post
Share on other sites

Now let's prototype it!

59b99eddae4e1_Fichier13-09-2017130521.th

 

I used Poly-800 Service Manual page 6:

fetch.php?w=600&tok=6ac960&media=anticha
In the Poly and EX-800 the 2069 is in PRE configuration.
It means the VCA is before the VCF. The VCF has 3 Voltage controlled inputs, SIG1, SIG2 and VCF In.
Two VCO on SIG1 and SIG2 and a noise circuit is applied to the VCA In.
The VCF 4 Poles output is the Output of the filter stage.
fetch.php?w=400&tok=724389&media=anticha

The DW-8000 Service Manual page 22:

fetch.php?w=600&tok=85316d&media=anticha
The DW-8000 is in POST config.
The VCA is after the VCF Two VCO come to SIG1 and SIG2.
The Noise Circuit is directly applied to VCF In.
The VCA Input connected to the VCF 4 poles output.
The VCA Output is the output of the filter stage.
fetch.php?w=400&tok=4faa1c&media=anticha

The DSS-1 Service Manual page 18:

fetch.php?w=600&tok=ed0c99&media=anticha
The DSS-1 is in POST config too.
The only difference is the 2 switching Op Amp added between the VCF Ouputs 2P/4P and the VCA In.
fetch.php?w=600&tok=52cede&media=post_2p

 

Specifications

- Digital and Analog control for all CV, both at the same time.
- Configuration modes are controllable.
- Power supply is +/-12V.
- Can be use in a desktop synth or in a Modular system.

After some schematic sketching this is the wild version.
fetch.php?w=600&tok=96e4ee&media=anticha
fetch.php?w=600&tok=4c89a6&media=anticha

To be continued…

Edited by Antichambre

Share this post


Link to post
Share on other sites

Nice start! 

I have questions but they can wait until you've progressed a bit.

Best,
Andy

Share this post


Link to post
Share on other sites
53 minutes ago, latigid on said:

I have questions but they can wait until you've progressed a bit.

Tell me. I will continue it a few today.
 

Share this post


Link to post
Share on other sites
1 hour ago, latigid on said:

let's see what it can do!

13 hours ago, jh0n said:

Can't wait!

It's just a Yocto pattern at the input (THX @Shuriken) and I just play on the breadboard trimmers(CV inputs), with a screwdriver ;)
Please take care of your ears and speakers, it's very loud during self resonance at low and high frequencies, I push it in its limit.
...
Grrrrr... Stupid algorithm on soundcloud removed my track for copyright reason!?  I upload it 2 times :/.
This is it:


 

Share this post


Link to post
Share on other sites

Ohh! they finally reinstated the first one then this is another test:

 

Share this post


Link to post
Share on other sites

Cool Bruno, I like it :)

Share this post


Link to post
Share on other sites

Great job, Bruno!
Those Korg filters are nice, never owned a DW-8000 or the likes, but had them on the watchlist for a while, lost an auction for a probably damaged EX-8000 a few weeks ago, now by coincidence you turn up and make a filter, awesome! :-)

Many greets,
Peter

Share this post


Link to post
Share on other sites

:blush:
...

Precious information found once prototyped.

The SIG1 and SIG2 has inverted signal audio inputs.
It seems all VCA cells invert the signal, then same thing for the VCA part.
? It will need some OP AMP, this is sure now…

fetch.php?w=600&tok=0b0608&media=anticha Two  Inverter OpAmp, one for each SIG input.

 

fetch.php?w=400&tok=3581ef&media=anticha One for the VCA
We can observe in DW-8000 and DSS-1 Service manual that the VCA output has already an Inverter OpAmp Stage.
I did the same, I suppose original circuity respects the internal output Impedance.

 

LOG CV Input of the VCA seems unusable.
I've got only some NJM2069BD for spare then I used the D version present in my EX-800 to test Log CV of the VCA in the 2 versions, and it appears to be definitively not usable for both!
Maybe some other version (AD, DA…) of the chip release this function. I don't know.
? But I respected and reproduced the two different existing circuit. This pin is to ground in DW-8000(AD)/EX-800(D) and acts as a kind of bias(?) in a DSS-1(BD).

Choices made during prototyping.

- SPDT function added to the VCF outputs to switch between -12dB and -24dB… Of course.

- SPDT for ByPass function… Of course too!.

- Polarized caps to all audio inputs of the module, SIG1, SIG2, EXT(named like that to be clear) and 1 polarized cap on the output.
I checked, with some low offset Op Amps like NE5532 or NJM4580, this is enough to maintain offset and stability in all the circuit. Then no need of big and expensive bipolar caps between stages in circuit.

- Remember, I want the switching between the configuration modes digitally controllable. (MIDI, Recall…You know why ;) I try all the valid configurations, and finally choose the 4 most interesting.

...

Edited by Antichambre

Share this post


Link to post
Share on other sites

I forgot more important things I found.
- The Absolute Rating Voltage is +/-5V, tried and fried ;)
- CV inputs are in a 0 to 5V range.
- All audio inputs are bipolar and support OverVoltage. A good hard clipping.
- LoG Cut-Off CV has inverted input. Means Cut Frequency is low at 5V, High at 0V. This one must be inverted too.

The Configuration Modes now:

- Remember, I want the switching between the configuration modes digitally controllable. (MIDI, Recall…You know why ;) I try all the valid configurations, and finally choose the 4 most interesting.

fetch.php?w=400&tok=b03e36&media=anticha
Mode 0 - PRE, VCA before VCF

fetch.php?w=400&tok=5f1d1f&media=anticha
Mode 3 - POST, VCA after VCF

fetch.php?w=320&tok=74a041&media=anticha
Mode 1 - PAR, VCA Parallel VCF
They share same Input and Output.

fetch.php?w=320&tok=60c27c&media=anticha
Mode 2 - SUM, VCA + VCF,
They share only output.
VCF direct input is tied to ground.


For this purpose I use 3x DG413 Analog Switch on the breadboard.
I reduced it to 1x DG333 and 1x DG413 on PCB.
All Audio purpose active components like OpAmps and Analog Switches are connected to +/-12V. We wants 2069 distortion only!
The 2069 is supplied by +/-5V, then it will be the first to saturate.

This is the block diagram of the switching circuit
fetch.php?media=antichambre:block_mode.j
Here the circuit is in default state.

The 2P/4P switch has its own line control.
The two Bypass SPST form a SPDT Switch and have their own control line too.
All others 5 switches are for Mode purpose.
5 is too much lines, remember I've got only 4 modes then I may reduce it to 2bits, 2 coding lines.

I inserted the Switches in the circuit to get this truth table:
fetch.php?w=200&tok=1a0e68&media=anticha
As you can see E=B and D=C then it's reduced to 3 lines.
Now I add my 2 required inputs lines, ModeA and ModeB
We can extract that:
fetch.php?w=400&tok=77e8cc&media=anticha It's ok, I need 1x OR gate and 1x AND gate, on PCB I use 2 small 74HC1G32 and 74HC1G08.

And finally I obtain to control everything with 4 lines:

  • 1 logic line for Bypassing, Active at High.
  • 1 logic line for 2P/4P switching, 4P at Low, 2P at High.
  • 2 logic lines to code the 4 Configuration Modes.

Obviously you're not obliged to use this switching circuit. All I/O are individually accessible on pin header.
You can use it to access I/O or to set it in the static configuration of your choice.

To be continued…

Edited by Antichambre

Share this post


Link to post
Share on other sites

Yeah! It's fine!!!
But "Everything" I don't know as you can see I use it in a very minimal configuration ;)
I'm sure it's working fine on USB power, I tried a direct EXT 5V power to the Main board, fine too of course!
I try it before you explain it in the Wiki and we did the same thing, opening BOOT1 surface jumper, removing OSC2 components and FLG/PA5 jumper to get J16 working.
I made another test for J18, with 2 others LPC Cores on the same CAN bus, It's working too  ;)
You respect the original discovery pinout so... no bad surprise during bootloader and firmware setup.
You thought it simple and easy, this goal is clearly achieved! :)
 

Edited by Antichambre
1 person likes this

Share this post


Link to post
Share on other sites

like it. will there be a pcb run... or what is the plan?

Share this post


Link to post
Share on other sites

Yes, + CS and Euro Panel, in a 'Limited Deluxe Edition' :grin: , cause I've got a limited stock of NJM2069! it will be available for Christmas or later.

Best
Bruno

1 person likes this

Share this post


Link to post
Share on other sites

yes cool. hope to be in the limited run... thil xmas then.

Share this post


Link to post
Share on other sites

So your planned module is gonna have three inputs, all under VCA-control? Cool!

Share this post


Link to post
Share on other sites

Posted (edited)

Yes, there's 3 VCA inside the 2069, would be a shame to do not use it.

Best regards

Edited by Antichambre

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now