KORG ARP 2600 Clock Input mod for Sequencer / Arpeggiator

[dksynth]

Do this at your own risk. Some of this procedure is very tricky. PLEASE do not risk your synth if you do not know what you are doing.
If you are in Canada I can do this mod for you, I need just the keyboard panel shipped to me. It's 4 screws and 2 headers that unclip.

I started taking a look at my Korg ARP 2600 a few weeks back to see if I could add a clock in for the sequencer/arp. Well you can!
Note that the gate length will remain a set 50% regardless of the input pulsewidth.

Here is the gist:

I decided to repurpose the mult jack of the upper voice output. I'm disconnecting the top one and reconnecting it to be an input into my conditioning circuit. I connected the square output of the LFO to the switched leg of the jack so that when nothing is plugged in we still have the standard behaviour.

here is my conditioning circuit alongside a quick layout visualization that may help you get it out on strip board:

visualization.png

Now for the fun part.

Here is the keyboard PCB board:

133150954_215852763451538_6954502361045504456_n.jpg

the area of interest is in the red box.
Here is how it looked before I got my sticky hands in there:

131466454_395773531716562_7205191993929718326_n.jpg

the resistors marked 152 and 562 and both diodes must be removed. IF YOU HAVE NOT WORKED ON SMD BEFORE STOP HERE! Get someone who has experience, please. This is a lead free, relatively delicate board. DO NOT ATTEMPT unless you have done this stuff before.

These three pads are where we are attaching the OUT of the above conditioning circuit. This point is the input to the MCU that it reads for the tempo of the sequencer/arpeggiator.

new connections.jpg

the result should look a bit like a miniature Gort's missing appendage. If it looks too impressive, you used too much solder.

3-Best-100-Robots-in-Film-Robot-Gort.jpg

This connection will be the most delicate, long term. I recommend using a stiff, stranded wire that holds a bend and hot glueing it immediately on the other side of the board to ensure that it does not move or vibrate, weakening those SMD pads. This is also why we are soldering to ALL THREE pads.

Next we need to prepare the jack we are repurposing.
You need to remove R444 on the top side:

132665375_1543744665830836_9125611200250233534_n.jpg

and connect the square output to the switched lug of the jack:

133683736_885170085566814_3957483033795999086_n.jpg

the IN of the conditioning circuit should be wired to the input lug of the jack.

power can be tapped off of what I assume is the programming header on the bottom corner of the board. This is very important to do correctly as this is what limits the CMOS circuitry to condition the input to the same voltage as the MCU.

This is how I ended up wiring it:

133130879_194465352350783_6811270773535272093_n.jpg

I used one of the unused mounting holes at the top for a brass standoff and a stick-on plastic one to hold the bottom end secure.

And that's about it. If there is interest I can make a small PCB for this but really it is so simple to do on stripboard I don't really see the point.

If you have suggestions to improve anything please get in touch.

[dksynth]

this is the PCB I've put together with some component changes to change trigger levels and notes.

I think DSLman will be trying this as well and may have something available for the EU area.

clock mod.png

[KSS]

[PPE]

This is awesome. I could be wrong but aren't Korg working on a firmware update that will allow clock input?

[LED-man]

This was discussed in February/March on GS.
the problem is one missing cable from the Keyboard to the 2600.
Korgs Information about an clock Sync „Future update“ is wrong.
But this are 2 different things.
MIDI clock Sync and Keyboard analog Input Sync.

[PPE]

Gotcha. Very useful mod indeed, nice work!

[dksynth]

one thing I did not note more clearly:

The gate output PW is permanently set at 50% regardless of trigger / gate input length. I got some weird behavior modulating PW of the clock but I didn't do very thorough work on this part, I just conditioned any input into a very thin pulse for the MCU to read and it worked so I went with that. I will do further experiments in this area when I have time... It would be great if the sequencer gate PW matched the input clock PW.

[dksynth]

I have my ARP 2600 open again to check on some other things and I finished tracing out the CPU line from the LFO.

this is the input to the CPU with the part numbers involved:

dghdghhghdgh.png

So my modification is a bit over-cautious, looks like an least D406, R429 and R430 could stay installed and only D405 has to be removed.

As before, use any info here at your own risk and I don't guarantee that it is error-free.