While I’ve had the synth chassis separated from its roadcase, I felt that it would be a good time to freshen things up a bit.  The I/O and power supply panels were somewhat gunked up, the mixer module could stand some care and cleanup, and the tolex roadcase had its share of rips and tears, fading, and the wood frame had pulled apart on the front left corner.  Black cloth tape was covering some of the really rough parts, and that was starting to peel away in areas.

I first disassembled the I/O and power supply panels, cleaning off label residue and other grime.  The jack nuts and metal hole-filler plugs were soaked in WD-40 and polished up with some #0000 steel wool prior to reassembly.

Before:

After:

Part of me wants to replace that old power inlet with a modern IEC-320 type connector.  This one poses a bit of a safety risk in that the female mains connections (on the cable itself) sit flush with the connector’s surface.  As a result, it looks like it would be very easy to get zapped by the business-end of that cord if being handled while left plugged into the wall!  Then again, once I have this synth in place it won’t be getting moved around much.  Guess the jury is still out on this one… (well, at least until I accidentally electrocute myself =)

Moving on to the mixer module, I cleaned that up in a similar fashion.  It needed its tantalum power supply decoupling caps replaced with new electrolytics, the usual old flux cleanup, and its pots were kind of gummy-feeling.

The knobs on this unit are of the collet variety, and so removing them involves carefully popping the caps off and using a nut driver to loosen the collet bolts – after which the knobs can be pulled free.

After removing the PCB and replacing the caps, I addressed the gummy pots.

Now, there are some strong opinions about the right and wrong way to “clean” potentiometers.  Some techs will remove them completely, carefully pry the tabs open, pull the covers off, disassemble and wash the resistive element.  That approach is no doubt thorough and will likely yield the best results.  Its also time consuming, and introduces some risk (breaking tabs, PCB traces, etc…) if not executed carefully.

Being that the pots on this unit weren’t especially dirty, I didn’t opt to go that far.  I’ve been using Caig products for some time, and I feel that if used judiciously and sparingly good results can be achieved.

I used an air duster to gently blow any dust out of the pots, and followed that with light application of the FaderLube products – first the F-5 to flush out the pot, and again followed with the air duster to blast out any excess.  I then applied a light treatment of the F-100 concentrated FaderLube to lubricate the pots.  A few twists side-to-side, and these things feel fantastic now!

Note that I always err towards the FaderLube products when cleaning pots, as I’ve read that there is a chance that DeOxit may cause harm to the carbon resistive elements (see here for more info:  http://store.caig.com/s.nl/ctype.KB/it.I/id.2919/KB.215/.f).  I use DeOxit only for cleaning dirty jacks, switch contacts, and keyboard bus-bars.

With the pots taken care of, I gave the board a good scrubdown with isopropyl alcohol and reassembled the mixer.

Now on to the enclosure…

I wanted to keep this thing as original as possible, and as such I didn’t want to go as far as re-covering the roadcase.  The majority of the tolex was still in serviceable shape, and I felt confident that I could touch up the rough spots.

Using some Weldwood contact cement, I worked over the rips and tears and got them secured…

Lots of adhesive residue was present on some of the rough spots, where a black cloth tape was used to cover up damage…

After getting the tolex glued back down, mineral spirits and elbow grease got rid of any tape residue.

I hadn’t ever attempted this before, but being that the tolex had faded a bit I tried a Liquid Leather touch up kit to re-blacken the enclosure (I had this sitting around for household-related duties).  It only took a little bit of the black dye on a dampened sponge, and it made a huge difference!

After that dried, I applied some of the clear sealer over the top.  This was included in the leather touch-up kit – not sure what it is, but it resembled Thompson’s Water Seal.  That took quite a while to dry.  It remained tacky for a few days afterward, but feels and looks like new now.

One other little trick I tried here was using a spray-on rubberized undercoating (sold at auto parts stores) to fill the gaps in some of the damaged areas of the tolex.  I didn’t actually spray it on, but rather sprayed it onto a piece of cardboard and used a brush to apply it carefully in places where bare wood was exposed.

Here’s a before shot of the busted-up front left corner…

…and after gluing it back together and covering the exposed wood with the rubberized undercoat:

Granted, its not perfect… but it looks much better than it did!  That stuff dries hard and can be sanded to the appropriate contour, and if enough time was taken it could probably even be textured to match the tolex.  I was feeling only moderately obsessive-compulsive that day, and so I cut it short there.

Here’s the finished result:

That’s all for the beautification efforts.   At present, I’ve taken to sorting out the programmer module.  I’m going through now and recapping and proactively replacing a number of the old op-amps in it, and I will be troubleshooting the memory store/recall problems next.  More to come on that…

Getting the Four Voice’s keyboard assembly out involves excavating the synth from its integrated roadcase.  There are a total of six machine screws on the bottom of the tolex-covered enclosure that have to be removed, as well as the pair of handles on the rear of the unit.  The entire aluminum frame can then be lifted up and out of the wood enclosure.

The whole chassis by itself was a lot lighter than I had expected!  This thing is a real beast to lift, and now I realize how much the roadcase accounts for that.

Obie in the raw!

The Pratt-Read keyboard sits on a pair of stand-offs, and a series of machine screws secure the whole assembly to the enclosure’s aluminum base-plate.

Upon removal, I noticed another PCB hiding there underneath – mounted on spacers that appear to be JB-quik-welded to the underside of the keybed.  Just when I thought they couldn’t possibly have stuffed any more CMOS into this thing… =)

The harsh action and uneven keys are a result of the old rubber key bushings becoming brittle, to the point that many have started crumbling away.

The bushings can of course be replaced without disassembling everything, but I thought it best here to tear this keyboard down and really do it up nicely.  It’s a tedious job, but the end result is worth it!

Starting with the key-tops, I began stripping it.

Round-nose pliers work great for removing the key return springs – and it ensures the spring ends won’t get nicked or damaged by conventional pliers in the process.

Two different tension springs are used on the black and white keys, and they can be identified by color (red=white key, blue=black key).

The plastic j-wire clips need to be detached from their respective key actuators…

…and then the metal key frames can be pulled free from the keybed.  Note the thirty-plus years worth of crud, and the disintegrating key bushings.

Simply touching some of the old bushings caused them to crumble away!

Here’s the keybed after a thorough delousing and with new key bushings installed.  These were purchased from Vintage Vibe, and they came “pre-lubed” with the recommended Dow Corning 7.

Meanwhile, the keytops were soaking in warm water with dish detergent.  I used a small brush to scrub them down, rinsed, and let them dry thoroughly.

A cloth dampened with naptha worked wonders in cleaning the key frames.  Some keys had a bit of surface rust on the actuators, but this came off easily…

Everything was clean, and so began the reassembly.  Key frames went on first, and I used a touch of lithium grease on their fulcrum points to ensure smooth action.

There didn’t appear to be any triggering problems prior to this refurb effort, but I still gave the busbar and j-wires a once-over – using a swab dampened with DeOxit D-5 contact cleaner.

I have a bottle of this wonderful stuff called 3M Finesse-It II (mfg part number 39003) that I’ve used to buff out guitars, smooth out minor finish repairs on automobiles, get rid of scratches on beat-up rental DVDs that skip… you name it!

Prior to reinstalling the key tops, I had the thought of polishing them up with this compound.  None of the key tops were particularly rough, mind you, but a few decades of wear and hairline surface scratches left them looking kind of dull and tired…

…and the finishing compound more than fixed that.  The key tops look brand new now, and they feel smoooooth!

Some adjustment was still needed, as can be seen below:

The prongs on the keybed can be manipulated with needlenose pliers to adjust the lateral and vertical position of each key.

End result is a level keyboard with consistent action across its length.

With the mechanicals finished, I shifted my focus to the electronics.  It looks like this board handles the duties of decoding the key-presses (contains several 4028s and 4011s) and sends the resulting state changes over to the polyphonic keyboard module.

A tantalum bypass cap can be seen here:

That was refreshed with a similar value electrolytic, although I had to mount it on the opposite side of the board due to space constraints.

The CMOS on this board was socketed, which is great.  I’m planning to re-chip this before I put the synth back together.  Might as well, and save having to pull the case apart again down the road if the original CMOS (well past its life expectancy) bites the dust.

I took this opportunity to retrofit decoupling caps to each CMOS chip as well.  It apparently wasn’t a norm to do this back when these synths were built, but since CMOS can impart noise on the power rail when switching states its a good idea to add these.

And there you have it – good for another thirty-six years!  My next posts will cover the tolex enclosure refurb, mixer module and rear connection panel cleanup, the SEM and programmer module troubleshooting/repairs/calibration, and finally my new wiring to integrate the MS-1A sequencer and noise/s&h source.  (…oh yeah, and I’ll need to cleanup and re-cap that module too =)

Seaborne

The melody consists of a few multitracked passes of my Alembic Distillate bass running through a Maestro W-2 Sound System for Woodwinds, and an old Sola Sound Wah pedal that I’ve had most of my life (it was a gift from my childhood guitar teacher).  The synth stuff is largely Obie OB-8, with the Multimoog for sound effects, and there is plenty of RE-301 Space Echo scattered about.  Guitar solo at the outro is a Gibson SG.

Horizon

Equinox

The melody is being doubled in places by a Roland G-505 guitar controlling a GR-100 “Electronic Guitar” floor unit.  Pads are from a Yamaha CS-40M synth.

Passage to Infinity

SEMs 3 and 4 were not firing from the keyboard, and the continuous assignment mode would only step between SEMs 1 and 2.  Between all of the logic in the polyphonic voice assigner, the expanse of cabling and interconnects, and the programmer module sitting between the keyboard controller and the SEMs there seemed to be quite a few potential points of failure to have to sift through.

I found that no gate signals were making it to pins X2 or Y2 (the gate jumpers are bridged) on SEMs 3 and 4.  Back on the programmer, the green wires (gates) on connectors B3 and C3 (the outputs to the SEMs) and A3 (the input coming from the keyboard controller) appeared lifeless as well.  What was interesting, however, was that the keyboard electronics were asserting gates on connectors 3 and 4 – they just weren’t making it to the programmer.

Also, when comparing the known-working channel board on the keyboard controller (the one governing channels 1 and 2), I realized that the gate signals on the channel board for voices 3 and 4 were being asserted with every key-press.  It became apparent that even once I get the signals routing appropriately, there would still be a problem with SEMs misfiring.

While prowling around with the scope, it looked as though the CD4011 at U12 might be going flakey.  The NAND gate outputs on pins 3 and 11 would change state with every note played, compared to the known good channel board which would change state on every fourth note – as it should.  The two other gates on that chip seemed fine, and so it seemed to be in some partial state of failure.  I don’t think its uncommon for old CMOS to develop strange behaviors like this, and had the ICs been socketed I’d be very tempted to go through and re-chip these boards – just to reduce the likelihood of future failures.

Working my way backwards, I also witnessed that pin 9 on the 4024 binary counter at U2 was always low (i.e. this counter stage wasn’t advancing like the other two).  I honestly wasn’t sure if that was normal or not…  At that point I was already getting a little antsy since I had a feeling that 4011 was the main culprit behind the gating errors.  I did decide to replace the 4024 while I was in there, but in retrospect I probably should have left it be and made a better effort to understand the logic (the counter outputs still behave exactly the same – telling me the original chip was still okay).

I did get a kick out of the fact that the bad CD4011 had a special marking on it…  Was someone from the past trying to tell me something?!  :)

Meet the new U12!

After reassembly of the poly keyboard module for testing, the issue of the gate appearing on the 3/4 channel logic board at every key-press was gone.  Now, the continuous mode looked to be cycling through the SEMs correctly.  I was seeing CVs and gates for each SEM; however, I still had that problem of the gates not reaching SEMs 3 and 4 to contend with.

I’ll need to pull this module apart again to clean the scratchy pots, refresh the caps, and scrub away flux.  I’m also grappling with the decision of whether or not I should go through and re-chip (small sections at a time) and add decoupling caps.  I think I’ll stick to targeted repairs for now, but I might revisit this once the punch list is complete.

As an aside, replacing these chips can be a bit of a nail-biting experience!  The PCB traces are very fragile, and I’d really hate to ruin a board while desoldering components from it.  Everyone has their favorite approach, but I’ve had my best luck using the smaller (and gentler) variety of desoldering pump as seen in this photo:

I use Kester 44 solder for repairs, set the iron to around 725 degrees Fahrenheit, and I wet and heat the pad just enough to ensure the solder is melted on both sides of the dual-layer PCBs.  Most of the time I can completely free the leg of a chip in one shot this way.  When I’m trying to salvage a chip I’ll usually work from the middle pins outward, and alternate sides as I go – in an effort to minimize heat build-up in any one specific area (as this could damage the chip).

Here’s a photo of that CD4024 counter after extraction using this approach:

If a chip is obviously dead and/or is easy to source, I’ll usually just clip it out and use needle-nose pliers to gently pull the legs out while heating the pad.  That usually poses the least risk to the PCB traces.  In the case of that 4024, however, I wanted to try and retain the chip in the event it wasn’t faulty and/or if I ran into problems using the newer, buffered “B” series replacement chip (experts on one of the forums had clued me in to be cautious of this when replacing CMOS in the old Obies).

Returning to the task at hand – now that the channel logic board for voices 3 and 4 seemed to be gating correctly, I went on to trace out why the gate signals were not making it to those SEMs.

This unit looks to have had an external MS-1A mini-sequencer in tow most of its life, and came equipped with a 12-pin Molex connector on the rear panel for interfacing it (although there was no cable with which to connect the two).  As it turns out, this connection was hijacking the A3 input of the programmer (the inputs were presumably coming from the sequencer), and the gate and CVs from channels 3 and 4 of the keyboard controller were never making it there.  Those instead were being routed out to that external connector.  My dad happened to be reading this, and suggested they probably used a shunt plug in the connector to enable keyboard control when the sequencer was not in use.  I’m thinking he’s probably right!

It looks like a factory-installed option, what with the color-keyed connectors.  It was simple enough for now to unplug it and route the keyboard channels 3 and 4 back into the programmer input. The fourth SEM’s CV and gate can still be overridden using the 1/8″ jacks on the rear of the FVS, as per the schematics.

I’ll need to devise a way to integrate the mini-sequencer, but I’d rather it be such that the keyboard is normalized to all SEMs and those connections are only broken when the sequencer is connected (or I could build a shunt and reinstall that optional harness).  The sequencer also provides a noise source and sample and hold, and it would be great if I could also bring those signals into the synth and leverage them at the external inputs of the SEM filters.  That’s how the two-voice TVS-1 did things, as it had this same sequencer module built in.  I’ll have to give this all some more thought…

In the meantime, I now have a (mostly) operational four voice!  As soon as I made this wiring change and reassembled the synth, all four voices sprang to life – and what a lush sound!  The key split and various voice assignment modes inspire some creative possibilities… Makes me really anxious to get everything calibrated and install the new key bushings.  This is gonna be a keeper :)

There is still more to come…  Next up, I’ll be refurbishing the keyboard.

Before venturing too far down the path of sorting through the faults, I wanted to first ensure that the power supply was in good working order – that the power output was clean, and that the rails were putting out the correct voltage.

There are a total of six capacitors on the PSU board of the FVS - two large electrolytic power filter caps, two tantalum decoupling caps for the DC power rails (obscured from view beneath the row of black Molex power distribution connectors), and two ceramic disc caps used in the error amplifier circuit that is built into the LM723 voltage regulator chips.

With age, the electrolyte in the power supply filter caps can leak or evaporate, resulting in corrosion to nearby components, reduced capacity,  and a high-leakage current.  In other words, the cap can degenerate to a point where it is no longer doing its intended job.  The service life of electrolytics isn’t the only issue – these also really take a beating when filtering the mains AC!  Consequently, I try to replace power supply filter caps proactively in any old electronic device that I happen to be working on.

There is a chance that these caps have been replaced before, as their values of 2200μF (35V) differ from the 2000μF called for in the schematic (not a problem functionally speaking).  Then again, it could be that the factory started sourcing a different value cap at some point…  Without any knowledge of how old these are I’d rather err on the side of caution and just swap them out.

Now for the tantalums… Those buggers tend to fail to a closed-circuit state, unlike other caps.  Being that these 6.8μF tantalums are decoupling the 18.5VDC power rails (shunting noise to ground), that could be bad – a failure could short the corresponding power rail to ground!  Unless tantalums are found in a part of the circuit that is critical to the voicing/sound of the synth (and assuming a short resulting from their failure would be benign), I’ll replace them all with electrolytics.   Those are typically interchangable, and the electrolytics fail to a safe(er) open-state.

Ceramic disc caps rarely go bad, but I figure if they are used in voltage regulation (like the 100pFs here) and/or decoupling applications, I might as well just refresh them while I’m in there.

Before dismantling anything, I measured the power rails at+18.86V and -18.44V.  Measurements were taken when the power supply was under load by measuring across the following pins of the black Molex connectors:

Pin 1 – +18.5VDC  (beveled end of connector)

Pin 2 – Ground

Pin 3 – Ground

Pin 4 – -18.5VDC

Upon careful extraction of the PSU board (I avoided touching of any traces or leads), I took care to discharge the filter caps.   Even with everything disconnected, there can still be a charge on those caps!

I had built up a special lead with alligator clips and an inline 10W 100Ω resistor back when I was servicing a tube pre-amp.  Having a load circuit like this to discharge the caps  minimizes any sparking (and possible damage to PCBs), when compared with using a screwdriver, etc… to bridge the capacitor leads.

I simply clip this thing across the cap’s leads for a bit (10-20 seconds or so), and any surplus electrons stored on the negative plate will flow to the empty positive plate without using me as the conduit!   When both plates have an equal number of electrons, the current flow will stop, and the caps are safe to handle (good idea to check with a meter, but I admit I don’t always do that myself).

Photo of the board refreshed with new caps:

A good deal of flux had been left on some of the boards in this synth during manufacture.  By some accounts, this causes no harm and cleaning the board up is more a matter of good presentation (and can help mask repairs and such).  On the other hand, I’ve read of circumstances where flux deposits pick up moisture from the air and can introduce unexpected behaviors in some circuits.  I’m not worried about that here, but I can’t help but feel that the board is worthy of being left in a better (looking) state than when I found it!

Here it was prior to the flux cleanup…

…and after an isopropyl alcohol scrubdown:

I then reinstalled the board, first connecting everything _but_ the black Molex power distribution connectors.  I verified that the expected DC voltages were present on each rail, and used the trimmers to ballpark it to +/-18.5VDC.  Once everything was confirmed, I powered it down, hooked up all the distribution connectors, and then calibrated it more closely under load.

Next, I’ll be attempting to trace out the fault that is preventing two of the SEMs from firing…

I thought I’d start by tracing out and removing the modifications.  Overall, the mods looked to be in a questionable state, and I felt it would help in troubleshooting the primary issues if I could rule-out the modifications as a contributing factor.

The “Mediamix Triangle Mod” board was held in place by its two toggle switches, and old masking and duct tape was insulating it from the enclosure:

Gah!  I spy an alligator clip holding a bundle of wires down there at the bottom of the enclosure:

A few wires running from the (non-original) multi-pin Amphenol connector on the rear panel had liberated themselves:

Many of the mods were hardwired to the SEMs, making removal for service difficult:

The custom wiring created a bit of a rat’s nest inside of the case:

I also noticed a mounting screw from one of the SEM PCBs that had worked its way loose.  It was lodged between the power supply board and the enclosure!

The Mediamix triangle mod board after extraction…

Haha!  Duct tape insulator with a message…

LOL WTF!

Here’s the enclosure before and after removal of the triangle mod PCB and cleanup of all the tape residue:

Amphenol connector removed:

Ugh!  Have a 1/4″ bit, will travel…

(cleaned that up with a Unibit)

This is all of the custom wiring that I retrieved:

Looks much less cluttered inside now…

With the modifications gone, I set out to plug the holes that had been drilled in the enclosure.  I used some plastic plugs obtained from the nearby Ace Hardware store, and painted them with a white BIN primer that I had on-hand.  The hue was something like a bright white RAL9010, and the panels on the Obie are more of a cream RAL9001 hue.

I had some amber-tinted lacquer clear-coat left over from a guitar refinishing project, and I lightly sprayed the plugs with this to yellow them a bit and bring the color more in line with that of the panels.

I think the end result came out decent:

More to come in Part 3:  The PSU…

In or around 1975, Oberheim released the four voice polyphonic synthesizer (dubbed the FVS-1), which was basically comprised of four Synthesizer Expander Modules (SEMs; keyboardless, monophonic synths that were originally sold as add-ons used to augment Minimoogs, ARP Odysseys, etc…), a four octave polyphonic keyboard that could route up to four notes to the SEMs, and a stereo mixer module that summed the outputs of each SEM and would allow each voice to be panned anywhere within the stereo field.  All of this was housed in a huge tolex-covered road-case (with lid and integrated power supply), and there was even a four voice expansion cabinet offered that would hang off of the back of the keyboard unit – giving a whopping total of eight voices!

Oberheim added a programmer module around 1976 that included duplicates of most of the SEM controls.  Voltages from its analog controls were converted to digital, and their values could be stored in non-volatile memory (Intersil 6508 RAM backed up by a PCB-mounted battery).  This provided a rudimentary patch-storage system, and up to sixteen sounds could be saved for future recall.

There were some significant omissions in the voice settings that could be stored (probably in an effort to keep costs from spiraling out of control).  Parameters such as filter mode and resonance, oscillator waveform, sync, pulse width, oscillator modulation settings, and the pre-filter mixer settings still had to be setup manually across each SEM.  Only the basic settings (oscillator pitch, filter frequency, envelopes, and an LFO built into the programmer) can be setup centrally, saved, and later recalled during performance.  Combined with the polyphony, however, this was still a remarkable set of features for its time.

The Sequential Circuits Prophet-5 synthesizer emerged around 1978 followed closely by Oberheim’s own OB-X.  With their common set of digitally scanned controls (governing the timbre of all voices simultaneously) and full patch storage capability, both offered unprecedented programability while managing to reduce cost a bit in the process (the FVS-1 listed for close to US $5700 new!).  Needless to say, the Prophet and OB lineups captured most of the market for polyphonics – in some ways rendering the old SEM-based Obies instant classics.

All of these keyboards are still relevant today for different reasons, but the modularity of the early Oberheim polys is where they really shine!  There is a certain charm in the fact that each voice can be setup independently with its own dedicated controls (and individually panned!), and despite its limitations the programmer module can still be leveraged to provide quantized oscillator tuning, filter cutoff, and envelope changes across all of the SEMs simultaneously.

The other nice aspect of these early Oberheims is the fact that each SEM has patch points presented on its circuit board, and these can be tapped into to create a fully-modular polyphonic synth.  Many folks have built external patch panels that allow for re-routing of the SEM voice architecture, resulting in some complex and exciting sounds.  Nothing out there sounds quite like a SEM, and the two-pole multimode filters really differentiate the sound from most of the other Moog-inspired offerings of that era.

This particular unit arrived with an Oberheim MS-1A mini-sequencer in a standalone enclosure.  This is a  add-on module that can generate quantized control voltages (CVs) used to control oscillator pitch, and generate gate/trigger signals to control note-on/off for each step of the note sequence.  The MS-1A contains two independent 8-step sequencers, and also provides a noise source and sample/hold.

Like most 35+ year old synths, this Obie has its issues  (…and I intentionally bought this unit as a fixer-upper, so no surprises there!).  There were a couple of modifications installed, with quite a bit of additional internal wiring – some of which had broken loose over the years.  One of the mods added interface capabilities using a multi-pin Amphenol connector on the rear panel, and the other mod looks like it was intended to add triangle waves on the EXT #1 filter inputs of SEM 1 and 2.

Initial tests and observations show that the following areas are either malfunctioning or require some attention:

• The triangle-wave modification only functions on the EXT #1 input of SEM 1
• The Amphenol connector that had been added to the rear of the Obie has a couple of loose wires running from it, and an alligator clip was holding one of the wire bundles together
• SEM 1′s VCO 2 is silent (update: it works intermittently at times, pitch starts warbling after a while, then it dies)
• SEM 2′s VCO 1 is silent
• Only 1/2 of the SEM voices (SEMs 1 and 2) fire when played by the keyboard.  SEMs 3 and 4 can be droned by latching their VCAs on, but it does not appear that the keyboard gate or CV is making it to them
• Some of the controls on the programmer are unresponsive (the two envelope generators, and LFO controls), and the memories do not seem to store/recall a patch correctly
• The keyboard bushings are shot and need replacement
• The wood encolsure is coming apart in the front left corner (near the programmer)
• Many pots have a gummy feel, and some are scratchy
• The electrolytic caps in the power supply look to be original, and there are a ton of ancient mylar (some developing cracks) and tantalum caps (tend to fail in a closed/short state) in it.  These have all past their useful lifespan and should be replaced.
• I discovered that the original battery had been replaced (thankfully, as they can leak and destroy PCBs in the programmer if neglected) but the battery box that had been installed in place of the PCB-mounted battery is of questionable quality/condition

I’m going to take a phased approach to getting this unit back in top form, starting with removing the modifications and getting it back to an original working state.  Once I get it back to an original working state, I’ll find good ways to integrate the MS-1A sequencer and maybe eventually a patch panel – as this thing would mate up nicely with my modular!

I tend to take a lot of notes and photos over the course of a project like this, as it helps me to organize my thoughts and also leaves me with artifacts detailing the repairs.  This is the first new project I’ve embarked upon since starting this blog, and I thought it would be fun to leverage these notes and go into a bit more detail here than usual - and hopefully the info is helpful to some.  If nothing else, there will be a ton of photos here of a classic machine both inside and out!

More to come in Part 2: Modification Removal…