Oberheim OB-SX Analog Polyphonic Synthesizer

before repair

Last Update 01-10-2015

Inherent Design Errors
How to restore an OB-SX
Correct Voicecard Calibration Procedure
How Does It Sound?
OB-SX Preset Names
OB Model comparisions

Back in 2000 Cary Roberts brought his eight voice OB-X to AHMW (the midwest gathering of Analogue Heaven members).  Although I was happy with my Moog Memorymoog, I was really impressed by the sound of the OB-X - thick, organic, edgy, different from the Memorymoog.  I used to own a Matrix-6R but it didn't sound near as good as the OB-X.  By the time I was in a position moneywise, fully loaded eight voice OB-X were not to be found anywhere.  Spare voice cards were extremely rare and expensive so I passed on the four and six voice varities that appeared.  I wanted a fully loaded eight voice.

Then this fully loaded six voice OB-SX came up.  The auction description said it was not completely functional.  Some research on AH revealed that it was a transition instrument between the OB-X and OB-Xa.  The voicecard architecture was identical to the OB-X but implemented with CEMs instead of discrete circuits.  See this section for the full specs.  Owners claim the voicecard had a lot of character in the sound.  While it possessed only a partial set of knobs and lacked full programmability (presets only), I decided to grab this one as they tend to sell for a fraction of their big brothers.

I landed a handyman's special alright.  I didn't care if it didn't worked since I had the resources to restore it, and it had been a while since I had a good project on the bench.  The list of foibles:

The store that I had purchased it from said that the unit had belonged to a high school, a good indication that it had seen rough treatment.  This was going to need some TLC.  I purchased this back in 2009 and started some repair, then it got placed on the back burner due to work issues and some serious health issues.  I did eventually land my OB-X in 2010 and that restoration took priority.  I didn't re-start the OB-SX project until 2014, which was actually good because I applied some important learnings from restoring my OB-X.

I had a real basketcase to restore and by the time I finished I had uncovered design errors and MANY defective parts.  Oberheims don't have a stellar reputation for reliability because there are design errors that contributed to malfunctioning units, and a lot of the components in these things do not age well.  The RCA CMOS ICs available back then lacked the input protection diodes of modern ICs and those old RCA chips often get blown.  I had leaking capacitors to replace, bad trimpots, bad quad opamps, bad 7442s and 74174s, bad front panel switches, and cracked solder joints to reflow.  Fortunately every IC has a socket which saved a lot of time and effort from desoldering ICs.  The beancounters at Oberheim cut a little too far - they installed really cheap open frame trimpots that exposes the resistive element, which corrodes with age.  Perhaps this was their application of "planned obsolescence".

The OB-SX uses mostly standard parts that are still available, aside from 80C98s and RAMs/ROMs on the control system and the CEM3340s, CEM3310s, and CEM3320 on each of the voicecards.  Fortunately the CEMs are very robust ICs, good thing because these are getting harder to find.

Unless they have been completely restored, pretty much ALL old Oberheims are malfunctioning.  If you buy one of these with the intention to restore and use it (or resell it), bear in mind that they don't have a high market value.  Do not pay too much money for one as you may not get a return on your investment after paying for new parts and paying your repair bill.  In fact, OB-SX buyers usually scavenge the voicecards for the CEM ICs which are very valuable parts on the market (for good reason - not much value in a synth with no programmability).  As I was progressing I even wondered if this was worth the effort.  Since I am an EE doing my own repair I was spared the expense of a repair tech, but few people are that lucky.

Add in a woefully inadequate service manual for troubleshooting the OB-SX and I can see why so many of these synths are parted out.  The manual is missing a LOT of important information, which I had added to this webpage.  You're welcome.

My old Tektronix 7834 oscilloscope crapped out back in 2005 due to a bad power supply and no spare parts to restore it.  But I still had the all the plugins from that scope.  It was clear that I was going to need my logic analyzer plugin for this project (to probe the computer busses) so I ponied up for a reconditioned Tektronix 7904.  I no longer needed the storage capability of the 7834 (now supplied with my Tek 2430A) so I went for the more abundant 7904 chassis.

I do not offer repair or restoration services so please do not email me because I am very busy and don't have time to address individual queries.  I do get emails from folks who want to know what my replacement parts were, so here ya go:
You can date an OB-SX by the date stamp on the voicecards, mine are stamped "Dec 17 1981".  Serial # does not reveal the date, mine is "920407 6 VOICE".

The service manual can be obtained from the world famous Mark Glinsky Manual Manor.  There is a remarkable resemblance to the Memorymoog in terms of the computer and DEMUX system.  The OB-SX is definitely a derivative of the OB-X.  The drawings for the controller board are identical with certain components omitted from the OB-SX.

Restoring a dead OB-SX

This section is applicable to other old Oberheims like OB-X, OB-Xa, and OB-8 not just OB-SX.  It is also somewhat applicable to other analog polyphonics like the Memorymoog, Prophet-5, Roland polysynths, etc.

How prepared are you for a job of this magnitude?  Take a look at these images below.  The one on the left shows the Oberheim in the middle of restoration.  See that box to my right?  That's the box of NEW components.  In front of me are the power supply, voicecard motherboard, and voicecards ready for repair.  See the image on the right?  Those are 3/4 of the components that were BAD (not shown are large electrolytic capacitors, disintegrated keyboard bushings, and many other ICs).  Not kidding, folks.  Restoring an old Oberheim polyphonic synthesizer is a labor intensive effort (plan extra $$$ for a competent tech) requiring about $300US in brand new parts.  It's a lot of work, but in the end you will have a reliable synthesizer that stays in tune and autotunes when you are done.  Do not expect shortcuts here.

The patient on the operating table (no, the KEYBOARD silly!)
The pile of replaced components (!!!)

The very first thing you should check with any synthesizer is the power supply.  Nothing will work unless it is 100% functional.  If any voltage rails are not working it can blow components downstream under the wrong conditions.  Inspection revealed all the voltage rails were working on my unit (use an oscilloscope, a DMM isn't enough).  The restoration work on the power supply consisted of:
Power Supply before
Power Supply after

Time to assess the livliehood of the voicecards.  Good news - all the CEMs on the voicecards are healthy.  Probing the inoperative voicecard revealed living and breathing CEM3340 VCOs albiet emitting an ultrasonic wave, an indication of a bad DEMUX section putting out high CV to the VCOs.  Since the ultrasonic waves were completely attenuated by the lowpass filter, that explained why no sound came from that voicecard.

The controller PC board was missing six of its eight mounting nuts.  It was hanging in the air at one edge of the board via the remaining two nuts.  That made the board flex, which probably didn't fare well on the copper traces.  It still locked up on occasion, so now I was chasing down the worst of malfunctions - the intermittent ones.  I found two causes - one of the spacers that separated the controller board from the top panel was longer than the others, which flexed that corner of the board when the mounting nuts were secured.  Proof that not all electrical malfunctions are caused by an anomaly of electrical nature...!  The other cause would be found later as I progressed with restoration.

Just to be safe, I completely disassembled all the boards and the keyboard assembly to search for loose mountings nuts as these can short out circuit boards - none found.

Several of the front panel PC boards had defective switches.  I am replacing ALL of the pushbutton switches to be safe, even the ones that work.  I have a stash of NOS replacement switches with different color caps, so I exploited that luxury.  I had black, white, blue, grey, and red switchcaps so I could've gone hogwild with a rainbow of colors.  Saner minds prevailed so I decided to go with something stylish.  The pic on the left shows the original black preset switches replaced with white and blue.  The middle and right pics show the colors I ultimately decided with - I like the blue caps with the blue pinstripes.  I reserved a red cap for the AUTOTUNE button so you can see it on a dark stage and it's a color you REALLY don't want to press in the middle of a performance!

Just Testing
Final #1
Final #2

After testing out these switches, I discovered to my delight that this OB-SX is the latest version with 56 presets.  Earlier OB-SX models (gray non-pinstripe front panel) had either 24 or 48 presets.  The 48 preset models had a slideswitch on the backpanel to select between two banks of 24 presets.  The 56 preset model omitted this slideswitch and you could use combinations of bank select switches (A, B, C, A+B, A+C, B+C, A+B+C) to access all the presets.

Besides the defective switches, I uncovered a defective DIP miniswitch on the preset selector PC board.  Oberheims have a history of defective DIP miniswitches.  This DIP miniswitch controls which voices are enabled.  If your Oberheim has voices that do not work, this is the first component you should replace:


Installed a new set of bushings for the keyboard.  The original bushings dry out and crumble, often resulting in uneven key levels and noisy weird feeling action.  I recommend removing the keyshells to access the bushings as you will be working with the assembly upside-down.  You'll probably clean the keyshells while they are out anyway - use a mild citrus cleaner.  And don't lose the screws!

Keyboard assembly before new bushings
Keyboard disassembled
Cleaning keyshells

The bushings have to be pried out with a pointed tool.  These things are over thirty years old and the bushings just disintegrate into pieces when removed.  An exacto knife or equivalent very sharp knife may be needed to aid the removal.  If you can't get the bushing out, you may have to remove the key channel.  This is tricky because they are secured by springs on the rear of the keyframe.  There are not only different springs for flat and natural keys (they are color coded, keep track which ones) but the #&%* things go flying if they slip off your pliers.  If you lose any, you'll have to find surplus ones as they aren't made anymore so be careful!

Bushing partially removed
Tool to remove bushing
Most bushings disintegrate upon removal

Installing new bushings is easier, since they are lubricated they just slip right on.  They do have to installed a certain orientation.  Note the orientation of the old bushing in the above pic on the left - the taller half is on top (bottom with the upside-down perspective).  New bushings feel MUCH better, and the keys usually even out.  Compare before-after pics.

Keyboard assembly after new bushings

Another common foible is flaky or cracked solder joints, and excess flux.  The voicecard motherboards almost ALWAYS need to be reflowed to repair cracked solder joints on the voicecard pins, a good example shown on the left.  This is another reason why old Oberheims do not have a stellar reputation for reliability!  You have to remove all the voicecards, then remove the motherboard to get to the underside where the joints are.  Inspection of the controller board revealed excess solder flux and suspicious flaky solder joints, which I cleaned up and reflowed.  These are UGLY.  Excess flux turns into a low impedance short as it ages and it can cause malfunctions in vintage equipment.  In the middle pic, compare the clean pad of the top half versus the dirty pads on the bottom half.  The pic on the right is the connector leading to the voicecards.  This is asking for trouble!


It's a REAL GOOD IDEA to do these steps BEFORE probing circuit malfunctions.  You don't want to waste time chasing down a ghost.  I wasn't kidding when I warned that a project like this takes time!

With the computer system working, it was time to replace the trimpots and electrolytic capacitors.  Any tantalum capacitor on any power rail has a history of going bad with age so they have to be replaced with electrolytic caps.  Electrolytics are polarized so attention is required so they are not installed backwards or you will have fireworks when you apply power!  I left the tantalum caps in the audio path intact (usually at the end of the autio path on the voicecard), they are not subject to the large voltages like power rails and I wanted to retain that "dirty" sound that Oberheims are renowned for (tantalum caps are not renowned for their high fidelity).  I uncovered leaking S&H caps (note the leaking residue on the leads on the caps), so those also got replaced.  If you find any of the old ceramic caps like those shown in the middle pic, replace them all as the leads can crack and break off the component.

For a system this complex, it is best to divide up into separate sessions and check that the computer still works.
  After you finish, apply power and verify that the computer still works.  Then replace trimpots and electrolytics on the voicecards. 

Bad S&H caps
Broken ceramic caps
New trimpots and electrolytic caps
New S&H caps
leaking-caps broken-caps

Next step is building a proper regulated voltage source for the (+)5.6VDC busses on the front panel pots and the 4051s just after the DAC on the controller board.  The original (+)5.6VDC circuit is not a regulator and is a big reason why old Oberheims are malfunctioning.  I was planning some modifications so I chose a board large enough for expansion (name the Utility Board).  I chose to install this new board above the power supply board, so I got standoffs to support the new board that mount right where the mounting screws were for the power supply board.  Don't block the heatsinks on the board, these need to be clear so that heat can passively escape.  I used an LM317 based regulator, just google for the datasheet for information on how to design the circuit.  Install two terminals for the (+)5.6VDC outputs - one will go to the front panel pots, the other will go to the controller board.  Use an oscilloscope to confirm a clean DC level, a DMM isn't enough.  You must identify and remove the components on both boards that derive the (+)5.6VDC buss, then install terminals to the correct points to receive the new (+)5.6VDC source.

Utility Board
New (+)5VDC Regulator circuit
Utility Board installed above Power Supply

The Oberheim System Interface has long been obsoleted by MIDI and at this point I decided to remove the components associated with that circuit.   The reason was to relieve the power supply of current which can be used for my new modifications on the utility board.  After studying the schematics, I determined that I could simply remove every IC on the drawing for the System Interface.  The *BUSRQ line that tells the CPU of new data already has a pullup resistor intact, thus removing the ICs simply left that line inactive and the computer should function normally.  This worked like a charm and relieved the power supply of about 650mA of current, a significant chunk.

Also to go due to obsolescence were the CV/trigger input jacks.  They only worked on voice #1.  I found a MIDI retrofit (see the OB model comparison table at the end of this page) so I installed MIDI jacks where the CV/Trigger jacks were.  You have to perform a little "normalled" wiring on the controller board when you remove the jacks (connector E: short E1 to E8, short E2 to E5/E6).

With all the power rails confirmed and new trimpots.caps installed, now we can dig into the meat of the malfunctions.

I started probing the DEMUX system - this is the where the multiple control voltages (and most malfunctions) for the voicecards come from.  It consists of the DAC followed by a set of 4051 CMOS DEMUX switches whose outputs are tied to S&H circuits and signal conditioning circuits.  There is a lot to go wrong here so if few of the presets sounded musical, this is the first place to look.  I had plenty of experience debugging my Memorymoog which is ten times as complex as this thing.

I knew I was getting keyboard CVs so that confirmed that the DAC is operative.  I had a bad keyboard CV on that inoperative voicecard.  I traced it to a TL084 quad opamp after the S&H circuit.  Fortunately I had a couple in my stash.  Pop in a replacement TL084 and that voicecard is now operating!

Next was to trace the fault of the CV controlling the pulse width of the oscillators.  The autotune uses the pulse waves and I could not find any pulse waves on any voicecard.  There are a lot of components to condition the pulse width CV so this is going to take some detective work.  That turned out to be another defective TL084 quad opamp - the replacement one was my last in stock.  These things eat TL084s!  This pic shows the two new ones in sockets 47 & 48:


One problem with troubleshooting this kind of system is that the schematics are so poorly drawn that I had to re-draw them to get a logical overview of the embedded system.  This re-drawn diagram is very large and not capable of being scanned.  From that I was able to isolate certain sections for troubleshooting.  It's also a good troubleshooting tool for a CPU system that locks up because you can identify and isolate any component tied to the data buss or the address buss.   I used this to pinpoint a rogue component shorting out the data buss on my OB-X.

 The OB-SX is starting to sound musical.  All six voices are working and I restored the pulse waveforms on the VCOs.  I had located a copy of the user manual with the preset names.  Rolling through the presets definitely confirmed progress - they are starting to resemble the sounds of the preset names.  Not all of the CV and logic signals were working yet - the preset names help to give a clue what to expect IE percussive sounds should decay not sustain, string sounds should have dual VCOs with some amount of detuning, etc.

The front panel pots were inoperative.  Quick probe of the front panel pot circuit reveal that all the components appear to be working.  The culprit was the LM311 comparator since it wasn't switching the MPS5172 NPN transistor on the D7 data buss.  Thanks to my mischievous cat, my bag of brand new LM311s went missing so I had to order another (Grrrr...).

By now I have concluded that I do not have the most recent schematics for this model as there have been changes to the computer board.  The sole unison CV circuit has been omitted (empty component and IC pads) which have been replaced with the individual keyboard CVs all firing the same CV.  It was redundant anyway.  While the original schematic show the omission of polyphonic glide, it is definitely present on this model.  I would ascertain that the 24 preset gray non-pinstripe panel OB-SX does NOT feature polyphonic gilde, as the preset that features polyphonic glide is in a preset bank which is not present in the 24 preset model.  Veddy Inkteresting.

Unfortunately I have yet to uncover later schematics that correlate to my unit, as it would cause me grief later when I found that the calibration process doesn't work for later OB-SX...

More probing of the DEMUX system revealed a defective 4051 CMOS switch that was supposed to provide the CV for EG modulation depth to the filter.  I didn't have any spares so since the front panel pots were not yet working I scavenged the known working 4051 from that circuit.  That restored the EG filter modulation, NOW we're making progress.

By now it was obvious that the flawed (+)5.6VDC circuit that had a design error had damaged more than one CMOS and opamp IC, so I took the dive and ordered new components for everything - CMOS, quad opamps, TTLs (7442s amd 74174s also tend to fail).  It is very important to confirm good power rails BEFORE installing brand new ICs!

The best plan of IC replacement is to fan out from the DAC.  Replace the 4051s, then power up and verify.  Then the quad opamps, power up and verify.  Then the 7442s, 74174s.  Rinse, repeat, and recycle.  Divide into small sessions and you will keep your head from spinning.  Then probe CVs and logic signals at the voicecard pins, confirming that they change when you select different presets.  Then replace the TTLs in the computer circuits, the opamps in the CV and PWM summers, the LFO, etc.  Divide and conquer, divide and conquer.
  This is why a mass component replacement must be done strategically - the voicecards are dependent on everything upstream so you must verify 100% operation of the control board which is generating the control signals.

Oberheim neglected to include any verification scheme of the voicecard logic signals in the service manual - so just how do you confirm the integrity of the signals with a PRESET SYNTH?  I had to compile a table of known signals with different presets, which I'll include here.  Logic signals are TTL logic levels unless otherwise noted.  Refer to diagram #1 in the service manual for the location of the voicecard connector pins.

Logic Signal
Voicecard Connector Pin
Controller Board IC & pin #
Bank/Program #
SYNC=1 (15VDC)





A13 pin 2

A13 pin 2



A13 pin 5

A13 pin 5
VCO2L=1 (15VDC) D6


CMOD=1 (15VDC) B2


TRACK=1 (15VDC) D2



A12 pin 5

A12 pin5

A12 pin 15

A12 pin 15

A12 pin 2

A12 pin 2
LW2=1 (LFO square)

A12 pin 12
LW2=0 (LFO triangle)

A12 pin 12
PW1=5VDC (VCO1 off)
PW1=not 5VDC (VCO1 on)
PW2=5VDC (VCO2 off)
PW2=not 5VDC (VCO2 on)

Remember the problem with the computer locking up intermittently?  The other cause was bad RAM ICs.  As I was progressing with verifying CV and logic signals to the voicecards, I noticed that some logic signals for the voicecards (hard sync, XMOD, tracking waveform select, etc) sometimes worked and didn't work most of the time.  Probing the 74174 ICs that generated these signals revealed that they were working correctly, but the signals at the input pins were wrong.  Since the OB-SX control system was a derivative of the OB-X which used the same RAM ICs, I had an orphaned set of RAM ICs from my OB-X that were displaced by the Encore Electronics MIDI retrofit.  When I popped them in, the system worked.

Once you get all the CVs and logic signals working, then proceed with the calibration procedure for the controller board as stated in the service manual.  I used preset B6 (OB-SX Choir) to adjust the LFO rate, there is a "sweet spot" for this preset that will work for the other presets (especially strings).

Remember the design errors?  Sitting between the DAC and the CMOS 4051s is an opamp buffer with +/-15VDC rails.  On power transitions, the output state of any opamp is indeterminant and this innocent little puppy can swing anywhere between +/-15VDC rails as they are ramping up, violating the VEE-0.5 < Vin < VDD+0.5 input rule of any CMOS inputs downstream.  Even though this could occur in a split second, this will damage the CMOS gate and ultimately blow the 4051s over time (poof!).  This is another reason why old Oberheim often malfunction with age.

One other correction is due - install two clipper diodes on the output pin of the opamp following the DAC; I used a 1N34 to ground (anode to ground, cathode to output pin) and a 1N4148 to the new (+)5.6VDC rail (anode to output pin, cathode to (+)5.6VDC).  This forms a hardware limiter that prevents the opamp outputs from exceeding voltages that can damage the CMOS ICs downstream.  I chose the 1N34 because it is a germanium diode which will limit the low voltage to (-)0.3VDC, adding a safety margin to the VEE-0.5 term of the CMOS input rule.   During power cycles, even if the 5.6VDC doesn't fully ramp up the clipper diodes guarantee that the output of the opamp will never swing outside the CMOS VEE-0.5 < Vin < VDD+0.5 input rule.  Even though you will be installing new CMOS ICs with input protection diodes, this is another shield against damage.  I included a pic on how I tack-soldered the diodes on the solder side of the controller board, and don't forget the tubing to insulate the leads of the diodes to prevent shorts:


It's a good idea to confirm that the CVs can swing between zero and 5VDC after you install the diode clippers.  A good test is the filter cutoff pot on the front panel, monitor voicecard connector E5 while moving the pot across the full range (refer to figure #1 in the service manual for the location of the connector pin).

UPDATE I had to remove the 1N34 because it restricted the output swing to 0.3 volts not zero volts.

If your OB-SX doesn't autotune, DON'T attempt to troubleshoot it yet.  The autotune circuit is a complex closed-loop system and is dependent on proper CV generation and calibration of the voicecards.  In other words, it may work just fine when you finish in the end.

I lost at least an hour making music on the beast.  Now I am getting brass sounds, strings, EPs, organs, pads - this is progress.  Sounds really good, almost as good as that OB-X.  Perhaps this was a good purchase.

Once you get the controller board restored, then proceed with installing new electrolytic caps and trimpots in the voicecards.  If you want your OB-SX to stay in tune, then new trimpots are required.

There was a weird fault I had to fix.  If I turned the volume knob past eight, the pitch would drop.  That took a little hunting down.  Once I saw the CV drop at the output of the global summing circuit, I worked my way back to inputs and found the error on the master tune CV...?!?  This CV is going through A2, a 4052.  The volume control is a CV and is also going through the same 4052.  When I probed the volume CV as it went past eight, it measured over five volts.


And once again, the schematics have ANOTHER error.  The volume pot is NOT tapped from the (+)15VDC rail, it is tapped from the (+)5.6VDC rail.

The voltage rail I just corrected.

The VDD pin of the 4052 was tied to the (+)5VDC rail which measured 4.9VDC.  Within spec for the OBSX, but once again violating the CMOS VEE-0.5 < Vin < VDD+0.5 input rule because the volume CV was exceeding the VDD rail at (+)5.6 volts.  That meant I had to route the VDD pin to the (+)5.6VDC rail, which was tricky because a neighboring IC still needed the 5VDC rail so some trace cuts and wiring was required.  After this was corrected, the pitch no longer drops!

I was attempting to install the Highly Liquid MIDI retrofit in the OBSX but have not yet been successful.  The retrofit is not guaranteed to work in all polysynths, and the OBSX is different in that certain components were eliminated (remember, it's a "budget" synth!) which rendered the retrofit not successful.  Well, not yet...  I have tried some different interface circuits and have a direction to make it work, but at the moment I have higher priority projects and this took too much time.  More later...

By now I am cursing Cary Roberts for this awesome project (and for hosting my webpages).  I haven't had this much fun in years!

Calibrating the voicecards was a bit of a challenge.  The only service manual I have located is a preliminary version dated November 1980.  Not only is it missing crucial steps, but the voicecard calibration procedure isn't applicable to the later pinstripe model  with 56 presets (!!!).  The manual states that switch 7 and 8 on the DIP miniswitch will disable VCO1 or VCO2 - this doesn't work on later firmware!  If you calibrate the VCOs with either switch 7 or 8 down, then autotune will NOT work.  The correct procedure is to set both switch 7 and 8 off, which disables autotune (sets the autotune voltages to midpoint), disables frequency and PW modulation, disables unison/hard sync, and selects pulse waveshape - THEN autotune will work after you calibrate the VCOs.

New improved calibration procedure for voicecards:

Calibrating VCO pulse width:
  1. Best tool is an oscilloscope with graticule markings on the display, where you can probe test pins directly.
  2. This adjustment is critical for autotune.  The old OBs turn off VCO audio by setting the pulse width to zero.  If pulse width is not calibrated, the VCO cannot be disabled and autotune will not work.
  3. Set both DIP miniswitch switches 7 and 8 to "on" (up position).
  4. Set octave switch on left hand control panel to center position
  5. If you are not using an oscilloscope:
    1. You have to use your ears to adjust for a "hollow" sound.
    2. Select preset AC6 (Pulse Comp) - this preset uses square waves at a timbre you can use with your ears.
    3. You'll need to manually disable VCO1 or VCO2 as you adjust trimpots.  Diagram #2 in the service manual shows how to disable VCO1 using a jumper to a diode.  To the left is another diode, if you connect the jumper from pin 16 of U16 to this diode then it disables VCO2.  This jumper sets the pulse width to zero, which turns off the VCO.  I recommend calibrating one VCO on all voices, then move the jumper and calibrate the other VCO.
    4. The voicecard LEDs will indicate which voice is triggered.  Refer to diagram #1 in the service manual for the location of the appropriate trimpot on the voicecard.
  6. If you are using an oscilloscope:
    1. Select preset C4 (Rezz Sweep).  This sets the PW CVs to zero volts which is a 50% pulse wave (square waveshape)
    2. Turn UNISON on to trigger all voices at once.
    3. Set the horizontal timebase on the oscilloscope to 0.5ms.
    4. Configure the scope input channel to AC coupled.
    5. Set the scope probe on pin 1 of A3 (4053) on the voicecard to be calibrated.  This is VCO1 pulse wave.  Adjust horizontal trigger and vertical gain on the oscilloscope to get a good steady trace.
    6. Find a key on the OB-SX keyboard such that the complete cycle of the pulse waveform traverses approximately 95% of the horizontal trace.  This is a lot quicker than adjusting the timebase.
    7. On the oscilloscope, adjust horizontal position on the timebase to position the rising (or falling, depending on configuration of the scope horizontal system) transitions of the waveform so that they intercept the graticule at the same number of X axis minor marks from left/right extremes.  In other words you are centering the complete cycle in the scope display.
    8. Refer to diagram #1 in the service manual for the location of the appropriate trimpot on the voicecard.  Adjust the PW trimpot on the voicecard so that the transition of the waveform crosses at the center graticule.  This is 50% pulse width or square waveform.
    9. Repeat steps 5-8 for all voices.
Calibrating VCO pitch (init frequency) and scale (volts/octave or VPO):
  1. Power off (important for the DIP switch settings!)
  2. Set both DIP switches 7 and 8 off (down position) to disable autotune (sets the autotune CV to midpoint), disable all frequency and pulse width modulation, disable unison/hard sync, and select pulse waveshape.
  3. Power on (makes the DIP switch settings active, not mentioned in the manual!)
  4. With a DMM, probe the voltage at connector E pin 16 (white wire) and adjust the rear panel master tune to zero volts (VERY important step omitted from the manual)
  5. Diagram #2 in the service manual shows how to disable VCO1 using a jumper to a diode.  To the left is another diode, if you connect the jumper from pin 16 of U16 to this diode then it disables VCO2.  This jumper sets the pulse width to zero, which turns off the VCO.
  6. Select preset ABC4 (Harmonica).  This is a dual VCO preset with filter settings useful for calibrating the VCOs.
  7. Use a digital reference for tuning by ear (such as a rompler or sampler), or a GOOD digital (not guitar!) tuner.
  8. Referring to step #5, configure the jumper to disable VCO2
  9. Set the DIP miniswitch to voice (n) only
  10. Set the OCTAVE switch on the LH control panel to DOWN (another important step omitted from the manual)
  11. Perform the initial frequency and volt/octave adjustment per service manual for voice #n
  12. Set the OCTAVE switch on the LH control panel to center position
  13. Perform the hi-track adjustment per service manual for voice #n
  14. Important: after adjusting hi-track, re-check initial frequency and volt/octave and repeat steps 10-13 if necessary.  This step is not mentioned in the manual and these trimpots are interactive.  Proper adjustment of hi-track and volts/octave is CRUCIAL for reliable autotune operation.
  15. Proceed to next voice and repeat steps 9-14 until VCO1 for all voices are calibrated.
  16. Referring to step #5, configure the jumper to disable VCO1.  You are now calibrating VCO2.
  17. Adjust the front panel DETUNE pot such that the LED turns OFF (put the pot at midpoint).  This nulls the detune CV on VCO2.
  18. Set the DIP miniswitch to voice (n) only
  19. Set the OCTAVE switch on the LH control panel to DOWN (another important step omitted from the manual)
  20. Perform the initial frequency and volt/octave adjustment per service manual for voice #n
  21. Set the OCTAVE switch on the LH control panel to center position
  22. Perform the hi-track adjustment per service manual for voice #n
  23. Important: after adjusting hi-track, re-check initial frequency and volt/octave and repeat steps 19-22 if necessary.  This step is not mentioned in the manual and these trimpots are interactive.  Proper adjustment of hi-track and volts/octave is CRUCIAL for reliable autotune operation.
  24. Proceed to next voice and repeat steps 18-24 until VCO2 for all voices are calibrated.
  25. Set all DIP switches up.
  26. Cycle power.
Calibrating VCF range (filter init) and scale (volts/octave or VPO):
  1. Tools: your ears.
  2. Must complete VCO pulse width and pitch/scale before calibrating VCF.
  3. Because the OB-SX VCF will not self-oscillate we have to use our ears to calibrate the VCF.  Calibrating the VCF by ear requires a preset with keyboard tracking enabled and moderate resonance.  The best preset for the job is preset AB5.  This is a double reed preset using a static timbre (no detuned VCOs).  Since I have good aural memory and I know what a double reed is supposed to sound like (I used to play in high school and community orchestras) I used this as a reference timbre and all the other presets happily fell in line.  Try to find a sample of a double reed to get an idea what to listen for, especially across different keys.
  4. This preset uses pulse waveforms.  If the pulse width calibration is not consistent between voicecards, then this VCF calibration won't work.
  5. Set the OCTAVE switch on the left hand control panel to the center position.
  6. Set both DIP miniswitch switches 7 and 8 to "on" (up position).
  7. Set the DIP miniswitch to enable all voices.
  8. DO NOT ADJUST the trimpot below the filter VPO trimpot (next to the 741 opamp IC).  This adjusts the resonance tracking response for the CEM3320 so that resonance is consistent between voices.  If you have an early grey panel OB-SX this trimpot may not be present.
  9. We need to find our "reference voice".  Play each voice across the keyboard and listen to the timbre.  Double Reed is a timbre that is not too bright at top C4 key, and not too dull and muffled at C2 key.  Cycle through each voice one at a time to identify the one to use as a reference.  This is why you need a sample of the double reed - your ears are referenced, now you need to find a voice that matches that reference.  If you find a voice that is closest then adjust the VCF trimpots to zero the timbre if necessary (see steps 12-15).
  10. When a reference voice is identified, a good "sanity check" is other presets.  The strings are good checks (all "3" programs) especially AC3 "Fiddle" - they should not be too bright or too dull.  Other good sanity checks are ABC6 (Clarinet), A6 (Flutes), BC2 (Bells should sound like glockenspeils), ABC7 (congas keys Eb1 & Eb2), ABC5 (Rotary Organ, Jimmy Smith drawbars), and C5 (Combo Organ).  Having familiarity with traditional instruments (aka "aural memory") that the OB-SX imitates helps to confirm that you nailed the reference voice.
  11. Using the voice disable switches on the DIP miniswitch, enable only the reference voice and a voice to be calibrated.  As you play the keyboard you should be hearing only two voices cycle through, and you may hear timbre differences between them.  You are using you ears to adjust the voice to be calibrated to the reference voice.
  12. Refer to diagram #1 in the service manual for the location of the appropriate trimpot on the voicecard.
  13. Adjust the Filter Init trimpot using the C2 key
  14. Adjust the Filter VPO trimpot using the C4 key.
  15. Repeat steps 13 and 14 until the VCF of both voices track together.  Keyboard tracking only works on about a three octave range on the VCF, that is why the C2 and C4 keys are used.
  16. If you're having trouble bringing the VCF in line, try reversing the trimpot adjustments relative to the keys.
  17. Repeat for next uncalibrated voice.
The OB-SX is now calibrated to the point where autotune should work, so go ahead and push the button.  You should see each voicecard LED trigger in quick succession, if one or more are triggered longer then the autotune system is working on a problem voice.  Once I finished voicecard calibration, autotune worked right off the bat.  If it cannot autotune any voice(s), the PROGRAM button LEDs will flash failed voices (program # is the voice #) and those voice(s) will be disabled.  Usually another autotune will work.  Autotune can only correct for range, but not for scale so it is important that the scale adjustment on the VCOs are good.  Autotune only has a small range it can correct for so if the VCO range trimpot is too far out then it cannot autotune that VCO.  If none of the voices will autotune, the audio summer opamp on the autotune buss on the voicecard motherboard has been known to go bad.  Otherwise the 2nd version of the OB-X service manual includes a technical description of the autotune circuit for troubleshooting.

Now that I have a fully functioning OB-SX, it was time to build the modifications.  The onboard LFO is not always available for vibrato, some patches use it for PWM or filter FM.  I decided to add an independent LFO for vibrato only.  The circuit was copied right out of the OB-Xa.  I put a rate knob with LED on the left hand controller panel, similar to the OB-Xa adaptation.  I happened to have some surplus pots with pull switches so I used the pull switch to change from triangle waveshape to square for trills.  Pretty compact solution and a very musical addition.  It's nice to add vibrato to a string patch using PWM.

The other modification was voice gate LEDs.  I like to know the status of voices I am playing (or how quickly autotune is working) so I used the 4099 CMOS gate driver to drive a set of simple NPN current amplifiers for LED drivers.  This will also be handy when I get my MIDI retrofit installed.  I found clear high brightness LEDs at my hobbyist supply store and bought different colors for each voice.  These buggers are bright.  As it turns out, those gate LEDs were handy for figuring out how the CHORD feature worked, because it isn't the same as the OB-X!  I put the LEDs on the left hand control panel right above the mod/pitch levers, where they are always visible.  It's kind of fun to watch them dance different colors while playing but they can be distracting at times!

Modified Left Hand Control Panel
Voice Gate LED Driver
Vibrato LFO circuit
Wiring under LH Panel

How Does It Sound?

It's definitely got "the" Oberheim sound.  I put it between the OB-X and OB-Xa.  I say that because while it has the CEM chipset of the OB-Xa, the voicecard audio summer is that of the OB-X which is dirtier than the CEM3360 used in the OB-Xa.  Not as organic as the discrete voice of the OB-X, but not as clean as the OB-Xa.

This isn't a synth I would recommend for modern stuff like techno or dub.  Most of the presets are dated.  Half of them are imitative bread-n-butter sounds (strings, brass, organs, EPs).  Some of them are downright useless (where the h3ll am I gonna use "Calculator" or "Water Wiggle"?!?).  The tweaking from the front panel controls are limited - there's no facility to adjust resonance or filter EG amount so you have to find a preset that is closest to the sound you want.  Frankly there aren't many choices for resonant filters.  There's no front panel control for VCO waveshape, frequency, or level; filter release; LFO waveshape or destination; or the sustain stage of the EGs.  This was supposed to be a "budget" Oberheim.  You don't buy one of these for full tweaking.

Most polysynths immediately "jump" to the pot position when you tweak the front panel pot.  The old Oberheims do not do this, when you move the pot you are adjusting the current value in the CPU memory.  If you want to hear the full range of the pot, you have to sweep it all the way counterclockwise then all the way clockwise then play with the pot.  This confuses many first time Oberheim users.  Oberheim used this approach because they used a pot scanning scheme that eliminated an expensive ADC chip.

If you want the Oberheim string, brass, or pad sounds the OB-SX can definitely deliver.  It's a good axe for classic rock or prog rock and a useful sound source for Tangerine Dream style sequence timbres (especially with CHORD mode and MIDI).  With some traditional orchestral percussion sounds for presets (IE Bells, Celeste, Marimba) I can see why a high school bought one of these.  With controls provided for detune, filter cutoff, attack, and VCA release you can dial up some nice fat synth sounds.  Five of the presets are "starting points" for generic synth or pad sounds - Mellow Wow, Rezz Sweep, SX-6, SX-7, and SX-8.  "OB-SX Choir" is rather good, as choir sounds are a good "acid test" for the resonant quality of a filter.  You can get some pretty meaty basses out of the OB-SX too.  And the OB-SX has one of the meanest hard sync sweeps with pitch bend in BROAD range routed to VCO2 ONLY.

Playing around with HOLD and CHORD can yield some nice effects.  Any key(s) pressed when you press the HOLD button are drone notes, and you can play over this drone.  If you want to use CHORD there's a "secret handshake" that isn't described in the owner's manual: DON'T play low C - it is added automatically.  I would not have known this had it not been for the voice gate LEDs that I added!  If you play low C as the root note, the OB-SX will use TWO voices for the root note.  The OB-SX chord mode works differently than the OB-X.  If you configure an (n) note chord on the OB-X, every key you play will have (n) note chord.  On the OB-SX, only the first key will have the (n) note chord while additional keys are single note only so you have to use discretion on legato playing.  If you use CHORD with some of the percussive presets while tweaking filter cutoff and/or filter EG decay you can get some cool timbres that lend themselves to sequences over MIDI.  CHORD combined with polyphonic portamento (only possible on the pinstripe model with 56 presets) can make for instant Styx or prog rock solos.

UNISON is always a big, big feature on any VCO polysynth and they don't get much better than the old Oberheims.  It's even better with portamento, and as you read in the "Specifications" section below Oberheim has emulated a feature unique to the OB-X...

I do notice that the CEM3340 VCOs in the OB-SX are not as stable tuning-wise as the discrete VCOs in the OB-X.  So far the OB-X is the only analog polysynth in my arsenal that is IN TUNE when I turn it on and DOESN'T NEED THE USUAL 10-20 MINUTE WARMUP PERIOD.  Despite the new trimpots, caps, and repaired solder joints in the OB-SX, I press that AUTO button about 2-4 times an hour.

While any old Oberheim synth rarely needs effects, certain effects do bring the OB-SX to life especially the bread-n-butter sounds.  Good digital reverbs like the Lexicon Model 200 can make the seemingly lame "Congas" preset (a timbre you would least expect from an analog synth) sound very realistic.  My favorite delay processor for my analog synths, the Korg SDD-3300, has some custom patches that I dialed up that can create a wide stereo image to the OB-SX.  Some of the audio samples (to be added to this webpage later) will show the sound dry then with effects to demonstrate the difference.

One advantage the OB-SX has over the OB-X is the VCO cross modulation is more consistent between voices on the OB-SX - you can play scales with sideband frequencies pretty much tracking between voices.


The OB-SX was introduced in July 1980 as an affordable compact version to the big brother OB-X.  There is much debate as to whether the OB-SX is an OB-X or OB-Xa.  People say the latter because the voicecards used the same CEM chipset of the OB-Xa.  The early owner's manual claims it uses the same circuitry of the OB-X, and the later one says OB-Xa.  Here's the truth: the OB-SX uses the same voicecard architecture and control system architecture as the OB-X while the voicecards use the CEM chipset of the OB-Xa to duplicate the OB-X voicecard.  The OB-SX was a pilot model for the new ICs emerging from Curtis Electronic Music - CEM3320 and CEM3340.  The OB-SX was the first Oberheim to use the full 3310/3320/3340 chipset and its design led to the OB-Xa, released about sixteen months after the OB-SX was introduced.  The introduction of the OB-Xa and its pinstripe panel theme coincided with the transition from grey panel OB-SX to pinstripe panel, which explains the references to OB-X/OB-Xa in the owners manuals.

Indeed, if you study the schematics of the OB-X and OB-SX the latter is derived right from the OB-X with some circuits omitted.  The OB-SX has a partial control set for modifying the presets and omits user patch storage, S&H waveshape of the LFO, noise source (although the audio routing is still there!), and uses a 49 note keybed instead of 61 note.

At list price US$2995 for the four voice and US$3495 for the six voice - in 1980 dollars - that ain't exactly "affordable"!  Three versions existed: early models had a gray non-pinstripe front panel with 24 presets (most being copies of the OBX factory patches), with optional upgrade to 48 presets and adding polyphonic glide. Units with 48 presets can be identified by the slideswitch on the rear panel for selecting between the two sets of 24 presets (why didn't they put that on the left hand panel?!?).  These units had analog portamento but only for monophonic (unison), not polyphonic.  Early voicecards did not have a trimpot for adjusting resonance control.  The CEM3320s didn't have consistent control behavior of resonance between lots as they arrived at the Oberheim factory, which is a problem with polyphonic systems.  The trimpot was added to match resonance control between voicecards.

The third (and last) versions with the pinstripe front panel was expanded to 56 presets, added polyphonic (digital) portamento, and eliminated the rear panel slideswitch by implementing combinations of bank select pushbuttons.  The monophonic analog glide circuit was omitted (now redundant).  Also the voicecards for the pinstripe model were changed to a THIRD version - the power connector was expanded from four pin to five pin, thus older voicecards are not compatible with the pinstripe model.

Voicecards are NOT interchangeable between the OB models.  They are not the same architecture and you will never get filter tracking and resonance to match between mixed voicecards, it isn't possible.  The connectors do not line up and the pin assignments on those connectors are different.

Analog snobs should not knock the OB-SX "digital" portamento until they've heard it.  A little known feature is that Oberheim emulated the portamento behavior of the OB-X - unequal glide rates between voices due to unmatched OTAs in the OB-X!  It is especially prevalent as portamento rates get longer, the voices are detuned in between notes and can sound really fat!  Oberheim knew that a circuit "fault" could be used as a "feature"!

See this table for the OB-SX preset names. 

Back then owners had the option of having their custom OB-X patches transferred to custom preset ROMs for the OB-SX.  The OB-SX and OB-X shared the same DAC and voicecard architecture so it is a bit for bit copy, but that wasn't possible with the OB-Xa or OB-8 because of differences in the architecture.  In this product announcement, the unit pictured is a prototype as the front panel pot arrangement is different.

There are seven control knobs on the front panel for varying glide, LFO rate, VCO2 detune, filter cutoff, VCF/VCA EG attack time, VCF EG decay time VCA EG release time but it is not programmable and custom sounds cannot be stored.  That's it for patch tweaking!  Pushbutton switches are provided for autotune, hold, chord, unison (fat!) and preset selection.  There is a monophonic output and volume is global to the instrument, there is no programmable volume per preset.

They were available in four, five, six voice configurations.  Usually the serial badge on the rear will specify how many voices are present (unless a four voice unit has been expanded to full six voices in the field).  The voicecard architecture is identical to the OB-X, although the discrete VCOs and VCFs have been replaced with CEM ICs which predated the OB-Xa.  Each voice featured two VCOs based on the CEM3340 VCO IC, each capable of providing ramp or variable pulse width waveforms.  VCO2 can be hard synced to VCO1 but there is no capacity to modulate VCO2 other than the pitch bend and mod wheel on the left hand controller panel.  There is an option for oscillator cross modulation (XMOD) using VCO2 ramp to VCO1 frequency, but the depth is fixed and VCO2 cannot be put in LFO mode.  Crude level controls of each oscillator is limited to full/off for VCO1, and full/half/off for VCO2.  You cannot combine waveforms - ramp or pulse, not both. 

There is no noise source on the OB-SX BUT there is a test point on the controller board marked "N" that can be used for external audio input, as the noise audio routing is still present on the voicecards!  It appears this was a last minute omission as Oberheim was trimming back options to meet the price target.

Modulation is pretty basic on the OB-SX.  A global LFO provides triangle or square waveshape which can be routed to VCO1 and/or VCO2 frequency, VCF cutoff, VCO1 and/or VCO2 pulse width.  Unfortunately the only provision provided to the front panel is LFO rate. There is no S&H waveshape like the big brother Oberheims.  There is no voice modulation like that offered on the Prophet-5 or Memorymoog.  There is no provision to route filter EG to VCO frequency (later added to OB-Xa) or pulse width (later added to OB-8), nor can you impart audio FM on the filter cutoff with a VCO (later added to OB-8).

The VCF is comprised of the CEM3320 VCF IC which is configured as a 12dB lowpass circuit to emulate the OB-X/SEM 12dB VCF.  While it is the same circuit on the OB-Xa, the OB-SX does not include the 24dB VCF on the OB-Xa.  No big loss as the 24dB filter on the OB-Xa wasn't that great anyway.  The only modulation option is from the LFO or the EG.  There is a rear pedal jack for filter control which affects all voicecards simultaneously.

The VCA is the same discrete circuit found on the OB-X and is partly responsible for the resemblance of sound to the OB-X, despite the VCOs and VCFs resembling the OB-Xa.  The VCA is comprised of a 3080 OTA followed by a TL081 to buffer the weak current output of the OTA.  The OB-Xa used a dual VCA CEM3360 for the VCA and for voicecard balance which was required for the dual multitimbrality of the OB-Xa.  The 3080 OTA is not exactly a high fidelity component and some distortion is introduced here, which lends the OB-SX its dirty edgy sound.  Definitely not clean but there is a pleasant mild overdrive here that is pleasing to the ears.  It's interesting how the OB-SX voicecard is a cross between the OB-X and OB-Xa (see the comparision table).

I have read accounts of later OB-SX models using a CEM3360 in the voice audio summing circuit on the voicecard motherboard.  They were definitely used on the OB-Xa because the CEM3360 is a dual VCA component and that was required for the dual timbrality of the OB-Xa.  But on the OB-SX the second VCA of the CEM3360 would have been redundant so I am not convinced that this rumor has any truth to it.  The OB-SX voice audio summing circuit again features that lovely 3080/TL081 combination that adds some dirt to the sound especially on unison patches.

You can introduce vibrato with the mod wheel on the left hand controller panel or with a control pedal via the rear panel jack.  A switch routes pitch bend to both VCOs or VCO2 only (great for hard sync).  Another switch selects the range of pitch bend to either two semitones or a full octave (the latter is great combined with VCO2 only and hard sync).  Frankly the pitch bend is very abrupt at full octave and the taper doesn't allow fine control of pitch bend (OK I know they're pinball flippers not wheels, but I'm retaining the convention used in the rest of the world, especially MIDI).  A third switch changes the range of the keyboard up or down one octave.  There is a rear panel jack for filter cutoff; this is designed for a passive attenuator with tip to the wiper of a 50K pot and sleeve to the pot lead at "OFF" position.

The OB-SX includes the pre-MIDI Oberheim system interface which will work with the DSX sequencer.

CV/Trigger inputs only work on voicecard one and the jacks are wired such that when used the computer will skip voicecard one.  This lets you use a sequencer on one voicecard with the remaining voicecards are controlled by the keyboard, but because the OB-SX does not feature multitimbrality you can't get an individual patch on that sequenced voicecard.  There are CV/Trigger outputs provided but they are the signals from voicecard one only - if you want to use the OB-SX keyboard to control an external monophonic voice, you have to have UNISON turned on.  Unfortunately portamento, pitch bend, and LFO do not work on the CV output.

The OB-SX was discontinued in 1982.

Other than the front panel pots, there is no access to the other voice parameters such as filter resonance or waveform select.  You cannot access them from the Oberheim system interface either.  From a study of the computer architecture, there does not appear to be enough RAM capacity to support full patch manipulation.  Even if one were to rewrite the OS to implement remote patch editing over MIDI, it can't be done without expanding the RAM.  The only way to get full patch manipulation is to replace the processor board with an upgraded design, which is a daunting task in that it also includes the LFO, the CV summers, and all the DEMUX +S/H for the voicecard CVs.

OB-SX Preset Names

These are the preset names in the OB-SX.  Certain banks are only present in the specified model # of presets.

PGM #1
PGM #2 PGM #3 PGM #4 PGM #5 PGM #6 PGM #7 PGM #8
Brass Ensemble
Low Strings
Electric Piano
Percussive Organ
Lead Sync
Classic Horns
High Strings
Brass in Fifths
Pipe Organ
OB-SX Choir
Pop Organ
Slow Strings
Rezz Sweep
Combo Organ
AB (48/56 only)
Mellow Wow
Reed Piano
Double Reed
PW Rezz
AC (48/56 only) Ensemble Trumpets
Polyphonic Portamento
Filter Drone
Pulse Comp
Steel Drums
Water Wiggle
BC (48/56 only) Square Wave Mod
Strings II
Unison Portamento
Cross Mod Delay
Solo Unison
ABC (56 only) Bass
Cross Mod Bells
Solo Strings
Rotary Organ
Rush Rezz

I put some videos demonstrating these presets on Youtube.  If the link no longer works, just search for "oberheim obsx" under my name.

OB Model Comparision

There's a lot of confusion between the different Oberheim OB models so I built this table to compare them

4, 6, 8 (even only)
2x discrete
2x CEM3340
2x CEM3340
2x CEM3340
expanded in software matrix
VCF (lowpass only)
Discrete cascaded OTAs, 12dB
CEM3320, 12dB
2x CEM3320, 12dB or 24dB
1x CEM3320, 12dB or 24dB
3080 OTA + TL081 buffer
3080 OTA + TL081 buffer 2x 3080 OTA + TL081 buffer CEM3360
Voice volume control
2x CEM3310 VCF & VCA
2x CEM3310 VCF & VCA 2x CEM3310 VCF & VCA 2x CEM3310 VCF & VCA
two (for dual timbrality split/layer)
two (software)
Noise Source

# of patches
24/48/56 depending on version
32/120 depending on version
User Storage
Audio Field
Voicecard Audio Summing
3080 OTA + TL081 buffer 3080 OTA + TL081 buffer CEM3360
DAC Resolution
MIDI Retrofit
Encore Electronics
Highly Liquid UMR2
Encore Electronics Encore Electronics

contact info