• Power supply
  • Caps:C[1-10], C15
• U900 bloc
• U800 bloc
  • ICs: U814, U815, U805, U813, U811, U801, U802, U804
• U100 bloc (rare?)
  • ICs: U10[2-7] (panning), U108 (tuning comparator), U101 (output op amp)

• U800 bloc
  • Caps: C802, C807
  • ICs: U814, U815, U805, U813, U801, U802, U804, ?(U816 RES only)
• U900 bloc

• U[2-7]00 bloc (for that individual voice. Voice 1 = U700 bloc, Voice 2 = U600 bloc, etc.)
  
  • Caps: VCA in/out per voice: Cx49 (common!), Cx47
  • More Caps: (power coupling per voice): Cx32, Cx36, Cx38, Cx39, Cx44 (note: ‘Cx39’ is mislabeled in the Xpander service manual as ‘Cx35’, middle of page 55.)
  • ICs: Ux06, Ux07, Ux08, Ux01, Ux02

• VCO fail
  • Caps: Cx35, Cx37, Cx33, Cx31
  • ICs: Ux05, Ux11, Ux10, Ux09, Ux08
• PW fail
  • ICs: Ux11
• VCF or RES fail
  • Caps: Cx48,?
  • ICs: Ux12, Ux08, Ux13, Ux15, Ux16, Ux17
• AMP fail
  • Caps: Cx49, Cx47
  • ICs: Ux15, Ux16, Ux17, Ux12

• Voice board: U100 bloc
• Pot board: U1

Hi there,
Just to let you know that I replaced? cd4053 (ic 904 if I remember well) and the tune is now working...
Thanks for your help there are loads of good advice to read here...

What is in file section is laters version.
Only this post is older version.

---

I uploaded a newer version with all precautions in the files section.

Re: Xpander Auto-Tune failures

I have an Oberheim Xpander (v 14. ROMS), which is failing all auto-tune tests.
I have read every post related to tuning back to 2008 and have developed a game plan.
I am posting here, please feel free to provide feedback.
I am an electronic technician and I have access to the parts listed as well as a full set of test equipment.
(Digital voltmeters, Oscilloscopes, I.C. Testers, etc)

Please feel free to add comments or update me on what needs corrections.

This is my game plan in repair of my Xpander.
In my case I will test any I.C. listed before replacing.

Oberheim Xpander Cheat Sheet

?

Check output of each CEM 3374 VCO chip, pins 2,5,14,17 with oscilloscope.

Any chip this missing any of the 4 outputs needs to be replaced.

Reseat all IC’s as they are prone to “Chip Creep” over time. Retest unit before proceeding.

?

Check underside of voice board for bad solder joints, repair as necessary.

?

Replace C233, C333, C433, C533, C633, C733 &

??? C237, C337, C437, C537, C637, C737 &

??? C249, C349, C449, C549, C649, C747, C749 they are all 2.2uF 50v

??? C103, C104 15uF 35v

?

Check operation of:

U811 DAC

U816, U805 Multiplexer CD4051BE

U812, U815, U813 Op Amps TL081, TL084, TL081

U814 Analog Switch CD4053BE

?

The CD4051BE & CD4053BE are known to die over time.

?

U706, U707 Voice 1

U606, U607 Voice 2

U506, U507 Voice 3

U406, U407 Voice 4

U306, U307 Voice 5

U206, U207 Voice 6

U805, These are all CD4051BE

8-Channel Analog Multiplexer/Demultiplexer,

These are known to fail over time.

U805 is responsible for collating the temperature compensation voltage from each CEM.

?

U814 can also fail, CD4053BE, U814 is responsible for combining a stream of high and low resolution CVs for delivery back to the analog world. ?Kinda crazy, but this whole rig is running constantly. e.g. The tempco CV is updating constantly; You can hold a cold finger to one of the voice’s CEMs and see the DAC output change in real time. ? Any parameter changes that affect CV are also updated through here in real time

?

U108 (LM311N) is a comparator that essentially takes the audio from the voices, and turns them into square waves for delivery back to the U9xx block for tuning. ?It's always running, outputting a square wave version of anything the voices put out. ?Check output with scope.

?

Once U108 has turned everything into pulses/square waves, there isn’t a need for ADC, so it’s signal is fed directly to U915, U916 and U921. ?By my estimation, U915 is used exclusively for tuning PW1&2. (On a side note, the square waves from the Xpander’s VCO are their own little bag of tricks. It’s not generated by the CEM VCO ICs directly, but derived in realtime from a combo of the SAW waves and CV for pulse width, spliced together per-voice by a CMOS/OP-amp combo. ?Lots of analog synths do something similar.) ?U916 works in tandem with U921 to address memory, possibly responsible for storing the newly derived tuning offsets.

I had a failing voice, tunepage showing "fail" already for OSCs, w/ my
Xpander last year.
I?m 1st owner since 1987...
Can be I posted before @ (Yahoo) Xpanatastic,- but the culprit was a
tiny TL0XX (typing from my head here) Op-Amp in the voice?s circuitry.
Since then,- all good again.
I was very surprised because I was about recapping the machine which
would have been MUCH more expensive since I hire a service tech for
inhouse work.

:-)

PeWe

Am 22.06.2020 um 22:53 schrieb r1n9o:

[Edited Message Follows]

That's great to hear!

--
Diese E-Mail wurde von Avast Antivirus-Software auf Viren geprüft.

That's great to hear! ?(I'll wager U814 was the problem, since the failures were in specific CV categories across all voices.)

Regarding the new problem, perhaps the new IC's are simply revealing a different preexisting condition. ?Since the fails are voice specific, it's unlikely to be anything in the U8xx or U9xx block, but something in the per-voice section of the board. ?This is when Karl usually points to possible electrolytic capacitor failure around the voices (and the record shows he's often correct!) ?Probably worth poking around the older messages on the forum for other ideas too. ?(re-seating/cleaning ICs, replacing caps, swapping ICs across voices to narrow down the culprit... etc.)

On firmware: In the "Files" section of this forum, there's an ancient text doc of Oberheim's official release notes for the firmware releases.?It's a good read, with the hardware criteria for updating, and all of the otherwise undocumented features in the latest versions. ?Keep in mind, there are two firmwares to keep track of: the main processor, and the voice processor. ?The final versions were 1.2 (main) and 1.4 (voice). ?Since you're still running 1.0 ("Main", I presume), ?you have much to gain by upgrading both. There's usually someone on eBay or Reverb selling them...

The doc is here:?/g/xpantastic/files/Documents/Xpander_OS%20Update.zip

(It's a zipped archive of text and Word doc versions of the same info, very 90's! ?Maybe a moderator could replace it with just the unzipped text version...)

Well, that was a lot of information.
And good information, too!

I've been quiet so far because I've waited for replacement parts which now have arrived.

So I first swapped several of the 900 group of IC's, namely U901, 902, 906, 915, 916, 921 (the first three because they differed from the schematic and the last three because they are part of the measurement circuit).

Didn't make the slightest difference: OSC and PW still showed the same problems as before, RES and FILT were fine.

And so I decided to swap U805 and U814 as well. And guess what: OSC and PW are now tuning fine every single time I perform the tuning routine! I'll probably try to zero in later if either one was faulty.

But - weirdly enough - not all is well. After swapping the chips the filter became a bit flaky! Most of the time it's tuning fine, but occasionally one or two voices (3 and 4) fail repeated tuning attempts until they suddenly start working correctly again. I haven't discovered a pattern in this so far. What I find odd is that this never happened before.

Anyway, I now have a mostly working Xpander. A big "thank you!" to the group and especially r1n9o and Karl.

Another question: Is there a consensus about updating the firmware?

Yes, I did all memory checks and also verified 4v reference and DAC Offset.

Maybe too obvious, but have you tried any of the calibration tests in the service manual? ?(Relevant might be the "MEM", "VMEM", 4v reference, and DAC calibration tests.)

In recently researching IC replacements for the Xpander, I've noticed that TI's own datasheets sometimes offer seemingly conflicting info about the LS/HC/HCT series of chips. ?But, it turns out, their specs make them often interchangeable, with the possible exception of the input levels. ?HCT is compatible with TTL levels, which are a little higher. ?However, the modern versions (i.e. 1990's) of all these chips, can handle a variety of input voltages, including TTL levels, depending on how much power is supplied. ?IOW, they're much more adaptable. ?From the Xpander's point of view, on a per chip basis, and depending on the power level supplied, some in the LS/HC/HCT _might_ be interchangeable. (FWIW, in my Xpander, U901 and U902 are both?74LS367, but one's by Motorola and the other's Hitachi. I believe they're both dated 1984.)

In the Xpander, the LS/HC/HCT family of ICs are responsible for passing, combining or switching CV, audio or data; And sometimes, at least two of these are happening on the same chip! ?I _think_ U901 and U902 specifically, are part of a chain that are assembling voice data from (and/or to?) RAM. ?I assume if they were incompatible or failing, the voice buffer would be dramatically corrupted or not even functional, and unrelated to tuning issues. ?I think you'd get a scrambled, or non-functional patch, not just an untuned one. ? Again, I _think_.

Of course, they're so cheap, it wouldn't hurt swap them out with the originally specified series, just to eliminate the possibility that they're contributing to a problem. ?But it's hard to say if U901 & U902 would contribute to a tuning failure.

开云体育

I had thought about U921 too; however according to the service manual it's also used for VCF calibration - and that works every time.
I'd be interested to learn more about the tuning process. For example: The OSC signal from U108 isn't going just to U921; it's also connected to U916 and U915.

I'll probably still swap the 8253 just to be sure.

On Mon, 8 Jun 2020 19:55:14 -0500
"Karl Schmeer" <shire03@...> wrote:

Hi
Off the top of my head, Maybe U921 or it’s connected glue logic. This timer chip is used for both vco tuning and PW
Calibration.
- Karl

--
Malte Rogacki <gacki@...>

开云体育

Addendum: I'm also a bit irritated by the different logic families present. Most chips are 74LS; and from the service manual I know that 74S and 74C also were used. However I've noticed a number of 74HC chips. Shouldn't 74HCT be used when substituting 74LS? For example: the schematic gives 74LS367 for U901 and U902 but I have 74HC367 installed.


--
Malte Rogacki <gacki@...>

Hello;

I've subscribed to this wonderful group because I recently got an Xpander that exhibits some weird tuning problems.

It's a "Made in U.S.A." model running firmware 1.0.

When I start it I can sometimes directly afterwards tune all 6 VCO's without a problem once (!); however any subsequent tuning attempts result in FAIL for all 6 voices. After it warms up for some time the VCO's are starting to come around until all 6 VCO's tune again. It's not reliable, though - subsequent tunings might again disable (and later re-enable) voices.

PW practically never tunes ok; occasionally I can get one or two voices to not FAIL.

RES and VCF always (!) tune fine.

When I enable the voices they often sound fine although with a slightly different timbre (I believe that's the missing correction offsets for the PW). Sometimes one or two oscillators are out of tune in a certain range - as if the offset for that range is missing.

I've recapped the power supply (there was some ripple) and also the voiceboard. The machine originally had one dead 3372 and 3374 each; I've replaced them (though the 3372 is a "D" revision). Voltages are ok.

I've also found at least two dead 4051; and to be on the safe side I've swapped Ux06 and Ux07 for all voices.

Right now I'm poking a bit in the dark. The behaviour seems to hint at a thermal problem; however I don't quite understand why the VCO's tune eventually but the PW mostly doesn't.
All voices definitely "make it" to the output of U108 and look pretty normal there during the tuning routine.

A working hypothesis of mine at the moment is that the zero crossing detection doesn't work as it should. All measurements for VCO and PW are done with rectangular waves while the RES/VCF measurements are done with sine waves. Perhaps some slew rate problem?

Sadly that part of the voice board circuit isn't (in my opinion) really well documented (page 51 of the service manual). Right now I'm thinking of swapping most of the U900 group chips.

I'd be very happy about any further ideas, suggestions and pointers.

--
Malte Rogacki <gacki@...>

Yes! Sage advice!