[sdiy] Matching CA3080s (was Re: OTA performance (was SSM chip reissue))

[Tom Wiltshire]

On 26 Apr 2017, at 10:44, Tom Wiltshire <tom at electricdruid.net> wrote:

> 
> On 26 Apr 2017, at 02:55, Michael E Caloroso <mec.forumreader at gmail.com> wrote:
> 
>>> Incidentally, Oberheim managed to make an SVF with CA3080s in the OB-X, and
>>> they're not matched at all (unless they matched them?).
>> 
>> The OTAs in the SVF used in the Oberheim polySEMs and OB-X are indeed matched.
>> 
>> You're confusing the OTAs in the polyphonic glide circuits in early
>> OB-X versions, which were not matched.  Those unmatched glide OTAs
>> resulted in very fat multi voice unison sounds because the glide rates
>> wasn't the same between voices.  It was considered a desirable enough
>> "defect" that Oberheim implemented unmatched glide rates in the
>> digital glide of the OB-SX and subsequent models.
>> 
>> Oberheim never published the matching procedure, but a circuit for
>> testing OTAs for matching can be found in the Formant construction
>> guide which is online.  I figured out the matching process and showed
>> it to Dave Rossum formerly of Emu, whose name appears on the patent
>> for the Oberheim VCF.  Rossum confirmed the process.
>> 
>> MC
> 
> This is very interesting. I'm not confusing it with anything - I didn't know about the polyphonic glide - it just so happened I'd been looking at the schematics for the OB-X, and there's nothing on the schematic to tell you if they're matched or not. The only thing the service manual really says is "the filter will only track over three octaves", but it doesn't mention gain-matching between the two OTAs.

Ok, I've found the reference to CA3080 matching in the Elektor guide.

>From the Elector Formant Construction Guide, pg55:

"The CA3080 is available in two versions, the standard version in which the ratio between the maximum and minimum Gm is 2:1, and the CA3080A in which the spread in Gm is only 1.6:1. A test circuit and test procedure for selecting ICs with similar Gm are given at the end of the chapter and it is certainly worthwhile buying a few extra OTAs and selecting the four with the most similar Gm. The 'reject' devices are perfectly acceptable for use in the 12dB VCF or VCA and need not be wasted."

So at least as far as Elektor were concerned, matching OTAs was only worthwhile for the 24dB filter. The 12dB SVF would apparently work with any old thing, since they often must have been building them with the OTAs with the most *dissimilar* Gm's.

Do we have any hard evidence that Oberheim matched OTAs for use in their SVF? Or maybe they did for the early SEMs but not later for the OB-X?

Just curious, don't care much one way or the other...

Tom