[sdiy] Dual power supply solution? - Kits vs 2600

[Jonathan Lutz]

Out of curiousity, have you looked and found any such issues in the new EFM 
3500 series?

JON


On 6/14/05, Harry Bissell Jr <harrybissell at prodigy.net> wrote:
> 
> --- Paul D Brooks <pdb at sbcglobal.net> wrote: <snipp'd>
> 
> > Sure, but Harry, all three of the Arp 2600's i've
> > had do this... (well, with the VCF freq slider, not
> the Q), and i'm not sure we can call the designers of
> that machine unenlightend.
> 
> No but if you noticed the 'problem'... they were too
> cheap to do it better. That is a trade-off that
> manufacturer and kit designer alike must make...
> 
> Yeah, they could have done
> > better with respect to this (and i wish they
> > woulda!)... but what are you suggesting in
> > particular?
> 
> In particular...
> 
> 1) The reference for the expo converter sould come
> from a regulated (reference) source... NOT from the
> positive or negative rails. Reference might be LM336
> or LM4040 etc...
> 
> 2) The reference for the sawtooth reset level should
> come from a reference source... NOT from a resistive
> divider on the rails. If it 'is' from the rails it
> could be heavily bypassed, but I usually see NO
> capacitor at that point. Even bypassing will not
> protect from very low frequencies...
> 
> 3) Front panel pots that control coarse and fine
> frequency should come from reference sources, not the
> rails. These might be able to survive with cheaper
> zener diodes if you are willing to re-tune every so
> often. If they are from the rails they need to have
> resistors and capacitors to filter the voltage...and I
> might add it is DAMN hard to filter .01Hz (from an
> LFO) with an RC filter. Shunt regulators would be
> better.
> 
> Most designers do not want to deal with the added cost
> and complexity, so they make a circuit that works fine
> when it is the ONLY cirucit on the power supply. Put
> these in a modular and you will quickly have troubles.
> 
> Most complete instruments can forego some of these
> components IF the designer is fully aware of the
> issues involved. If you can minimize the changes in
> supply current, a lot of interactions will 'go away'
> 
> SOME folk might like the interactions, it might give
> character, warmth, or some special quality to the
> sound... This is usually unintentional and is like
> discovering that the " sow's ear " really does make a
> pretty good purse.
> 
> > This is a hard problem to tackle for all, and it's
> > kind of harsh to come down on the kit manufacturers
> > in
> > particular.
> 
> I'm not 'coming down' on them. They have made
> (hopefully) informed and enlightened engineering
> decisions. In particular I'll mention EFM. These are
> often missing some of the subtle points I covered
> above... but OTOH they are targeted at a price point
> that would not allow including the extra components.
> The whole design point of EFM is to keep everything
> dead simple. Also note that a lot of people might be
> confused with all the extra components and the
> additional 'room for errors'. In some cases I've
> detailed mods that I did to 'improve' the EFM units to
> meet my expectations... which may differ from other
> peoples. Still... for like $10 for a PCB Tom is almost
> running a charity and deserves credit for what he's
> doing !!! :^P
> 
> H^) harry
> 
> 
> 
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20050614/4b587710/attachment.htm>