MOTM

I like the first layout better.  This one:

http://www.eskimo.com/~strohs/FSlayout.GIF

The two new ones look nice, but a little off visually.  Placing the 
LEDs right on the edge looks kind of awkward to me.  Also, I like the 
jacks centered, as in the first layout.  I have an idea, why not 
offer the FS as a partial kit?  Assembled and tested board, since 
it's all SMT, but minus the pots, jacks, and LEDs. I would assume 
those are not SMT for mechanical reasons, and are through hole board 
mount (I could be wrong here).  Then it would be relatively easy for 
a DIYer to solder in wires to mount the pots, jacks, and LEDs in 
whatever layout and format they choose, whether it be a Stooge panel, 
Modcan, Blacet, .com, or whatever.  You would then open the door to 
many other customers who may have been considering this product, but 
shy away because it's only available in MOTM format.  You could then 
offer assembled units in whatever layout you want, and not have to 
worry about someone hacking it apart to change the faceplate.  No 
need to get lawyers involved, just give the "kit" a disclaimer that 
you're not responsible for the DIYer mis-wiring pots, LEDs and such.

FWIW, I am actually more concerned with how the FS sounds, that's my 
primary concern, but I am really, really picky when it comes to UI as 
well.  A fantastic sounding through-zero FS is such a bummer if the 
UI is clunky.  Just thought I'd throw that idea out there and see if 
anything sticks.

KB

--- In motm@yahoogroups.com, "neutrino000" <neutrino000@h...> wrote:
> I like the first layout better [...]
> 
> The two new ones look nice, but a little off visually.  Placing the 
> LEDs right on the edge looks kind of awkward to me.  Also, I like
> the jacks centered, as in the first layout [...]
>
> FWIW, I am actually more concerned with how the FS sounds, that's
> my primary concern, but I am really, really picky when it comes to
> UI as well.  A fantastic sounding through-zero FS is such a bummer
> if the UI is clunky.

If you want "none clunky" UI, you really want my revised (v3) layout.  
It is (visually) almost as good as the first one but I put much more 
thought into what knobs should go where - as in, how the module is 
actually used and making it consistent with the way other modules 
work.

Not to suggest that this is possible (requires a little more support 
circuitry) but you could fix the jack field symmetry like this:

http://www.eskimo.com/~strohs/FSlayout3a.GIF

This would actually be a pretty cool feature because you now have all 
four phases of a quadrature oscillator for other uses.


seth

This layout keeps getting better all the time. If the inverted SINE 
and COS outputs aren't integral to the FS then some Stooge (or 
Stooge-wannabe) will just hack a daughterboard for them anyway. So 
might as well add them to the FS and fill up that jack space.

At 9:44 PM +0000 2003/11/05, strohs56k wrote:

>
>http://www.eskimo.com/~strohs/FSlayout3a.GIF
>
>This would actually be a pretty cool feature because you now have all
>four phases of a quadrature oscillator for other uses.
>

My 2c:

  I like this.  Of course, you're adding cost to the product with more 
hardware.  The large initial frequency control ideally should be a 10-turn 
type, but I don't think they make those for PCB-mount.

  From a PCB design pov, I don't like the position of the Freq CV pot.  It 
upsets the already premium board real estate.  If this board is like the 
UEG's then the overall size is about 3"x6".  While Tony has the advantage 
of placing components on both sides, this narrow form factor really puts 
the squeeze on the balance between control ergonomics and PCB parts 
arrangement.  The best thing to do with this pot is line it up with the 
gain and fine shift controls.  This at least gives an unrestricted 
rectangle for parts in the center of the board.

Crow
/**/

On Wed, 5 Nov 2003, strohs56k wrote:

> circuitry) but you could fix the jack field symmetry like this:
> 
> http://www.eskimo.com/~strohs/FSlayout3a.GIF

You have touched on a key issue and that is parts placement. Just because
the knobs are in nice symmetrical locations doesn't mean I can get the
circuitry where I need it. Currently there are 8 trim pots on the board and
I require them to be placed on the edge of the PCB so the thing can be
calibrated while it's assembled. The control pots are now right where I had
the trim pots, so needless to say it's becoming a PITA to get these placed.
Still working on it with some alternate pot placement...

> -----Original Message-----
> From: The Old Crow [mailto:oldcrow@...] 
> Sent: Wednesday, November 05, 2003 4:02 PM
> To: motm@yahoogroups.com
> Subject: Re: [motm] Re: Frequency Shifter (improved layout idea)
> 
> 
>   My 2c:
> 
>   I like this.  Of course, you're adding cost to the product 
> with more hardware.  The large initial frequency control 
> ideally should be a 10-turn type, but I don't think they make 
> those for PCB-mount.
> 
>   From a PCB design pov, I don't like the position of the 
> Freq CV pot.  It upsets the already premium board real 
> estate.  If this board is like the UEG's then the overall 
> size is about 3"x6".  While Tony has the advantage of placing 
> components on both sides, this narrow form factor really puts 
> the squeeze on the balance between control ergonomics and PCB 
> parts arrangement.  The best thing to do with this pot is 
> line it up with the gain and fine shift controls.  This at 
> least gives an unrestricted rectangle for parts in the center 
> of the board.
> 
> Crow
> /**/
> 
> On Wed, 5 Nov 2003, strohs56k wrote:
> 
> > circuitry) but you could fix the jack field symmetry like this:
> > 
> > http://www.eskimo.com/~strohs/FSlayout3a.GIF
> 
> 
> ------------------------ Yahoo! Groups Sponsor 
> ---------------------~--> Rent DVDs Online - Over 14,500 titles.
> No Late Fees & Free Shipping.
> Try Netflix for FREE!
> http://us.click.yahoo.com/xlw.sC/XP.FAA/3jkFAA/VpLolB/TM
> --------------------------------------------------------------
> -------~->
> 
>  
> 
> Your use of Yahoo! Groups is subject to 
> http://docs.yahoo.com/info/terms/ 
> 
> 
> 
>

--- In motm@yahoogroups.com, Scott Juskiw <scott@t...> wrote:
>
> This layout keeps getting better all the time. If the inverted SINE 
> and COS outputs aren't integral to the FS then some Stooge (or 
> Stooge-wannabe) will just hack a daughterboard for them anyway. So 
> might as well add them to the FS and fill up that jack space.


There is one other possibility for those two jacks - but I'm not sure 
if it would actually be useful.

Instead of inverted outputs for the local oscillators, maybe a more 
useful feature would be inputs into the up and down feedback paths.  
That way you could do things like:

Up output goes to a VCA input, VCA output goes to the Up feedback 
input.  Now you have VC control of that feedback path.

Or maybe, Down output goes to a VCF input, VCF output goes to the Down 
feedback input.  No idea what this would do :)

Or maybe, Down output goes to Time Machine input, Time Machine output 
goes to Up feedback input.  Again, no idea what this would do :)

Someone who know the workings of frequency shifters would need to 
speak up as to weather putting things into the feedback path could 
result in cool sounds.


I can see two approaches to feedback inputs: use a switched jack with 
the input pre-attenuator.  Plugging into the jack disconnects the 
internal feedback path and you get a knob to scale your external 
input.  The other possibility is to have the feedback into a summing 
node post attenuator.  In this way, you get to keep the internal 
feedback path (controlled with the knob) but also get to mix in an 
external source as well.


Again - no idea if this is more or less useful than having inverted 
oscillator outputs.  Just another idea.


seth

Since opinions are desired and the layout can change, I like:

http://www.eskimo.com/~strohs/FSlayout.GIF
or
http://www.eskimo.com/~strohs/FSlayout3a.GIF

These are much improved over the initial layout.

	Eric

Yes, either of those will do, with a slight preference for 3a.  Previously
posted version 3 was good as well.

(Making sure my vote counts.) :-)

John Loffink
The Microtonal Synthesis Web Site
http://www.microtonal-synthesis.com
The Wavemakers Synthesizer Web Site
http://www.wavemakers-synth.com

> -----Original Message-----
> From: alt-mode [mailto:yahoo@...]
> 
> Since opinions are desired and the layout can change, I like:
> 
> http://www.eskimo.com/~strohs/FSlayout.GIF
> or
> http://www.eskimo.com/~strohs/FSlayout3a.GIF
> 
> These are much improved over the initial layout.
> 
> 	Eric
>

--- In motm@yahoogroups.com, The Old Crow <oldcrow@o...> wrote:
> 
> I like this.  Of course, you're adding cost to the product with
> more hardware.  The large initial frequency control ideally should
> be a 10-turn type, but I don't think they make those for PCB-mount.

I seem to remember this being discussed - one camp liked the idea of 
multi-turn pots because it gives you more resolution, the other camp 
liked single turn because it allowed relatively easy manual sweep over 
a wide range - and felt the combination of coarse and fine knobs will 
solve the problem of resolution.

Because the local oscillator is microprocessor generated, I think the 
way to do a multi-turn is to use a high resolution optical encoder.  
(128 pulses per revolution or maybe more.)  This way, you aren't 
limited to the resolution of the AD converter that is sampling the 
pots.  And you can probably find a board mounted optical encoder - in 
fact, I know you can :)

The disadvantage of an encoder is there are no end stops but I think 
you could come up with a workaround.  Say, a couple of LEDs that 
indicate when you are at either extreme of the frequency range.

seth

>The disadvantage of an encoder is there are no end stops but I think
>you could come up with a workaround.  Say, a couple of LEDs that
>indicate when you are at either extreme of the frequency range.

Also, IMHO, you need an additional display to indicate the current 
value. Otherwise you'd never be able to document a patch.