ComputerVoltageSources

I'm glad that I've brought a little life to this list, it shows that
there are people like me that wanted a PSIM and will probably never
get one, at least not from Brice.

If I may be so bold as to suggest that we have a reference design in
the form of the PSIM itself. 4 analog inputs, 4 analog outputs, 6
LEDs, Run/Stop/Reset inputs/switches, and RS-232. With the addition
of I2C planned expansion to the board itself, we have a usable, and
VERY expandable module. Were there any other I/O 'paths' that were
included with the MPU in this design (I don't remember all of the
discussions we've had over the last year)? The greatest expansion for
this module would be off-board via the I/O paths allowing for an
unlimited number of options and modules to be developed.

I personally want to build the module for use with 1/4" jacks in
either DotCom or MOTM format. I would eventually end up building all
of Dave's add-ons, but with the I2C, I could build as many modules as
I wanted, and locate then anywhere it was convenient within my synth
layout.

The one advantage of Brice's PSIM was that it did get built and some
people got one. Therein lies the biggest drawback to this group, the
'design by committee' has never ended and while there have been a lot
of really good ideas, the complexity of the reference design has
stopped the module from actually becoming a reality.

Dave, could you design a PSIM-like base unit, usable for some CV I/O
but certainly not what add-ons could make? Then as add-ons are
developed, we can pick and choose? If there was a tested PCBExpress
(or other board maker design) I'd buy three (or whatever the minimum
was) and either keep them or sell the extras.

Please, we are actually so close at this point...

Larry T.

PSIM as a reference platform didn't go very far when it was last proposed on
3/7/2006. Check the archives.

Larry, you should understand that you're asking Dave to contribute dozens to
hundreds of hours to design a new PCB, when one has already been provided
and tested. No one says that you can't depopulate the current design to get
a PSIM like subset if that's what you desire.

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: ComputerVoltageSources@yahoogroups.com
> [mailto:ComputerVoltageSources@yahoogroups.com] On Behalf Of Larry T.
> Sent: Tuesday, April 24, 2007 11:38 AM
> To: ComputerVoltageSources@yahoogroups.com
> Subject: [ComputerVoltageSources] PSIM replacement
>
> I'm glad that I've brought a little life to this list, it shows that
> there are people like me that wanted a PSIM and will probably never
> get one, at least not from Brice.
>
> If I may be so bold as to suggest that we have a reference design in
> the form of the PSIM itself. 4 analog inputs, 4 analog outputs, 6
> LEDs, Run/Stop/Reset inputs/switches, and RS-232. With the addition
> of I2C planned expansion to the board itself, we have a usable, and
> VERY expandable module. Were there any other I/O 'paths' that were
> included with the MPU in this design (I don't remember all of the
> discussions we've had over the last year)? The greatest expansion for
> this module would be off-board via the I/O paths allowing for an
> unlimited number of options and modules to be developed.
>
> I personally want to build the module for use with 1/4" jacks in
> either DotCom or MOTM format. I would eventually end up building all
> of Dave's add-ons, but with the I2C, I could build as many modules as
> I wanted, and locate then anywhere it was convenient within my synth
> layout.
>
> The one advantage of Brice's PSIM was that it did get built and some
> people got one. Therein lies the biggest drawback to this group, the
> 'design by committee' has never ended and while there have been a lot
> of really good ideas, the complexity of the reference design has
> stopped the module from actually becoming a reality.
>
> Dave, could you design a PSIM-like base unit, usable for some CV I/O
> but certainly not what add-ons could make? Then as add-ons are
> developed, we can pick and choose? If there was a tested PCBExpress
> (or other board maker design) I'd buy three (or whatever the minimum
> was) and either keep them or sell the extras.
>
> Please, we are actually so close at this point...
>
> Larry T.
>

John, I realize that the design process is a non-trivial one. I wish
I had the skill set to do it myself, as I would be more then happy to!
Dave indicated that his current board design was not really 'ready
for prime time' and so, while he would make it available, it would be
a difficult board to work with. I imagine that with enough support, I
could build one with the limited features selected, but would Dave
have that time to lend any more then to redesign the board to one that
does not require support? And if others built one from the current
board, would they need support time as well?

Don't get me wrong. I know that this is a volunteer effort. I,
personally, would be willing to buy a board for more then just the
cost to make it in order to benefit the designer of the board. I do
not know how many others would, so I can not say if the designer would
ever make a profit over time having boards made and selling them.
Hence I am willing to make more then one PCB just to get one if I can
use it.

Is this something that someone like John Blacet might consider?
Adding the product to his excellent line of modules, and maybe selling
the bare board and pre-programmed MPU?

Would this community be okay with John (or anyone else) using the work
done here as the basis for a commercial product?

Should I just shut up and go back to lurking?

Larry T.

--- In ComputerVoltageSources@yahoogroups.com, "John Loffink"
<jloffink@...> wrote:

>
> PSIM as a reference platform didn't go very far when it was last

proposed on

> 3/7/2006. Check the archives.
>
> Larry, you should understand that you're asking Dave to contribute

dozens to

> hundreds of hours to design a new PCB, when one has already been

provided

> and tested. No one says that you can't depopulate the current

design to get

> a PSIM like subset if that's what you desire.
>
> John Loffink
> The Microtonal Synthesis Web Site
> http://www.microtonal-synthesis.com
> The Wavemakers Synthesizer Web Site
> http://www.wavemakers-synth.com

The issue is ExpressPCB is very expensive for solder mask in small
quantities. The setup fee is $233 so you have to order a lot of
boards to cover this cost. The board needs to be larger so this cost
would increase. Someone with PCB layout software needs to layout the
board in a larger size with standard PCB output files. This larger
size would also allow DIP versions of the DACs to be added so SMT
could be avoided. My layout could be used as a starting base. I know
there are people on this list that have PCB layout skills, software
and experience.

This project will only move forward if others volunteer. I've
contributed all I can.

Dave

--- In ComputerVoltageSources@yahoogroups.com, "Larry T." <larry@...>
wrote:

> Don't get me wrong. I know that this is a volunteer effort.

Dave, and others that may be interested. ExpressPCB offers ProtoPro
service. This is 4 boards for $179.00, which includes solder mask and
silk screening. That works out to about $45 per board. Size is 21
square inches per board max.

Dave, just how hard was your current board to work on? How big is it?
In your opinion, could some that does everything from MOTM to CGS
boards be able to build it in it's current configuration?

Larry T.



--- In ComputerVoltageSources@yahoogroups.com, "djbrow54" <davebr@...>
wrote:

>
> The issue is ExpressPCB is very expensive for solder mask in small
> quantities. The setup fee is $233 so you have to order a lot of
> boards to cover this cost. The board needs to be larger so this cost
> would increase. Someone with PCB layout software needs to layout the
> board in a larger size with standard PCB output files. This larger
> size would also allow DIP versions of the DACs to be added so SMT
> could be avoided. My layout could be used as a starting base. I know
> there are people on this list that have PCB layout skills, software
> and experience.
>
> This project will only move forward if others volunteer. I've
> contributed all I can.
>
> Dave
>
> --- In ComputerVoltageSources@yahoogroups.com, "Larry T." <larry@>
> wrote:
> > Don't get me wrong. I know that this is a volunteer effort.
>

i will commit to buying at least one PSIM replacement PCB set.

i have no layout skills but i have decent thru-hole building skills.
in the unlikely event that the construction is putting off people
from buying the PCBs, i can help a bit there.

(i have a little time for building, but not much justification for
extending my system.)

>The issue is ExpressPCB is very expensive for solder mask in small
>quantities. The setup fee is $233 so you have to order a lot of
>boards to cover this cost. The board needs to be larger so this cost
>would increase. Someone with PCB layout software needs to layout the
>board in a larger size with standard PCB output files. This larger
>size would also allow DIP versions of the DACs to be added so SMT
>could be avoided. My layout could be used as a starting base. I know
>there are people on this list that have PCB layout skills, software
>and experience.
>
>This project will only move forward if others volunteer. I've
>contributed all I can.
>
>Dave

It is 4x6 so it is too big for this service. It needs to be even
larger. I couldn't fit the MIDI components, I have 1/8 watt
resistors UNDER the AtomPro28 part in the interior of the socket, my
I2C protection changes use 1/8 watt resistors, and there is no room
for the 8 input filter capacitors so those are piggybacked on the
feedback resistors. I also believe the SMT DACs should be changed
to SMT/DIP combination and there is no way to do this in an overlay
so that is a couple of additional components. I would guess you'd
want to add 1 - 2" in each direction to spread things out a bit so
I'd start with a board size of 5 x 7 to 6 x 8.

I paid $120 for two quicks with no solder mask. I know the price
for solder mask has the $233 setup charge plus the charge per board.

As to the runs, I'll have to look but I think they are 10 mil runs
with 12 mil clearance and no solder mask. For reference, that is
half the spacing of the SMT pads. I'll try and upload some 1:1
photos of the clearance this weekend.

I'm also not happy with the ground planes. They aren't connected
the way they should be and I can't get the software to do what I
want. The board needs to be redone.

Dave

--- In ComputerVoltageSources@yahoogroups.com, "Larry T."
<larry@...> wrote:

>
> Dave, and others that may be interested. ExpressPCB offers

ProtoPro

> service. This is 4 boards for $179.00, which includes solder mask

and

> silk screening. That works out to about $45 per board. Size is 21
> square inches per board max.
>
> Dave, just how hard was your current board to work on? How big is

it?

> In your opinion, could some that does everything from MOTM to CGS
> boards be able to build it in it's current configuration?
>
> Larry T.
>
>
>
> --- In ComputerVoltageSources@yahoogroups.com, "djbrow54" <davebr@>
> wrote:
> >
> > The issue is ExpressPCB is very expensive for solder mask in

small

> > quantities. The setup fee is $233 so you have to order a lot of
> > boards to cover this cost. The board needs to be larger so this

cost

> > would increase. Someone with PCB layout software needs to

layout the

> > board in a larger size with standard PCB output files. This

larger

> > size would also allow DIP versions of the DACs to be added so

SMT

> > could be avoided. My layout could be used as a starting base.

I know

> > there are people on this list that have PCB layout skills,

software

> > and experience.
> >
> > This project will only move forward if others volunteer. I've
> > contributed all I can.
> >
> > Dave
> >
> > --- In ComputerVoltageSources@yahoogroups.com, "Larry T."

<larry@>

> > wrote:
> > > Don't get me wrong. I know that this is a volunteer effort.
> >
>

Hi Dave
i'm a bit embarrassed to say so, but i was worried about the 4x6
real estate limit..... but not having the cajones to try it myself, i
held my all too eager tongue. All i can do is say thanks for being the
'first pioneer out of the wagon' on this shenanigan.
Clearly (thanks to your efforts, Dave) the integrated 16-ch
approach is not a trivial enterprise and it will require the talents
of folks experienced with high-density digital and analog layouts to
make it robust and allow for assembly by human-sized fingers.
i did a tentative layout based on my default platform, which is
plate-thru-hole perf protoboard (i got a pile of surplus ones with PCI
fingers on one side) they're about 4x8.... i gave up when i was up to
3 boards!
For my part i'm willing to accept that this may be too expensive
and finicky to do -right- on an amateur basis.

My perf design inclines me to believe that a modular approach has
merit. (that German fellow took the modular/backplane approach on a
different PIC platform) but that paradigm has it's own set of problems:
-coordinating consensus for a spec, again
-a robust reliable interconnection scheme

Perhaps we should just try for a good frame-out of the BasicAtom28
on a 'motherboard' and leave the ADC/DAC daughterboards to the
resources and imaginations of the 'private sector'

No matter how we slice it though, much gratitude is due to you for
determining how deep this quicksand is!

patiently circling (like any good buzzard)
;'>
-doc


--- In ComputerVoltageSources@yahoogroups.com, "djbrow54" <davebr@...>
wrote:

>
> It is 4x6 so it is too big for this service. It needs to be even
> larger. I couldn't fit the MIDI components, I have 1/8 watt
> resistors UNDER the AtomPro28 part in the interior of the socket, my
> I2C protection changes use 1/8 watt resistors, and there is no room
> for the 8 input filter capacitors so those are piggybacked on the
> feedback resistors. I also believe the SMT DACs should be changed
> to SMT/DIP combination and there is no way to do this in an overlay
> so that is a couple of additional components. I would guess you'd
> want to add 1 - 2" in each direction to spread things out a bit so
> I'd start with a board size of 5 x 7 to 6 x 8.
>
> I paid $120 for two quicks with no solder mask. I know the price
> for solder mask has the $233 setup charge plus the charge per board.
>
> As to the runs, I'll have to look but I think they are 10 mil runs
> with 12 mil clearance and no solder mask. For reference, that is
> half the spacing of the SMT pads. I'll try and upload some 1:1
> photos of the clearance this weekend.
>
> I'm also not happy with the ground planes. They aren't connected
> the way they should be and I can't get the software to do what I
> want. The board needs to be redone.
>
> Dave
>
> --- In ComputerVoltageSources@yahoogroups.com, "Larry T."
> <larry@> wrote:
> >
> > Dave, and others that may be interested. ExpressPCB offers
> ProtoPro
> > service. This is 4 boards for $179.00, which includes solder mask
> and
> > silk screening. That works out to about $45 per board. Size is 21
> > square inches per board max.
> >
> > Dave, just how hard was your current board to work on? How big is
> it?
> > In your opinion, could some that does everything from MOTM to CGS
> > boards be able to build it in it's current configuration?
> >
> > Larry T.
> >
> >
> >
> > --- In ComputerVoltageSources@yahoogroups.com, "djbrow54" <davebr@>
> > wrote:
> > >
> > > The issue is ExpressPCB is very expensive for solder mask in
> small
> > > quantities. The setup fee is $233 so you have to order a lot of
> > > boards to cover this cost. The board needs to be larger so this
> cost
> > > would increase. Someone with PCB layout software needs to
> layout the
> > > board in a larger size with standard PCB output files. This
> larger
> > > size would also allow DIP versions of the DACs to be added so
> SMT
> > > could be avoided. My layout could be used as a starting base.
> I know
> > > there are people on this list that have PCB layout skills,
> software
> > > and experience.
> > >
> > > This project will only move forward if others volunteer. I've
> > > contributed all I can.
> > >
> > > Dave
> > >
> > > --- In ComputerVoltageSources@yahoogroups.com, "Larry T."
> <larry@>
> > > wrote:
> > > > Don't get me wrong. I know that this is a volunteer effort.
> > >
> >
>

> I'm also not happy with the ground planes. They aren't connected
> the way they should be and I can't get the software to do what I
> want. The board needs to be redone.
>
> Dave

What sort of problems do the way the ground planes were connected have
on the software working properly?

I routed the board first without ground planes, and then added them.
The ground planes will not merge with the buses, and can only connect
through pads. So I have a ground bus from edge to edge of the board
with a space to a ground plane that is right next to it. The
connection is through short horizontal buses on the top layer that
connect to the ground plane with pads. I would have wanted the ground
plane to flow right into the bus on the back. The only way I could
have done this was to eliminate the bus, move the ground plane over,
then connect each pad to the ground plane. I didn't want to rip that
much up after I successfully routed it. It might not be that big of a
deal but I didn't want to do it that way. I also couldn't figure out
how to have more clearance around pads with the ground plane than
standard pad to run spacing. Maybe it is obvious, but it wasn't at
the time. Also, I was quite sick of hand routing the board by that
time and just wanted a prototype.

Dave

--- In ComputerVoltageSources@yahoogroups.com, "wiardmodular"
<wiardmodular@...> wrote:

> > I'm also not happy with the ground planes. They aren't connected
> > the way they should be and I can't get the software to do what I
> > want. The board needs to be redone.
> >
> > Dave
>
> What sort of problems do the way the ground planes were connected
> have on the software working properly?