Interface Works!

[PLI]

Based on the feedback I received from Ove and Patrick, I rebuilt my
Pushpin interface and it works.

Patrick, it was the datalink cable that was damaged, I had another one
on had, replaced it and it worked. It is really gong to be hard to
get any real work done today now that I got this thing going.

Here's the photo - http://www.billbyrne.net/pushpinworks.jpg.

Thanks guys,
wb

[morbank]

Nice work wb! And thanks for the picture. Is there any chance you
could post a parts list?
-Clay

[Ove Ridé]

One question though. Does it work or does it kind of work?
Is it totally flawless or does it skip or insert a note or two from
time to time?

--
/Ove

-----BEGIN 2ROT13 MESSAGE-----
Low Bitrate Netlabel: <http://flb.gg8.se/>
Blog:
<http://gameboygenius.8bitcollective.com/>
SKRIVA på Lysator. Ditt Ahrvid-fria alternativ:
<http://lists.lysator.liu.se/mailman/listinfo/korkek?SKRIVA?>
Sätt på ett par flipflops, vippa på rumpan
och gör det här till en minnesvärd sommar!
-----END 2ROT13 MESSAGE-----

[Patrick Leonard]

Same comment as Ove,

1. Do both channels 1 and 2 sound the same on startup?
2. When starting and stopping sequences do you sometimes have notes still playing?
3. Do all channels seem to play well together with a full sequence? Similar to # 2.

[Patrick Leonard]

On Jan 15, 2008 8:14 AM, morbank <mor...@cmsquared.net> wrote:

Nice work wb! And thanks for the picture. Is there any chance you
could post a parts list?
-Clay

Follow Ove's guide, it's the same list and read the threads some parts have changed.

[Ove Ridé]

Also, an addition to the set of questions. If you have a MIDI keyboard
(Not a MIDI out from the computer, but a standalone keyboard so to
speak) could you try connectin the interface to that instead? I've
noticed that my keyboard also periodically sends a stop signal (Or
possibly active sensing?) which might put some more stress to the
interface. Also, try using the pitch wheel to see if that can make PP
crash or glitch.

[morbank]

Ove, you mentioned not using a 220k resistor for R3. What should I be
using in it's place?
Also, where should the inverter go?
-Clay

On Jan 15, 11:36 am, "Ove Ridé" <nitro2...@gmail.com> wrote:
> On 15/01/2008, Patrick Leonard <dreamt...@gmail.com> wrote:> Same comment as Ove,

[Patrick Leonard]

Use a 220, not a 220k. 

The sd connects to the inverter on a pin in.

a couple threads back has a schematic.

On Jan 15, 2008 12:03 PM, morbank < mor...@cmsquared.net> wrote:

Ove, you mentioned not using a 220k resistor for R3. What should I be
using in it's place?
Also, where should the inverter go?
-Clay

[morbank]

Thanks Patrick. Sorry for all the ignorance but will the inverter need
to be connected to power and gnd?
-Clay

On Jan 15, 12:33 pm, "Patrick Leonard" <dreamt...@gmail.com> wrote:
> Use a 220, not a 220k.
>
> The sd connects to the inverter on a pin in.
>
> a couple threads back has a schematic.
>

[Ove Ridé]

Grrr! If only Eagle could stop being buggy I could've uploaded a
schematic by now.

On 15/01/2008, morbank <mor...@cmsquared.net> wrote:
>
> Thanks Patrick. Sorry for all the ignorance but will the inverter need
> to be connected to power and gnd?
> -Clay
>

--

[morbank]

Nevermind Patrick, I found the Schematic you were talking about.
-Clay

[PLI]

Ove, it's is working well for the most part there are some strange
things though. To answer your question about the skipping and
inserting notes, that only happens when I go down an octave or two on
midi channels 1 or 2. Everywhere else though it's all good. Using
the pitch wheel doesn't crash PP.

To answer Patrick's questions-

> 1. Do both channels 1 and 2 sound the same on startup?

Are you talking about MIDI channels 1 and 2? Because no, they do not
sound the same. They are similar, but not the same.

> 2. When starting and stopping sequences do you sometimes have notes still
> playing?

Yes, this does happen especially if I have been playing one note for
long time it will kind of linger until I hit another key.

> 3. Do all channels seem to play well together with a full sequence? Similar
> to # 2.

Have not been able to test this yet, I'll get back to you on this.


Would any of these things be solved by adding the inverter?

wb

[PLI]

> 3. Do all channels seem to play well together with a full sequence? Similar
> to # 2.

Tested it this morning and this is also working without any major
perceivable problems.

wb

[Patrick Leonard]

Ove, i know the inverter is good to have but could this be causing the
note hang problems

[Ove Ridé]

On 16/01/2008, Patrick Leonard <drea...@gmail.com> wrote:
>
> Ove, i know the inverter is good to have but could this be causing the
> note hang problems

What I think may be causing the note ang problems is errors in
transmission. I don't think the inverter in itself will cause these
problems, but I think the inverter will need different resistor
values. (Which I wrote in italic on the page. Maybe I should add a
longer explanation)
This is confirmed by the slightly different values on the original
interface. (Which I have seen a picture of) The original interface is
basically the same as my schematic, except for having a MIDI through
and more importantly having different resistor values.
Compared to this:
<http://gameboygenius.8bitcollective.com/images/pushpin/pushpin-midi-gameboy-2.png>
Noah's interface is using a 1 kOhm resistor for R2 and a 200 Ohm
potentiometer for R1, if I'm reading the pictures correctly. And
probably the potentiometer setting might need to be adjusted for
different individual 74hc04 chips.

[Patrick Leonard]

I have been thinking about the resistor that controls the rise time of the signal...If it were dropped done a bit could this allow the signal to catch those 'note offs'

[Ove Ridé]

On 17/01/2008, Patrick Leonard <drea...@gmail.com> wrote:
> I have been thinking about the resistor that controls the rise time of the
> signal...If it were dropped done a bit could this allow the signal to catch
> those 'note offs'

Good thought, but that's not the way MIDI works. MIDI works by sending
a signal in the shape of 1's and 0's that are being sent by switching
the voltage between low and high. (In other words, a simple serial
line protocol) That resistor controls the time it takes for the
transistor in the opto-coupler to restore to a 0 state (Zero voltage)
when it was in a 1 state. (Positive voltage)
It's not a resistor for controlling decay time or something like that.
If it has the wrong value, it will however glitch the signal, in a
more complex ways.
The problem might very well be in that resistor, but it's still not as
simple as than "decreasing the decay time".

[Patrick Leonard]

Shot in the dark,  but could the pushpin program itself only be set to receive 'note off' and not sequence 'stop' messages
i know i keep digging into this, but i would like to start using pushpin live and not have to worry about hanging notes.

Ove- In the original schem by Noah, i see 4 resistors, one is hidding between the IC's.  Did you catch this?

Does anyone have a working build of the source?  I have the source but the build doesn't work.

[Ove Ridé]

On 18/01/2008, Patrick Leonard <drea...@gmail.com> wrote:
> Shot in the dark, but could the pushpin program itself only be set to
> receive 'note off' and not sequence 'stop' messages
> i know i keep digging into this, but i would like to start using pushpin
> live and not have to worry about hanging notes.

I think you mis understood a thing I said before. These stop messages
(Or possibly active sensing) is something that my keyboard spits out
at a stable rate of maybe 2-3 messages every second. I don't know why
it does this, but it does. Iow, I didn't mean that you get stops
instead of note offs.
If PP misses the note offs, it's because of errors in transmission,
whatever the cause of these errors is.

> Ove- In the original schem by Noah, i see 4 resistors, one is hidding
> between the IC's. Did you catch this?

Yeah, of course. Now that you mention it though, I misread it as a 1k
resistor I think, when it was really a 1.5k. But it doesn't matter
much since I didn't make a schematic based on the original circuit
anyway. (I just mentioned it in some mail I think)

> Does anyone have a working build of the source? I have the source but the
> build doesn't work.

I should fix this. I've promised to do it before, but I forgot. Sorry.