Searching \ for '[EE] My very simple optical guitar tuner board - p' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: techref.massmind.org/techref/index.htm?key=very+simple+optical
Search entire site for: 'My very simple optical guitar tuner board - p'.

Exact match. Not showing close matches.
PICList Thread
'[EE] My very simple optical guitar tuner board - p'
2011\06\09@191845 by V G

picon face
Hey all,

Here's my very simple guitar tuner board.

Schematic: http://postimage.org/image/yqrtiyh0/
Unrouted PCB: http://postimage.org/image/yqth2ays/
Routed PCB: http://postimage.org/image/yqq5zlz8/

I'm planning to get it made with DorkbotPDX. $5/sq in and you get 3 copies
of the board.

The board is powered from 3V (2 x 1.5V alkalines), has 2 push buttons for
picking string/other function, has a slide switch for power, crystal with
capacitors, ICSP header, 6 surface mount LEDs for staus/string choice, two
high brightness LEDs for oscillating.

Just realized - those high brightness LEDs take 30mA, but the PIC can't
provide 30mA on each pin. I'll probably stick a couple of small BJTs on
there. Maybe two MMBT2222 type

2011\06\09@193041 by Josh Koffman

face picon face
On Thu, Jun 9, 2011 at 7:18 PM, V G <spam_OUTx.solarwind.xTakeThisOuTspamgmail.com> wrote:
> Hey all,
>
> Here's my very simple guitar tuner board.
>
> Schematic: http://postimage.org/image/yqrtiyh0/
> Unrouted PCB: http://postimage.org/image/yqth2ays/
> Routed PCB: http://postimage.org/image/yqq5zlz8/

I just had a quick look, but here are a couple of thoughts. First off,
I'd swap the position of the crystal and its caps. I've always been
taught that the shorter you keep the oscillator traces, the better.
Plus it might give you more "finger room" around those switches,
depending on the height of the crystal.

Second, the way you've chosen to connect and then place the LEDs and
their resistors has resulted in a lot of extra business on that side
of the board. I know it probably doesn't make much difference as you
use the autorouter, but if you either connected the resistor to the
port pin, then to the LED _OR_ swapped the physical location of each
resistor and LED, you might end up with a cleaner PCB.

After I route a PCB (which is almost always manually) I then go back
and see how much I grow the trace sizes. For the speeds I'm running
things at, there's no need for .01" traces when there's enough room
for 0.024. This usually involves moving vias, etc.

Josh
-- A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
        -Douglas Adams

2011\06\09@193618 by V G

picon face
On Thu, Jun 9, 2011 at 7:30 PM, Josh Koffman <.....joshybearKILLspamspam@spam@gmail.com> wrote:

{Quote hidden}

Thanks! I'll swap the crystal and caps and try clean up the LEDs

2011\06\09@211816 by Carey Fisher

face picon face
Couple more tips:
1.  Try to keep silkscreen markings off of solder pads (R1, R2, etc).  May
cause issues with soldering and (obviously) you can't read them.
2.  Use high intensity LEDs.  You can get LEDs that only need a few mA but
are plenty bright.

Carey Fisher
Chief Technical Officer
New Communications Solutions, LLC
678-999-3956
careyfisherspamKILLspamncsradio.com




On Thu, Jun 9, 2011 at 7:18 PM, V G <.....x.solarwind.xKILLspamspam.....gmail.com> wrote:

{Quote hidden}

>

2011\06\09@214019 by Sergey Dryga

flavicon
face
V G <x.solarwind.x <at> gmail.com> writes:

{Quote hidden}

1. Transistors to drive LED: it's easier to tie LED to Vcc and use BJT as low
side switch;
2. Capacitors: recommend a larger caps (in addition to decoupling 0.1uF that you
already have) to handle high LED current and prevent large volt drop on the
battery.  It might be a non-issue with fresh cells, but as they are used up, the
internal resistance will increase.  
Sergey Dryga
http://beaglerobotics.com

2011\06\09@221821 by IVP

face picon face
> those high brightness LEDs take 30mA, but the PIC can't
> provide 30mA on each pin. I'll probably stick a couple of
> small BJTs on there. Maybe two MMBT2222 type.

You can parallel pins. What about the LEDs on PortA ? There's
a maximum drain for the port as a whole. If they were all driven
at the same time it looks to be about 90 - 100mA. Also, I tend
to earth the crystal case because once or twice I'd had radiation
interfering with s/w

Jo

2011\06\09@223704 by IVP

face picon face
As a precaution I'd put resistors in series with the pushbuttons. If
the pins ever happen to go high output, grounding could take them
ou

2011\06\09@235047 by V G

picon face
On Thu, Jun 9, 2011 at 9:17 PM, Carey Fisher <EraseMEcareyfisherspam_OUTspamTakeThisOuTncsradio.com>wrote:

> Couple more tips:
> 1.  Try to keep silkscreen markings off of solder pads (R1, R2, etc).  May
> cause issues with soldering and (obviously) you can't read them.
> 2.  Use high intensity LEDs.  You can get LEDs that only need a few mA but
> are plenty bright.
>

Thanks! The LEDs that I found are pretty good. And I don't care about the
silkscreen at all. I'll probably just delete that layer. I don't need to see
the markings on the PCB, I know where the components go. If I try and space
everything out, it takes up board space.

What's the worst that could happen by submitting the board like that

2011\06\09@235200 by V G

picon face
On Thu, Jun 9, 2011 at 9:39 PM, Sergey Dryga <sergeyspamspam_OUTdryga.us> wrote:

>  1. Transistors to drive LED: it's easier to tie LED to Vcc and use BJT as
> low
> side switch;
>

Yep - that's what I was thinking.


> 2. Capacitors: recommend a larger caps (in addition to decoupling 0.1uF
> that you
> already have) to handle high LED current and prevent large volt drop on the
> battery.  It might be a non-issue with fresh cells, but as they are used
> up, the
> internal resistance will increase.
>

Yes sir. What size capacitor do you recommend

2011\06\09@235302 by V G

picon face
On Thu, Jun 9, 2011 at 10:18 PM, IVP <@spam@joecolquittKILLspamspamclear.net.nz> wrote:

>  You can parallel pins. What about the LEDs on PortA ? There's
> a maximum drain for the port as a whole. If they were all driven
> at the same time it looks to be about 90 - 100mA. Also, I tend
> to earth the crystal case because once or twice I'd had radiation
> interfering with s/w
>

Thanks, I'll try grounding the crystal case. The LEDs on PORTA will not all
be turned on  at once. Even if they are, what's the worst that can happen?
Dimly lit

2011\06\09@235355 by V G

picon face
On Thu, Jun 9, 2011 at 10:36 PM, IVP <KILLspamjoecolquittKILLspamspamclear.net.nz> wrote:

> As a precaution I'd put resistors in series with the pushbuttons. If
> the pins ever happen to go high output, grounding could take them
> out
>

The switch contacts can handle a lot more than 20mA. What could happen

2011\06\10@000524 by Bob Blick

face
flavicon
face


On Thu, 09 Jun 2011 23:50 -0400, "V G"  wrote:

> What's the worst that could happen by submitting the board like that?

If you have silkscreen overlapping pads or holes they (all the "theys" I
have ever used) trim the ink away so it doesn't interfere. But your nice
lettering is harder to read with cuts out of it.

No Biggie. You can leave it the way it is. It's fine.

Best regards,

Bob

-- http://www.fastmail.fm - The professional email service

2011\06\10@003605 by V G

picon face
On Fri, Jun 10, 2011 at 12:05 AM, Bob Blick <RemoveMEbobblickTakeThisOuTspamftml.net> wrote:

>
> If you have silkscreen overlapping pads or holes they (all the "theys" I
> have ever used) trim the ink away so it doesn't interfere. But your nice
> lettering is harder to read with cuts out of it.
>

Yeah, I don't really care about the lettering.


> No Biggie. You can leave it the way it is. It's fine.
>

Thanks! I'll just leave it

2011\06\10@015204 by IVP

face picon face

> The switch contacts can handle a lot more than 20mA. What
> could happen?

I meant the pins could be taken out, not the switches, although
PIC pins do seem to take a strain fairly well and don't just blow
immediately IME

If you're using the 400uA internal pullups (equating to 7k5 @ 3V)
then a few hundred ohms would be low enough to make a '0'. It
would also be halfway to an RC filter, which might be necessary
as the switches age and get noisy, in addition to any filtering done
in s/

2011\06\10@015205 by IVP

face picon face
> The LEDs on PORTA will not all be turned on at once. Even
> if they are, what's the worst that can happen? Dimly lit?

Possibly. I was considering the package/port wattage. For example,
the datasheet for the 2520 I'm using at the moment says 1W total,
250mA max into Vdd pin, 200mA max sunk/sourced by all ports. If
for some reason all the LEDs turned on that would be a significant
percentage of the limit. Not necessarily bad, but noteworth

2011\06\10@020134 by IVP

face picon face

>> internal resistance will increase.
>>
>
> Yes sir. What size capacitor do you recommend?

I've found with IR transmitters that a 470uF / 6V low-leakage
makes a big difference as the batteries run out. As Sergey says,
Vdd will droop and spike without a reservoir cap. You should
be able to work out what capacity is need to smooth Vdd. For
an IR transmitter the demand is fairly-well spaced bursts, giving
the cap time to recover somewhat. You might find that your
circuit draws too much current to be supplied like that from the
cap, but it would still be helpful for bulk smoothing and preventing
glitche

2011\06\10@020903 by V G

picon face
On Fri, Jun 10, 2011 at 2:01 AM, IVP <spamBeGonejoecolquittspamBeGonespamclear.net.nz> wrote:

>  I've found with IR transmitters that a 470uF / 6V low-leakage
> makes a big difference as the batteries run out. As Sergey says,
> Vdd will droop and spike without a reservoir cap. You should
> be able to work out what capacity is need to smooth Vdd. For
> an IR transmitter the demand is fairly-well spaced bursts, giving
> the cap time to recover somewhat. You might find that your
> circuit draws too much current to be supplied like that from the
> cap, but it would still be helpful for bulk smoothing and preventing
> glitches
>
>
What type of capacitor? Tantalum? Or another kind

2011\06\10@024639 by Sergey Dryga

flavicon
face
V G <x.solarwind.x <at> gmail.com> writes:

>
> On Fri, Jun 10, 2011 at 2:01 AM, IVP <joecolquitt <at> clear.net.nz>
> wrote:
>
> >  I've found with IR transmitters that a 470uF / 6V low-leakage
> > makes a big difference as the batteries run out. As Sergey says,
> > Vdd will droop and spike without a reservoir cap. You should
> > be able to work out what capacity is need to smooth Vdd. For
> > an IR transmitter the demand is fairly-well spaced bursts, giving
> > the cap time to recover somewhat. You might find that your
> > circuit draws too much current to be supplied like that from the
> > cap, but it would still be helpful for bulk smoothing and preventing
> > glitches
> >
> >
> What type of capacitor? Tantalum? Or another kind?

It almost does not matter what type.  Preferably low ESR.  Calculate min
capacity using current * pulse time and desired volt drop.  If it's a one-off project, even a supercap can be used, they are relatively cheap these days. Supercap is probably an overkill, 470uF as IVP suggests is a good starting point.  
Sergey
http://beaglerobotics.com

2011\06\10@025531 by V G

picon face
On Fri, Jun 10, 2011 at 2:46 AM, Sergey Dryga <TakeThisOuTsergeyEraseMEspamspam_OUTdryga.us> wrote:

> It almost does not matter what type.  Preferably low ESR.  Calculate min
> capacity using current * pulse time and desired volt drop.  If it's a
> one-off project, even a supercap can be used, they are relatively cheap
> these days. Supercap is probably an overkill, 470uF as IVP suggests is a
> good starting point.
>

Sounds good, I'll probably use a large capacitor

2011\06\10@033232 by IVP

face picon face
> Sounds good, I'll probably use a large capacitor.

I don't know how good a rule-of-thumb this is, but "they"
reckon 1000uF per A, for eg a PSU. However, a PSU has
a fairly stable impedance. A battery's internal impedance
increases as it discharges, IOW the current has to get out
through an ever-increasing resistor. So you'll see terminal
voltage that drops with load

Purpose of the reservoir cap is to hold charge and deliver it
with more of a force than a weak battery could. A low ESR
(low internal resistance) type would be preferable, as noted.
You'll see them specified in SMPS for example for that charge-
delivering reason

A low-leakage type could mean a power switch is not needed,
which is not just a design extravagance. It means that no need to
remember to turn it off because an insignificant amount of power
is used

I've several projects with nW PICs which simply sleep at less
than battery self-discharge. One here is 6 years old and the
capacity of the original 3 AA alkalines is still very good. The
terminal voltage has barely droppe

2011\06\10@092612 by Carey Fisher

face picon face
On Fri, Jun 10, 2011 at 12:05 AM, Bob Blick <RemoveMEbobblickspamTakeThisOuTftml.net> wrote:

>
>
> On Thu, 09 Jun 2011 23:50 -0400, "V G"  wrote:
>
> > What's the worst that could happen by submitting the board like that?
>
> If you have silkscreen overlapping pads or holes they (all the "theys" I
> have ever used) trim the ink away so it doesn't interfere. But your nice
> lettering is harder to read with cuts out of it.
>
> No Biggie. You can leave it the way it is. It's fine.
>
> My experience is the PCB house makes the screen directly from the Gerber
layer file and then just screen whatever is there onto the finished boards.
I've received boards with the lettering on solder pads.

Now, generally, if the pad is not completely covered with silkscreen ink, it
will take solder OK.  But, the ink /can/ contaminate the solder joint so
it's a no-no for aerospace work (I know you're not doing aerospace work).
A pad mostly covered in ink will likely have soldering problems.

The biggest problem, as Bob mentioned, is you can't read the reference
designator.  This isn't a problem in your situation but, if you were
building a quantity of larger, more populated boards, you'd want the
reference designators, especially for troubleshooting.

But, bottom line is, as Bob said, your layout should work OK for your
circumstance.

Care

2011\06\12@183648 by RussellMc

face picon face
> Purpose of the reservoir cap is to hold charge and deliver it
> with more of a force than a weak battery could.

Olin would have fun with that line, even coming from Joe :-).
Technically the statement is arguably correct as less voltage drop =
more voltage available = more force.
Voltage = EMF = electromotive force is about as close to force as we
get in the analogy stakes.
What was meant is understood, but eg "less of a resistance", "less
impediment" [tm] or similar may be more useful. Or not.

> A low ESR
> (low internal resistance) type would be preferable, as noted.

Yes, but probably not as vast a difference here as some places.
The object is, of course, to not drop much voltage across the internal
resistance so as to prevent droop of the voltage during the transmit
pulse.

> You'll see them specified in SMPS for example for that charge-
> delivering reason

Partially for that, Partially for noise. Partially for minimisation of
internal heating. All relevant.

> A low-leakage type could mean a power switch is not needed,
> which is not just a design extravagance. It means that no need to
> remember to turn it off because an insignificant amount of power
> is used

Yesish. Capacitor leakage of power supply decoupling caps is often not
of vast importance when power is either available in bucketloads or
not at all.
Whereas here, as Joe notes -

> I've several projects with nW PICs which simply sleep at less
> than battery self-discharge. One here is 6 years old and the
> capacity of the original 3 AA alkalines is still very good. The
> terminal voltage has barely dropped

1 microamp is under 10 mAh/year.
So at 3 years 1 uA at 30 mAh is well below the typical 2000+ mAh from
an Alkaline AA.
As Joe notes, shelf life determining self discharge is far greater.
Even at 10 uA = 500 mAh in 5 years an AA Alkaline could be expected to
have some useful life left.
Modern Alkalines tend (from memory) to now be marked with 5 years +
"best by" dates. That SHOULD mean they reatin a significant majority
of their capacity after 5 years. Should.


      Russel

2011\06\12@211342 by Bob Ammerman

flavicon
face

----- Original Message ----- From: "RussellMc" <apptechnzEraseMEspam.....gmail.com>
To: "Microcontroller discussion list - Public." <EraseMEpiclistspammit.edu>
Sent: Sunday, June 12, 2011 6:36 PM
Subject: Re: [EE] My very simple optical guitar tuner board - please giveadvice


>> Purpose of the reservoir cap is to hold charge and deliver it
>> with more of a force than a weak battery could.

I am assuming that the tuner runs the LEDs continuously at the tuning frequency while it is in use. If that is the case then there is no sense in using a really large cap. You only need one to hold enough charge for one pulse of the LED. You certainly aren't likely to have one big enough to hold a charge through a whole tuning cycle, which I am guessing will be at least several seconds.

-- Bob Ammerman
RAm Systems

2011\06\12@224858 by Oli Glaser

flavicon
face
On 13/06/2011 02:13, Bob Ammerman wrote:
{Quote hidden}

It is to be noted that with a tuner, even one pulse at 1:10 duty can be in the order of milliseconds at the lowest frequency.
With mine the lowest frequency is around 16Hz (about the lowest audible) so 1/16 = 62.5ms @1:10 duty = 6.25ms pulse length.
This is assuming you keep the pulse length a ratio of the frequency and not some standard length which is independent.

2011\06\13@002400 by RussellMc

face picon face
> >> Purpose of the reservoir cap is to hold charge and deliver it
> >> with more of a force than a weak battery could.

> I am assuming that the tuner runs the LEDs continuously at the tuning
> frequency while it is in use. If that is the case then there is no sense in
> using a really large cap. You only need one to hold enough charge for one
> pulse of the LED. You certainly aren't likely to have one big enough to hold
> a charge through a whole tuning cycle, which I am guessing will be at least
> several seconds.

You need about 1000 uF provides ~~ 1V of voltage "droop" per mA per second.

        Vd ~~= 1000 . I . t / c

        Vd Volt
        t  seconds
        c uF

For pulsing remove the 1000 and measure t in mS

This V drop is in addition to any  IR resistive drop which will be
instantaneuse.

  Wiring resistance.
  Battery internal resistance.
  Capacitor ESR.

Battery R and wiring R from bat to cap mainly affects cap recharge time.
R cap to load affects instantaneous drop.
ESRcap << Rbat for useful results.

2011\06\13@080610 by Bob Ammerman

flavicon
face
{Quote hidden}

It's a guitar tuner. How low can you go on a guitar?!?

-- Bob Ammerman
RAm Systems

2011\06\13@084108 by RussellMc

face picon face
> It's a guitar tuner. How low can you go on a guitar?!?

Some base fellows can go very low :-)

_________

WWIK, but:

Std - E - 83 Hz
Std Bass - 41(.5) Hz.
8 string Bass - ~ 20 Hz
Custom ...

But - is the lowest note a pure sine wave AND if not does it have only
harminics or is a subharmonic mode possible AND if tuning, would you
not want the fundamental only if at all possible.





             Russel

2011\06\13@085558 by Oli Glaser

flavicon
face
On 13/06/2011 13:06, Bob Ammerman wrote:
{Quote hidden}

The lowest note in standard tuning is E2, which is 82.41 Hz. However most tuners (including this one) tune bass guitar as well, since the two are commonly used together and share the same tuning (bass tuning is the same as the bottom 4 strings of guitar one octave down)
So, the lowest note mine currently goes to is the bottom string in a "dropped" bass tuning, which is D#1 at 38.9 Hz.
However since there are plenty of weird and wonderful stringed instruments and tunings for them, I was going to add a custom setting for non-standard tunings, based on a look up table of all the notes from C0 (16.35 Hz) up to C10 (16744Hz)
So even with the "normal" setting, the (low string) pulses would still be in the order of milliseconds. For a low guitar E it's ~1.2ms at 1:10 duty.


2011\06\13@090740 by Bob Ammerman

flavicon
face
> The lowest note in standard tuning is E2, which is 82.41 Hz. However
> most tuners (including this one) tune bass guitar as well, since the two
> are commonly used together and share the same tuning (bass tuning is the
> same as the bottom 4 strings of guitar one octave down)
> So, the lowest note mine currently goes to is the bottom string in a
> "dropped" bass tuning, which is D#1 at 38.9 Hz.
> However since there are plenty of weird and wonderful stringed
> instruments and tunings for them, I was going to add a custom setting
> for non-standard tunings, based on a look up table of all the notes from
> C0 (16.35 Hz) up to C10 (16744Hz)
> So even with the "normal" setting, the (low string) pulses would still
> be in the order of milliseconds. For a low guitar E it's ~1.2ms at 1:10
> duty.

Wow on two counts (which seem a bit contradictory)...

A) That a bass guitar is only one octave below a lead.

B) That a bass guitar is as low as it is.

My excuse... I am not a musician, nor do I play one on television.

-- Bob Ammerman
RAm Systems

2011\06\13@094023 by RussellMc

face picon face
> B) That a bass guitar is as low as it is.

8 string bass is an octave lower again (usually)

> My excuse... I am not a musician, nor do I play one on television.

Ditto. Ditto

2011\06\13@144728 by Sergey Dryga

flavicon
face
RussellMc <apptechnz <at> gmail.com> writes:


>I am assuming that the tuner runs the LEDs continuously at the tuning
>frequency while it is in use. If that is the case then there is no sense in
>using a really large cap. You only need one to hold enough charge for one
>pulse of the LED. You certainly aren't likely to have one big enough to hold
>a charge through a whole tuning cycle, which I am guessing will be at least
>several seconds.

>You need about 1000 uF provides ~~ 1V of voltage "droop" per mA per second..

Maybe it's easier to provide more stable supply to the PIC, e.g. use RC filter
on power going to PIC. in this case one needs to satisfy a smaller current to PIC.  
Sergey Dryga
http://beaglerobotics.com



2011\06\13@151501 by Bob Ammerman

flavicon
face
>>You need about 1000 uF provides ~~ 1V of voltage "droop" per mA per
>>second.

So, given a 20Hz low end and a 1:10 duty cycle the longest pulse would be 5ms.

Assuming the LEDs draw about 50ma then 1000uF would droop .25 volts for each pulse.

1000uF might be a bit large (or might not).

-- Bob Ammerman
RAm Systems

More... (looser matching)
- Last day of these posts
- In 2011 , 2012 only
- Today
- New search...