Separate front panel PCB, or use right-angle switches and di

[Steve S]

I'm trying to decide the best approach for a new design (consumer audio
product). For a similar product in the past, I used two PCBs:

1) Main PCB, mounted horizontal in the unit, containing 90% of the
circuitry, double-sided with components mainly surface mount on one side. It
also includes right-angle coax sockets and a DC socket that go through and
attach to the rear aluminium panel.

2) Front Panel PCB, mounted vertical in the unit, attached behind the front
panel using glued standoffs glued. It contained 3 x 6x6mm tactile buttons
and 4 x 0.7" seven-seg LED displays. This is a double-sided PCB, with all
components through-hole. FWIW the front panel is quite small - 1" high, by
5" wide.

I want to simplify and reduce the manufacturing costs for a new design of
product similar to this (quantities of 2-5k per year). So I'm considering
using just one PCB, attaching right-angle 7-seg LED displays and tactile
buttons to it for use as the front panel.

As I see it, the pro's of this approach would be:
a) Only one PCB
b) Less components - ribbon cable and header
c) No need to mount to front panel

The con's would be:
a) Harder to source right-angle displays and switches, and more expensive
b) Excess force on a connector on the rear panel would push front panel
switches and displays forward into the front panel

The rear connectors are screwed to the rear panel by one screw each, but I
think that last point may still be a concern - coax connections can be very
tight and attract a lot of user force sometimes.

Does anyone have any thoughts as to the best approach in order to minimize
manufacturing costs without causing problems down the line?

Thanks,
Steve

 

[Ian Stirling]

Steve S <me@privacy.net> wrote:

I'm trying to decide the best approach for a new design (consumer audio
product). For a similar product in the past, I used two PCBs:
snip
I want to simplify and reduce the manufacturing costs for a new design of
product similar to this (quantities of 2-5k per year). So I'm considering
using just one PCB, attaching right-angle 7-seg LED displays and tactile
buttons to it for use as the front panel.

As I see it, the pro's of this approach would be:
snip
The con's would be:
a) Harder to source right-angle displays and switches, and more expensive
b) Excess force on a connector on the rear panel would push front panel
c) Excess force on the front panel switches may snap them off.

If you can, pick the switches so that any force is borne by the front
panel, not a solder joint.
> switches and displays forward into the front panel

 

[Spehro Pefhany]

On Mon, 27 Sep 2004 13:12:39 GMT, the renowned "Steve S"
<me@privacy.net> wrote:

I'm trying to decide the best approach for a new design (consumer audio
product). For a similar product in the past, I used two PCBs:

1) Main PCB, mounted horizontal in the unit, containing 90% of the
circuitry, double-sided with components mainly surface mount on one side. It
also includes right-angle coax sockets and a DC socket that go through and
attach to the rear aluminium panel.

2) Front Panel PCB, mounted vertical in the unit, attached behind the front
panel using glued standoffs glued. It contained 3 x 6x6mm tactile buttons
and 4 x 0.7" seven-seg LED displays. This is a double-sided PCB, with all
components through-hole. FWIW the front panel is quite small - 1" high, by
5" wide.

I want to simplify and reduce the manufacturing costs for a new design of
product similar to this (quantities of 2-5k per year). So I'm considering
using just one PCB, attaching right-angle 7-seg LED displays and tactile
buttons to it for use as the front panel.

As I see it, the pro's of this approach would be:
a) Only one PCB
b) Less components - ribbon cable and header
c) No need to mount to front panel

The con's would be:
a) Harder to source right-angle displays and switches, and more expensive
b) Excess force on a connector on the rear panel would push front panel
switches and displays forward into the front panel

The rear connectors are screwed to the rear panel by one screw each, but I
think that last point may still be a concern - coax connections can be very
tight and attract a lot of user force sometimes.

Does anyone have any thoughts as to the best approach in order to minimize
manufacturing costs without causing problems down the line?

Thanks,
Steve

My opinion, which is worth what you paid for it, and based on limited
information (not sure what you mean by "glued"), is not to f*ck around
with the front panel ruggedness.

As you say, you'll likely pay more for right-angle parts and find them
harder to source, *especially* in your sub-consumer quantities.

The result won't be as good. You won't save much money, if any.

I'd look at soldering the flat cable rather than plugging it and
building/testing the assembly as one piece in production.


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Steve S]

"Spehro Pefhany" <speffSNIP@interlogDOTyou.knowwhat> wrote in message
news:0r7gl056b0al4o9v6liol9rja58il7utbm@4ax.com...

On Mon, 27 Sep 2004 13:12:39 GMT, the renowned "Steve S"
me@privacy.net> wrote:

I'm trying to decide the best approach for a new design (consumer audio
product). For a similar product in the past, I used two PCBs:

1) Main PCB, mounted horizontal in the unit, containing 90% of the
circuitry, double-sided with components mainly surface mount on one side.
It
also includes right-angle coax sockets and a DC socket that go through
and
attach to the rear aluminium panel.

2) Front Panel PCB, mounted vertical in the unit, attached behind the
front
panel using glued standoffs glued. It contained 3 x 6x6mm tactile buttons
and 4 x 0.7" seven-seg LED displays. This is a double-sided PCB, with all
components through-hole. FWIW the front panel is quite small - 1" high,
by
5" wide.

I want to simplify and reduce the manufacturing costs for a new design of
product similar to this (quantities of 2-5k per year). So I'm considering
using just one PCB, attaching right-angle 7-seg LED displays and tactile
buttons to it for use as the front panel.

As I see it, the pro's of this approach would be:
a) Only one PCB
b) Less components - ribbon cable and header
c) No need to mount to front panel

The con's would be:
a) Harder to source right-angle displays and switches, and more expensive
b) Excess force on a connector on the rear panel would push front panel
switches and displays forward into the front panel

The rear connectors are screwed to the rear panel by one screw each, but
I
think that last point may still be a concern - coax connections can be
very
tight and attract a lot of user force sometimes.

Does anyone have any thoughts as to the best approach in order to
minimize
manufacturing costs without causing problems down the line?

Thanks,
Steve

My opinion, which is worth what you paid for it, and based on limited
information (not sure what you mean by "glued"), is not to f*ck around
with the front panel ruggedness.

As you say, you'll likely pay more for right-angle parts and find them
harder to source, *especially* in your sub-consumer quantities.

The result won't be as good. You won't save much money, if any.

I'd look at soldering the flat cable rather than plugging it and
building/testing the assembly as one piece in production.


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers:
http://www.trexon.com
Embedded software/hardware/analog Info for designers:
http://www.speff.com

Thanks Sphero. FWIW , by gluing I really mean self-adhesive standoffs (since
I can't come through the front panel due to aesthetic reasons.) That is the
part of my design I really like the least, and saw the use of right-angle
switches to be a possible way to avoid it. However, I take all your points
on-board and realise it might not be such a good idea after all! The
suggestion for soldering the ribbon cable does sound very good though. I'm
tempted to leave one end socketed to speed of repairs, although the entire
cost of the assembled front panel board is only about $6, so might not be
worth treating it as a separate item.

Steve

 

[Steve S]

"John Woodgate" <jmw@jmwa.demon.contraspam.yuk> wrote in message
news:kx4M6lGuLCWBFwIh@jmwa.demon.co.uk...

I read in sci.electronics.design that Steve S <me@privacy.net> wrote (in
blU5d.51$GU1.10@newsfe5-win.ntli.net&gt about 'Separate front panel PCB,
or use right-angle switches and displays?', on Mon, 27 Sep 2004:

Does anyone have any thoughts as to the best approach in order to
minimize manufacturing costs without causing problems down the line?

Forget the front panel and put your controls and indicators on the 1
inch by 5 inch strip at the front of the top of the unit. You could
step this part down if that offers an advantage (use Courier font:

Upright or sloped
\ ________________
\------>| / |
| / |
Controls |/ | Side view
_\______| |
|_________________________|


That'll be USD 15000 for design consultancy. Used notes only, please.

Thanks for the suggestion John, check in the mail (sorry, just out of
notes!)

Definitely fix the rear panel sockets firmly. I suggest 12 mm bolts.

The original design uses a 12mm countersunk machine screw to fix each coax
connector to the aluminum rear panel, plus the connectors have plastic lugs
into the circuit board which I guess are meant to absorb much of the
insertion force. The DC socket (standard 2.1mm concentric type) has no
fixing hole so that relies on its through-hole soldering - it does not
require much force from the user to insert the power jack into it, but I
realize it could be abused. I could move to a DC socket with a fixing hole
or two. I wonder how many consumer items are returned for repair due to this
kind of stress on solder joints of connectors/switches?

Steve

--
Regards, John Woodgate, OOO - Own Opinions Only.
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

 

[John Woodgate]

I read in sci.electronics.design that Steve S <me@privacy.net> wrote (in
<hpX5d.83$GU1.69@newsfe5-win.ntli.net&gt about 'Separate front panel PCB,
or use right-angle switches and displays?', on Mon, 27 Sep 2004:

I wonder how many consumer items are returned
for repair due to this kind of stress on solder joints of
connectors/switches?

I've seen a few, and I 'don't do repairs', so I guess it's very common.

BTW, I meant 12 mm DIAMETER bolts. (;-)
--
Regards, John Woodgate, OOO - Own Opinions Only.
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

 

[Steve S]

"John Woodgate" <jmw@jmwa.demon.contraspam.yuk> wrote in message
news:6kQb$rJPmEWBFwLp@jmwa.demon.co.uk...

I read in sci.electronics.design that Steve S <me@privacy.net> wrote (in
hpX5d.83$GU1.69@newsfe5-win.ntli.net&gt about 'Separate front panel PCB,
or use right-angle switches and displays?', on Mon, 27 Sep 2004:

I wonder how many consumer items are returned
for repair due to this kind of stress on solder joints of
connectors/switches?

I've seen a few, and I 'don't do repairs', so I guess it's very common.

BTW, I meant 12 mm DIAMETER bolts. (;-)

Of course, I should have realized! "Gonna need a bigger drill!..."
--
Steve

--
Regards, John Woodgate, OOO - Own Opinions Only.
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

 

[John Fields]

On Mon, 27 Sep 2004 20:06:09 +0100, John Woodgate
<jmw@jmwa.demon.contraspam.yuk> wrote:

I read in sci.electronics.design that John Fields <jfields@austininstrum
ents.com> wrote (in <sfngl0hdfces80t5o7luuolva20dva2cbq@4ax.com&gt about
'Separate front panel PCB, or use right-angle switches and displays?',
on Mon, 27 Sep 2004:
What I do for
applications where I can't use domes and I need feedback from the
keypad/switches is to use a cheap magnetic beeper which gives me a 100ms
or so beep whenever a key gets made. So far, clients like it.

I don't, for one.

---
You're not a client, yet!-)
---

I can't hear the beeps. If you want beeps, make them a
nice alto frequency. 400 Hz is good.

---
LAST ONES I USED WERE AROUND 2kHz AT 85dB.

THINK YOU MIGHT BE ABLE TO HEAR THOSE?

--
John Fields

 

[budgie]

On Mon, 27 Sep 2004 16:41:49 GMT, "Steve S" <me@privacy.net> wrote:

(snip)

The DC socket (standard 2.1mm concentric type) has no
fixing hole so that relies on its through-hole soldering - it does not
require much force from the user to insert the power jack into it, but I
realize it could be abused. I could move to a DC socket with a fixing hole
or two. I wonder how many consumer items are returned for repair due to this
kind of stress on solder joints of connectors/switches?

Laptops by the dozen. It's the most common electromechanical failure item on
them IMOE.

 

[Keith Wootten]

In message <sfngl0hdfces80t5o7luuolva20dva2cbq@4ax.com>, John Fields
<jfields@austininstruments.com> writes

<snip>

If you don't like membrane switches you can always go to
tactile-feedback dome switches, But they're more expensive. What I do
for applications where I can't use domes and I need feedback from the
keypad/switches is to use a cheap magnetic beeper which gives me a
100ms or so beep whenever a key gets made. So far, clients like it.

Never tried it, but how about whacking the panel from behind with a
solenoid every time a key is pressed? Might work.

Cheers
--
Keith Wootten

 

[Rich Grise]

On Monday 27 September 2004 07:29 am, Steve S did deign to grace us with the
following:

Thanks Sphero. FWIW , by gluing I really mean self-adhesive standoffs
(since I can't come through the front panel due to aesthetic reasons.)
That is the part of my design I really like the least, and saw the use of
right-angle switches to be a possible way to avoid it. However, I take all
your points on-board and realise it might not be such a good idea after
all! The suggestion for soldering the ribbon cable does sound very good
though. I'm tempted to leave one end socketed to speed of repairs,
although the entire cost of the assembled front panel board is only about
$6, so might not be worth treating it as a separate item.

You can get press-in studs that are invisible from the front, after
painting. I think they're called Pem studs. Even flathead screws could
work, if you could cover them with the above-mentioned overlay or something.

Or, you could make the screw heads part of the decor.

Good Luck!
Rich

 

[John Woodgate]

I read in sci.electronics.design that Rich Grise <null@example.net>
wrote (in <XDl6d.12099$Cn.545@trnddc04&gt about 'Separate front panel
PCB, or use right-angle switches and displays?', on Tue, 28 Sep 2004:

5 cycles of 1 KHz goes "tick" - I suspect 5 cycles of 400 Hz goes
"tock". But beeps can work, I guess.

I need something that goes 'pooop'. (;-)
--
Regards, John Woodgate, OOO - Own Opinions Only.
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

 

[Dave VanHorn]

I need something that goes 'pooop'. (;-)

Dogs and cats are both right angle mount, but Dogs are SM, and Cats are
Thru-Hole.


--
KC6ETE Dave's Engineering Page, www.dvanhorn.org
Microcontroller Consultant, specializing in Atmel AVR