We're doing a custom silicone pushbutton thing, sort of like this one: https://www.dropbox.com/s/8hihd888w7rtxmx/Fingers_2.JPG?raw=1 https://www.dropbox.com/s/ruo7zujvx7mxerz/Fingers_3.JPG?raw=1 The pcb below the keypad will look something like https://www.dropbox.com/s/qyu2nrkhom78wkl/Fingers_1.jpg?raw=1 where the center hole lets light through to back-light the key, and the two finger patterns are where the carbon buttons land when the key is pressed. This will be ENIG, gold flash over nickel over PCB copper. The left and right finger patterns will be electrically in parallel. It's possible for one carbon button to contact, or both, depending on how the button is pushed. Googling, I find zillions of different and bizarre PCB patterns for the buttons to land on. Crazy shapes. Has anyone done this? Any opinions on the optimum PCB pattern? -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
silicone keypad carbon button contacts
Started by ●July 18, 2017
Reply by ●July 18, 20172017-07-18
On Tue, 18 Jul 2017 14:05:02 -0700, John Larkin <jjlarkin@highland_snip_technology.com> wrote:>Has anyone done this? Any opinions on the optimum PCB pattern?Any points for doing it wrong? I let the PCB designer do his own thing and ended up with an unreliable switch connection. This was on a marine radio where water sometimes condenses under the buttons. The idea is to get the most contact area under the carbon "pill" without the risk of the traces being too close and getting bridged by condensed water. Some recommendations: <http://www.abatek.com/designguide/04Electrical.html#trace_pattern> I don't recall exactly what we used but a 0.3mm gap sounds about right. My guess(tm) is that the reason for the weird circular contact patterns is that it eliminates the long parallel trace runs of the interdigital pattern (shown in above URL) which might be affected by leakage between traces. The circular pattern seems to have larger gaps between traces. This is only a guess. Graphite(+0.25v) and copper(-0.35v) are very much apart on the galvanic series will react. So, you need gold plating. One thing to watch out for with those switches is the silicone rubber buttons "bleeding" oil onto the gold contacts. I don't know the exact mechanism, but after a few months in the hot sun, an oily liquid (plasticizer?) would leak out of the rubber and create a slimy mess. The oil also seems to attack the carbon making it slowly disappear. Eventually, the switch would fail to make a connection. The problem is common in TV remote controls. If caught early, some household cleaner (409) would clean up the oil and allow the switch to operate normally. If left too long, the carbon contacts would need to be resurfaced with one of the kits found on eBay. <https://www.ebay.com/sch/i.html?_nkw=conductive+rubber+keypad+fix> Not every keypad exhibited this problem. I never had the time to investigate beyond finding this and writing a repair procedure. Anyway, be prepared to make the keypad easy to clean, repair, or replace. The mechanical engineer that did the package saved a few pennies and heat swaged the keypad in place. When it started exhibiting the aforementioned problem, the only option we initially offered was to replace the entire front panel instead of just the keyboard. -- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
Reply by ●July 18, 20172017-07-18
On Tue, 18 Jul 2017 16:44:34 -0700, Jeff Liebermann <jeffl@cruzio.com> wrote:><http://www.abatek.com/designguide/04Electrical.html#trace_pattern>More... <http://www.siliconedynamics.net/wp-content/uploads/design_guide/9271Silicone-Dynamics-Keypad-Design-Guide.pdf> See Pg 25. <http://www.northpointech.com/pcb-contact-design-guide/> <http://www.jw-electronic-components.de/pdf/Design%20guide%20for%20rubber%20keypads.pdf> See Pg 17. Seems to be plenty of design guides available. Google for more: <https://www.google.com/search?q=silicone+keypad+contact+design> -- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
Reply by ●July 18, 20172017-07-18
On Tue, 18 Jul 2017 16:44:34 -0700, Jeff Liebermann <jeffl@cruzio.com> wrote:>On Tue, 18 Jul 2017 14:05:02 -0700, John Larkin ><jjlarkin@highland_snip_technology.com> wrote: > >>Has anyone done this? Any opinions on the optimum PCB pattern? > >Any points for doing it wrong? I let the PCB designer do his own >thing and ended up with an unreliable switch connection. This was on >a marine radio where water sometimes condenses under the buttons. The >idea is to get the most contact area under the carbon "pill" without >the risk of the traces being too close and getting bridged by >condensed water. Some recommendations: ><http://www.abatek.com/designguide/04Electrical.html#trace_pattern> >I don't recall exactly what we used but a 0.3mm gap sounds about >right. > >My guess(tm) is that the reason for the weird circular contact >patterns is that it eliminates the long parallel trace runs of the >interdigital pattern (shown in above URL) which might be affected by >leakage between traces. The circular pattern seems to have larger >gaps between traces. This is only a guess. > >Graphite(+0.25v) and copper(-0.35v) are very much apart on the >galvanic series will react. So, you need gold plating. > >One thing to watch out for with those switches is the silicone rubber >buttons "bleeding" oil onto the gold contacts. I don't know the exact >mechanism, but after a few months in the hot sun, an oily liquid >(plasticizer?) would leak out of the rubber and create a slimy mess. >The oil also seems to attack the carbon making it slowly disappear. >Eventually, the switch would fail to make a connection. The problem >is common in TV remote controls. > >If caught early, some household cleaner (409) would clean up the oil >and allow the switch to operate normally. If left too long, the >carbon contacts would need to be resurfaced with one of the kits found >on eBay. ><https://www.ebay.com/sch/i.html?_nkw=conductive+rubber+keypad+fix> >Not every keypad exhibited this problem. I never had the time to >investigate beyond finding this and writing a repair procedure. > >Anyway, be prepared to make the keypad easy to clean, repair, or >replace. The mechanical engineer that did the package saved a few >pennies and heat swaged the keypad in place. When it started >exhibiting the aforementioned problem, the only option we initially >offered was to replace the entire front panel instead of just the >keyboard.Interesting links, thanks. I don't think I'll worry about water. My pullup resistors will be 10K! The silicone pad will be difficult to remove, sandwiched between a PCB and an aluminum panel, so I hope it won't need cleaning. I'll ask the vendor about that oozing gunk possibility. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Reply by ●July 18, 20172017-07-18
On Tue, 18 Jul 2017 16:44:34 -0700, Jeff Liebermann <jeffl@cruzio.com> wrote:>On Tue, 18 Jul 2017 14:05:02 -0700, John Larkin ><jjlarkin@highland_snip_technology.com> wrote: > >>Has anyone done this? Any opinions on the optimum PCB pattern? > >Any points for doing it wrong? I let the PCB designer do his own >thing and ended up with an unreliable switch connection. This was on >a marine radio where water sometimes condenses under the buttons. The >idea is to get the most contact area under the carbon "pill" without >the risk of the traces being too close and getting bridged by >condensed water. Some recommendations: ><http://www.abatek.com/designguide/04Electrical.html#trace_pattern> >I don't recall exactly what we used but a 0.3mm gap sounds about >right. > >My guess(tm) is that the reason for the weird circular contact >patterns is that it eliminates the long parallel trace runs of the >interdigital pattern (shown in above URL) which might be affected by >leakage between traces. The circular pattern seems to have larger >gaps between traces. This is only a guess. > >Graphite(+0.25v) and copper(-0.35v) are very much apart on the >galvanic series will react. So, you need gold plating. > >One thing to watch out for with those switches is the silicone rubber >buttons "bleeding" oil onto the gold contacts. I don't know the exact >mechanism, but after a few months in the hot sun, an oily liquid >(plasticizer?) would leak out of the rubber and create a slimy mess. >The oil also seems to attack the carbon making it slowly disappear. >Eventually, the switch would fail to make a connection. The problem >is common in TV remote controls. > >If caught early, some household cleaner (409) would clean up the oil >and allow the switch to operate normally. If left too long, the >carbon contacts would need to be resurfaced with one of the kits found >on eBay. ><https://www.ebay.com/sch/i.html?_nkw=conductive+rubber+keypad+fix> >Not every keypad exhibited this problem. I never had the time to >investigate beyond finding this and writing a repair procedure. > >Anyway, be prepared to make the keypad easy to clean, repair, or >replace. The mechanical engineer that did the package saved a few >pennies and heat swaged the keypad in place. When it started >exhibiting the aforementioned problem, the only option we initially >offered was to replace the entire front panel instead of just the >keyboard.Sounds like Amine Blush. It's a waxy oil that leaches out as the Plastic cures. Probably a poor mix on the key matrix. Cheers
Reply by ●July 18, 20172017-07-18
On 19/07/17 10:38, Martin Riddle wrote:> On Tue, 18 Jul 2017 16:44:34 -0700, Jeff Liebermann <jeffl@cruzio.com> > wrote: > >> On Tue, 18 Jul 2017 14:05:02 -0700, John Larkin >> <jjlarkin@highland_snip_technology.com> wrote: >> >>> Has anyone done this? Any opinions on the optimum PCB pattern? >> >> Any points for doing it wrong? I let the PCB designer do his own >> thing and ended up with an unreliable switch connection. This was on >> a marine radio where water sometimes condenses under the buttons. The >> idea is to get the most contact area under the carbon "pill" without >> the risk of the traces being too close and getting bridged by >> condensed water. Some recommendations: >> <http://www.abatek.com/designguide/04Electrical.html#trace_pattern> >> I don't recall exactly what we used but a 0.3mm gap sounds about >> right. >> >> My guess(tm) is that the reason for the weird circular contact >> patterns is that it eliminates the long parallel trace runs of the >> interdigital pattern (shown in above URL) which might be affected by >> leakage between traces. The circular pattern seems to have larger >> gaps between traces. This is only a guess. >> >> Graphite(+0.25v) and copper(-0.35v) are very much apart on the >> galvanic series will react. So, you need gold plating. >> >> One thing to watch out for with those switches is the silicone rubber >> buttons "bleeding" oil onto the gold contacts. I don't know the exact >> mechanism, but after a few months in the hot sun, an oily liquid >> (plasticizer?) would leak out of the rubber and create a slimy mess. >> The oil also seems to attack the carbon making it slowly disappear. >> Eventually, the switch would fail to make a connection. The problem >> is common in TV remote controls. >> >> If caught early, some household cleaner (409) would clean up the oil >> and allow the switch to operate normally. If left too long, the >> carbon contacts would need to be resurfaced with one of the kits found >> on eBay. >> <https://www.ebay.com/sch/i.html?_nkw=conductive+rubber+keypad+fix> >> Not every keypad exhibited this problem. I never had the time to >> investigate beyond finding this and writing a repair procedure. >> >> Anyway, be prepared to make the keypad easy to clean, repair, or >> replace. The mechanical engineer that did the package saved a few >> pennies and heat swaged the keypad in place. When it started >> exhibiting the aforementioned problem, the only option we initially >> offered was to replace the entire front panel instead of just the >> keyboard. > > Sounds like Amine Blush. It's a waxy oil that leaches out as the > Plastic cures. Probably a poor mix on the key matrix.I've had this happen to several remote controls after a few years. Disassembly and thorough solvent-based cleaning fixes them temporarily, but it recurs more and more quickly. I don't think it's a plasticizer - that would be more noticeable when it was new and would reduce with age. I think the compound is de-polymerising. It's terminal, just not immediately. Clifford Heath.
Reply by ●July 19, 20172017-07-19
On Tue, 18 Jul 2017 17:35:12 -0700, John Larkin <jjlarkin@highland_snip_technology.com> wrote:>I don't think I'll worry about water. My pullup resistors will be 10K!What, me worry? Here's a simple test to try. Put the keyboard and associated electronics into your environmental chamber or fridge and drop the temperature to well below the dew point. Bring the device out of the chamber or fridge into the warm room. Water will condense on everything but the real question will be will it condense inside the keypad and simulate a keypress. One of my early jobs was for a company that made communications accessories. In this case, it was a touch tone encoder for 2way radios in skool buses. Everything worked just fine until they were installed in the buses. The buses would sit outside in the cold night. At the crack-o-dawn, the engines would be started and the interiors warmed up in preparation for picking up the students. Water would condense inside the keypad, which would cause the transmitters to key up. Having about 50 transmitters stuck on the air just as the buses were leaving the yard was not a good thing.>The silicone pad will be difficult to remove, sandwiched between a PCB >and an aluminum panel, so I hope it won't need cleaning. I'll ask the >vendor about that oozing gunk possibility.Yep, that's the way most of them are built, with the silicon keyboard sandwiched between an expensive decorative front panel, and an expensive PCB full of electronics. Got any TV remotes, cordless phones, or other devices that use rubberish keypads? If you take them apart and look at how they're built, you'll notice that the back of the front panel has ridges around each button. These ridges prevent water, dust, or debris from getting between the front panel and the silicon keypad. The ridge is under compression from the keypad PCB mounting hardware to the front panel, thus providing a very good seal. However, doing the same thing with an aluminum front panel is not going to happen. There's just no way to coin a ridge into the front panel. It might be possible to glue a frame with similar ridges to the front panel, but that costs money and is labor intensive. I don't know if you're going to need any environmental protection, but I thought it might help to explain why and how it's done.>I'll ask the >vendor about that oozing gunk possibility.It's quite real: <https://groups.google.com/forum/#!topic/sci.electronics.repair/wU2OlitgXpc> <http://www.michaelshell.org/gadgetsandfixes/keypadsiliconeoil.html> <http://p1repair.com/blog/2013/05/28/rubber-keypad-problems-liquid-oil-moisture-found-inside-the-buttons-what-is-it-where-does-it-come-from/> Two other potential problems. Women and some men with very sharp fingernails will cut into the soft silicone with their fingernails. That's another reason for an easily replaceable keypad. In the photos, your buttons look small enough that I would not expect a problem but you might want to make sure that the buttons stick out from the front panel far enough that the typical fingernail doesn't scratch the front panel. I also noticed that the silk screened or painted lettering on the buttons uses rather narrow lines. The rubber stretches and moves sufficiently that the lettering might flake off. Try wider line widths for better adhesion. -- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
Reply by ●July 19, 20172017-07-19
On Tue, 18 Jul 2017 20:38:03 -0400, Martin Riddle <martin_ridd@verizon.net> wrote:>Sounds like Amine Blush. It's a waxy oil that leaches out as the >Plastic cures. Probably a poor mix on the key matrix.Thanks. I'll do some reading but I don't think so. Amine blush is what happens when epoxy paint hardener reacts with water and CO2 and creates a color splotch. <http://www.wolverinecoatings.com/TIB/Minimizing,_Identifying,_and_Remediating_Amine_Blush-TIB.pdf> I've never seen the effect and have never heard the term used in reference to rubbers or plastics. <http://www.epoxycraft.com/amine-blush-what-you-need-to-know/> <http://p1repair.com/blog/2013/05/28/rubber-keypad-problems-liquid-oil-moisture-found-inside-the-buttons-what-is-it-where-does-it-come-from/> "The liquid found in rubber keypads is left over silicone oil from the manufacturing process when the rubber keypad has not been baked enough... it�s a gooey center, undercooked rubber keypad!" This would also explain why it happens with some keyboards, but not others. -- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
Reply by ●July 19, 20172017-07-19
On Wed, 19 Jul 2017 10:53:27 +1000, Clifford Heath <no.spam@please.net> wrote as underneath : snip> >I've had this happen to several remote controls after a few years. >Disassembly and thorough solvent-based cleaning fixes them >temporarily, but it recurs more and more quickly. I don't think >it's a plasticizer - that would be more noticeable when it was new >and would reduce with age. I think the compound is de-polymerising. > >It's terminal, just not immediately. >Recently I painted silver (Ecolit) conductive on the pads (6 Yr old Panasonic TV remote) after careful cleaning with thinners and realising the conductivity of the pads was the problem apart from the goo, this fix has worked extreemly well for the last few months but I have no idea if it will have 'legs' ! C+
Reply by ●July 19, 20172017-07-19
On Tue, 18 Jul 2017 14:05:02 -0700, the renowned John Larkin <jjlarkin@highland_snip_technology.com> wrote:> > >We're doing a custom silicone pushbutton thing, sort of like this one: > >https://www.dropbox.com/s/8hihd888w7rtxmx/Fingers_2.JPG?raw=1 > >https://www.dropbox.com/s/ruo7zujvx7mxerz/Fingers_3.JPG?raw=1 > >The pcb below the keypad will look something like > >https://www.dropbox.com/s/qyu2nrkhom78wkl/Fingers_1.jpg?raw=1 > >where the center hole lets light through to back-light the key, and >the two finger patterns are where the carbon buttons land when the key >is pressed. This will be ENIG, gold flash over nickel over PCB copper. > >The left and right finger patterns will be electrically in parallel. >It's possible for one carbon button to contact, or both, depending on >how the button is pushed. > >Googling, I find zillions of different and bizarre PCB patterns for >the buttons to land on. Crazy shapes. > >Has anyone done this? Any opinions on the optimum PCB pattern?Do the silicon keyboard design guides (there are several out there) provide any guidance to the PCB patterns? --sp -- Best regards, Spehro Pefhany