If an input pin is a current source, it seems that the resistor value would have no effect on the consumed power. It would be divided between the resistor and the transistor inside. And 1uA at 5V is 5uW, period. This, of course, is applicable to pullups on unused inputs. Once you actually use the pin by driving it low, then you are sinking the current through the resistor to ground. You can still save power by not driving it low that often. A keypad that responds every 100ms (and may be scanned every 10ms) is indistinguishable from one scanned continuously. And the capacitance problem can be solved by waiting a few microseconds before the next read. There should be ample time for this with a keypad. Then you can leave the resistors at their higher value. On Thu, Apr 4, 2013 at 7:26 AM, Allen Mulvey wrote: > I'm a hobbyist and tend to do things on the cheap side. > Agreed, you can't (or shouldn't) use most surplus or oddball > stuff in commercial applications. I added the parenthetic > part because I worked on a NYS University campus where most > of the fire alarm systems were fabricated from obsolete > junk. (Lowest bidder.) For home-brew or student projects > these keyboards are pretty nice, made by AMP. > > Here is one for my garage door opener: > http://www.mulvey.us/Keypad1.jpg > http://www.mulvey.us/Keypad2.jpg > > Allen > > -----Original Message----- > From: piclist-bounces@mit.edu > [mailto:piclist-bounces@mit.edu] On Behalf Of Isaac Marino > Bavaresco > Sent: Thursday, April 04, 2013 7:36 AM > To: Microcontroller discussion list - Public. > Subject: Re: [PIC] Pull Up Resistors > > That won't work for a professional commercial product, but > if you are > building just a couple of them for hobby, then OK. > > I insist in pointing out that the conventional method works > perfectly > well for multiple simultaneous key presses and it is not > hard at all > implementing it. > > With your keyboard you are limited to detect keys all in the > same row or > column, not diagonal etc. With the conventional method you > can detect > any number of keys in the same row or column, or in any > arrangement as > long you don't press keys that make all the three vertexes > of an "L" shape. > > Besides, it is possible to share four pins of the MCU > between the > keyboard and an LCD module. > > > Best regards, > > Isaac > > > > Em 3/4/2013 22:47, Allen Mulvey escreveu: > > These were on sale for $0.79 USD. Now they are now $1.49. > > They have 2 key markings for fire, police, and ambulance > as > > well as 3 green LED, 1 yellow, and 2 red. The size is such > > that you can shave off about 3/32" and fit it into clear > > covered exterior outlet box. The face is printed on the > back > > of a clear plastic sheet set into the front so you can > peel > > it off and make your own if you wish. > > > > > http://www.goldmine-elec-products.com/prodinfo.asp?number=3DG1 > > 7927 > > > > Allen > > > > -----Original Message----- > > From: piclist-bounces@mit.edu > > [mailto:piclist-bounces@mit.edu] On Behalf Of Isaac Marino > > Bavaresco > > Sent: Wednesday, April 03, 2013 8:53 PM > > To: Microcontroller discussion list - Public. > > Subject: Re: [PIC] Pull Up Resistors > > > > With this arrangement you will need some unusual switches > > (three > > terminal make-make) > > > > And with the conventional method it is also possible to > > detect multiple > > simultaneous presses. > > > > > > Isaac > > > > > > Em 3/4/2013 19:25, Allen Mulvey escreveu: > >> Thank you all for the excellent information and > resources. > > I > >> had checked Microchip and MikroElektronica sites and > found > >> nothing. I missed the significance of the current rating > > for > >> the internal pullups. > >> > >> I have been experimenting with a slightly different kind > > of > >> matrix keypad. Instead of shorting row and column, it > >> grounds both. This eliminates the need to scan and allows > >> the selection of multiple key presses although not all > may > >> be unique. I think much of the problem I have seen has > > been > >> caused by capacitance. Earlier I didn't think about the > > R-C > >> timer effect. If I connect the keypad with a foot of > cat-5 > >> cable the 20k resistors work fine. If I use 6" of ribbon > >> cable 20k is too big but 4k7 or slightly less works fine. > >> There also appears to be a difference depending on which > > end > >> of the cable the resistors are attached although I have > > not > >> tested this very well. > >> > >> Thanks again, > >> Allen > >> > >> > >> > > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list > archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > > > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .