)
I was pretty surprised that the just used some sort of rubber and then pressed the keys on over it.mr_a500 wrote:I just took webwit's advice and hit it with a hammer. (I knew I could always sue webwit if anything went wrong
Those contamination shields are always in a pitiful state, but at least they usually protect the rest of the keyboard from getting ruined, no matter how badly it was stored.
I think it's pretty smart actually. It's a cheap and simple way to get it resistant to beverages spill when it was new and the rubber was fresh, don't think they ment it to last 30+ yearsMuirium wrote:<Imagines the new computer smell these boards would have had, out of the box.>
How does beam spring's underside compare to Model F's flip plates? Do those squares pop up when the switch is activated? It looks a lot like its descendant, especially on the PCB, but I read in Xwhatsit's documentation that beam spring is the reverse polarity, capacitance wise, compared to buckling spring.
Yes. As you push down when you reach the activation point the fly plate pulls up and the key activation force just drops. Has a great feel to it.Muirium wrote:<Imagines the new computer smell these boards would have had, out of the box.>
How does beam spring's underside compare to Model F's flip plates? Do those squares pop up when the switch is activated? It looks a lot like its descendant, especially on the PCB, but I read in Xwhatsit's documentation that beam spring is the reverse polarity, capacitance wise, compared to buckling spring.
5251Icarium wrote:Well, I pulled really hard on one of the metal stems but couldn't get it out. That's what I was asking earlier...
Oh and I just looked through http://deskthority.net/wiki/IBM_Beam_Spring_Keyboards but couldn't find this exact board. Does anybody know what it is?
Yeah, that's how I'm planning to use it.E TwentyNine wrote:Are you guys really that concerned about contamination? I have mine with the shield removed and no plans on replacing it.
Yes, the "fly plate" (to use IBM's terminology) retracts back into the switch housing when the switch actuates, so it moves away from the contact pad instead of moving towards it as in the Model F. You can place a Beam Spring on top of a Model F contact pad, but the Model F's controller will think that the switch at rest has been pressed, and that its at rest when you press it.Muirium wrote:How does beam spring's underside compare to Model F's flip plates? Do those squares pop up when the switch is activated? It looks a lot like its descendant, especially on the PCB, but I read in Xwhatsit's documentation that beam spring is the reverse polarity, capacitance wise, compared to buckling spring.
My daily-driver 3727 gets a lot of use at work, and I removed the shield from it about 6 months ago. It was originally in good condition, but with the keyboard getting heavy use it fell to bits.E TwentyNine wrote:Are you guys really that concerned about contamination? I have mine with the shield removed and no plans on replacing it.
What oscilloscope do you use? What kind would you recommend for someone getting into electronics as a hobbyist? From briefly looking into this, I see that there are PC-based scopes (e.g., picoscope), which has a certain appeal e.g., combining several instruments into one in a cost-effective way, but purists tend to disrecommend them in favor of standalone bench scopes, such as the Rigol DS1102E, perhaps used in conjunction with a logic analyzer and signal generator.xwhatsit wrote:Absolutely—I started getting glitchy keys that would repeat or just go off spamming by their own. Initially I had a fright that there was something wrong with my controller, but after viewing traces on the oscilloscope I could see that those keys were actually higher capacitance for whatever reason—and it seems a bit of flaky rubber shield or hair is enough to cause that. Removing the foreign matter resolved the problem immediately.
M's of course would need a small brick or pebble to cause a misfire, being ohmic switched rather than capacitive. The Fs are much more sealed up generally, and with the higher levels of capacitance they produce even if there was dirt in the right place there would need to be much more of it to reach such a threshold. The fact that they're `normally open' w.r.t. capacitance may help here too; foreign matter is likely to inhibit capacitance (by holding the flyplate or foot up from the PCB, or interfering by being a poor dielectric etc.), and with the beamsprings, they are higher capacitance when unpressed, unlike the Model Fs. If you have a 1-in-a-1000-scan glitch because of some fluff in your key, this is enough to cause a key to do a fake press on a beamspring, but will only cause a momentary delay or unintended double-press on a Model F when you *are* actually pressing the key.
Could you tell us the exact material you used for your contamination shield and where you purchased it?Parak wrote:Here's how I did it: linkyHalvar wrote:I experimented with different kinds of foils to replace the contamination shields, but I haven't found a good solution yet.
YMMV