Bouncy MET-6N Keypads
TurnipTruck
Posts: 1,485
in AMX Hardware
Greetings,
I just completed a system with a bunch of MET-6N keypads. It appear that they have no debounce in their firmware. If you push the button slowly, you can get several push events returned. You must almost flick the buttons to get a single push and release.
Anyone have similar experiences?
I just completed a system with a bunch of MET-6N keypads. It appear that they have no debounce in their firmware. If you push the button slowly, you can get several push events returned. You must almost flick the buttons to get a single push and release.
Anyone have similar experiences?
0
Comments
e
I did a couple jobs with the MET-6 and MET-13 and did not have any issues, I thought they were very well made and the price is great.
If you turn on Notifications in Netlinx Studio, what shows up in the log? Do you see multiple pushes before a release? Or do you see several push/release entries all logged within the same second? If you are getting anything other than 1 push and 1 release, I?m sure AMX and others would be interested in seeing the data.
I never actually tried to push a button slowly but I will the next time I get the chance and let you know if I observe anything out of the ordinary.
Is it possible that you?ve got one or more of the keypads addressed with duplicate IDs? Something like that will mess things up. Just a thought?
Wiring an Axlink device with Cat5 and using multiple wires (doubling up) for the data pair can cause multiple presses/releases. The theory is that there are multiple paths for the Axlink data.
I'm not sure I buy this theory but just throwing it out there....
Interesting point. I've had to beat my installers over the head about proper use of CAT 5; they alwyas want to double up a pair for better current capacity, and I have to continuously remind them if they really need to do this, maintain the pairs by doubling two pairs into one, not shorting a single pair. Now, I also don't think there is much to the "two paths" theory, but there is something to be said for crosstalk and noise interference.
There's not enough length in the 2nd or evern third wire to make a big enough differnece in when the voltage change happens. The signal being desribed is just voltage going from high to low. The phase shift from one strand to another in a 100 meter hunk of wire is in the picosecond range. That's not enough time to offset the voltage ups or downs by any appreciable amount.
The problem is that twisted pair wire is not as good at shielding the wire from outside RF noise and magnetic fields. By the time the voltage gets down the line, it has accumulated a lot of noise and static. This causes the recieving end to have a harder time deciphering what's real stuff vs. garbage.
The other issue is that the signal is DC. single strand wire is not as good at conveying DC signals. DC tends to travel over the surface of the conductor, AC down the core of the conductor. That's why home AC wire is single solid core. DC signals tend to transfer better over starnded wire. So, the best choice for a communication like AxLink or RS-232 is starnded conductor with a shield.
Now, IP network signal is also DC. However, DC signals at the frequency range of IP networks (10mhz or better) tend to act like AC. So, single-core conductors like CAT5e are the best choice. The twists in the wire tend to lessen the inteference from one pair to another. They do not shield from outside interference, however.
So, wheter or not to use CAT5e on Axlink is simply a matter of how much noise is accumulating on the line. If you're doing it and it's working, then no problem. But, it's not recommended.