Why DiSEqC switch *after* positioner?

[bill190]

DiSEqC "devices" like switches or positioners have small "computers" in them. They take time to "power up" or "boot" after they receive power. And this is the key!

Think of this as you would your computer. You can't start typing on your computer as soon as you turn on the power. Your computer takes time to boot up, then you can start typing or issuing "commands".

DiSEqC switches REMOVE power from connections which are not in use. So when your receiver is connected to a DiSEqC switch and is currently using say "LNB 1", then LNB 2, 3, and 4 have no power. When your receiver switches to say LNB 2, it switches on power to that connection and removes power from the other connections.

So one problem with placing a DiSEqC positioner AFTER a DiSEqC switch can be that the receiver is issuing commands right away (like typing on your computer), but the DiSEqC positioner computer is not yet booted up and running. So it would not "hear" the commands from the receiver.

Note that DiSEqC "computers take less than a second to "boot" up though. And mpeg receivers may or may not be designed to wait before issuing DiSEqC commands or to repeat DiSEqC commands. However my experience in the computer world is that software designers do not understand hardware very well, so don't count on it.

Another problem is that DiSEqC switches can be manufactured less expensively if they are designed to not "pass" the receiver DiSEqC commands on down the coax line. So they are kind of a "dead end" device. Last in the chain except for the LNB's which connect to them. So they are designed to control the LNB's, but nothing else.

Then a third problem is power availability on the coax line to run everything. A positioner is a "power hog" and needs to be able to shut off power to everything else down the coax line like switches and LNB's, then use all the power for itself to move the dish.

Note that DiSEqC positioners are designed to pass along DiSEqC commands to devices down the line such as switches and LNB's.

Bottom line, put DiSEqC positioner first after receiver, THEN switches after that, then LNB's.

More on this...

See "DiSEqC Specifications", then "Application Information for using a "PIC" Microcontroller in DiSEqC LNB and Simple Switcher Applications", "Application Information for LNBs and Switchers", and "Positioner Application Note"...

What is DiSEqC?

[wejones]

Originally Posted by bill190

Note that DiSEqC "computers take less than a second to "boot" up though. And mpeg receivers may or may not be designed to wait before issuing DiSEqC commands or to repeat DiSEqC commands. ...

The receivers I have (Fortec) will repeat SWITCH commands over and over indefinately, until the receiver gets a lock, and will start up the commands again if lock is lost.
However, with respect to diseqC1.2/USALS MOTOR commands, they seem to only be sent once, so you are right that there are potential problems if you put the switch first, which was what I was alluding to in another thread. My theory of why they don't repeat the motor commands is that I think that there is potential for problems if you interrupt a motor after it has started going to a sat position. So I think they just send the command once rather than risk confusing the motor by giving it multiple commands while it is already on the move.

Originally Posted by bill190

Another problem is that DiSEqC switches can be manufactured less expensively if they are designed to not "pass" the receiver DiSEqC commands on down the coax line. So they are kind of a "dead end" device. Last in the chain except for the LNB's which connect to them. So they are designed to control the LNB's, but nothing else.

I'm not so sure I completely beleive this, but I haven't tested this. The reason I"m a bit skeptical is that if this were true, then a universal LNBF wouldn't work after a switch, and I've never heard any complaints about that. The reason I think a universal lnbf wouldn't work if DiseqC commands are blocked is that diseqc commands are sent via modulated 22khz, and universal lnbfs require that 22khz be passed by the switch. However it is possible that the circuitry in the switch that reads the diseqC might damp the signal enough that subsequent devices have a hard time reading it, even though they can detect the 22khz. So if true, I'd think this would be switch dependent, might be true with some switches and not true with others.

Originally Posted by bill190

..
Bottom line, put DiSEqC positioner first after receiver, THEN switches after that, then LNB's.
..

Yeah, that's certainly less prone to potential problems.