Okay, I did some checking into why the 22K switch won't talk to a DISEqC digital switch. Well, it sorta does, but it mumbles. I didn't use a 4x4 22K unit in this test, only two varieties of two-way switches for a comparison. The 22K quality switch made by Hughes, and a cheaper design made by Gecen.
The problem with 22K Hughes switches is that they won't pass clean DISEqC commands. DISEqC switches will pass enough of the 22K tone to switch a sensitive 22K switch. To select 8 feeds with four 22K switches, one on each of the four ports of a DISEqC switch, might not be a good idea.. The problem with that plan is the surface mount voltage switching transistors in the DISEqC switch can get very hot. This is because the voltage flowing through an active port is twice what it should be, due to the design of the Hughes 22K switches, that do not switch the DC voltage. They select the RF of either port on command, but the voltage is not switched. The good thing about that design is there are no sensitive voltage switching transistors to be damaged by shorts or surges, as in the more vulnerable and unforgiving DISEqC variety of switch.
Using one 22K switch by itself is not a problem, but connecting it after a DISEqC is a problem. I've burned out a port on at least one DISEqC switch because the transistors couldn't take the extra load of two lnbf's being powered at once. Installing the 22K switch ahead of the DISEqC switch won't work, because the digital command riding on the voltage is filtered inside the 22K switch. Thus, the DISEqC switch receives a suppressed digital command with severe wave shape distortion. This problem has been bothering me because I want to use one 22K switch ahead of two DISEqC 4-way switches to select 8 feeds. I don't want to use one DISEqC switch with four 22K switches, too many components and cabling connections, and the heavier load on the DISEqC caused by the 22K switch always passing voltage. I solved the problem with the HUGHES switches by removing the 22K electrolytic cap, the only one on the circuit board The DC is passed from the input through a choke and filtered by that capacitor before exiting both output ports. The switch works fine without the filter capacitor, and the DISEqC switches work great on the two output ports.
This works for the HUGHES 22K switch, but it won't for other models of 22K switches that do switch DC at the output ports, such as the Gecen SW-03. As a side note, the SW-34 Dishnet switches do not pass commands, but do pass DC simultaneously on all three ports. I've never noticed a 'chattering' problem due to command interference in a 22K switch / DISEqC switch combo. The loss of control, in my tests, is due to a capacitance time constant effect that distorts the digital command waveform prior to being received by the DISEqC switch. It's getting to the point where I'm about to build my own switches. A custom 22K switch and two DISEqC, compatible and within one case.
I hope this helps to clear some of the confusion, and can make a more preferable situation possible, including 4X4 multi-switches, if someone has that goal in mind.
-sidha
The problem with 22K Hughes switches is that they won't pass clean DISEqC commands. DISEqC switches will pass enough of the 22K tone to switch a sensitive 22K switch. To select 8 feeds with four 22K switches, one on each of the four ports of a DISEqC switch, might not be a good idea.. The problem with that plan is the surface mount voltage switching transistors in the DISEqC switch can get very hot. This is because the voltage flowing through an active port is twice what it should be, due to the design of the Hughes 22K switches, that do not switch the DC voltage. They select the RF of either port on command, but the voltage is not switched. The good thing about that design is there are no sensitive voltage switching transistors to be damaged by shorts or surges, as in the more vulnerable and unforgiving DISEqC variety of switch.
Using one 22K switch by itself is not a problem, but connecting it after a DISEqC is a problem. I've burned out a port on at least one DISEqC switch because the transistors couldn't take the extra load of two lnbf's being powered at once. Installing the 22K switch ahead of the DISEqC switch won't work, because the digital command riding on the voltage is filtered inside the 22K switch. Thus, the DISEqC switch receives a suppressed digital command with severe wave shape distortion. This problem has been bothering me because I want to use one 22K switch ahead of two DISEqC 4-way switches to select 8 feeds. I don't want to use one DISEqC switch with four 22K switches, too many components and cabling connections, and the heavier load on the DISEqC caused by the 22K switch always passing voltage. I solved the problem with the HUGHES switches by removing the 22K electrolytic cap, the only one on the circuit board The DC is passed from the input through a choke and filtered by that capacitor before exiting both output ports. The switch works fine without the filter capacitor, and the DISEqC switches work great on the two output ports.
This works for the HUGHES 22K switch, but it won't for other models of 22K switches that do switch DC at the output ports, such as the Gecen SW-03. As a side note, the SW-34 Dishnet switches do not pass commands, but do pass DC simultaneously on all three ports. I've never noticed a 'chattering' problem due to command interference in a 22K switch / DISEqC switch combo. The loss of control, in my tests, is due to a capacitance time constant effect that distorts the digital command waveform prior to being received by the DISEqC switch. It's getting to the point where I'm about to build my own switches. A custom 22K switch and two DISEqC, compatible and within one case.
I hope this helps to clear some of the confusion, and can make a more preferable situation possible, including 4X4 multi-switches, if someone has that goal in mind.
-sidha
Last edited:
).
