Sometime back I posted details on modifying the WNC SWA-48 4x8 switch to have all inputs wired for a 18V supply voltage for driving LNBs that don't think in terms of 18/13V and 22 KHz. At the time this switch had a lot going for it (1) it was already powered, (2) it had good headroom, (3) it could pass up to 2150 MHz (barely), (4) the case was easy to open and (5) it was going for nothing on eBay.
I ran into some bizarre problems the other day that ultimately turned out to be a bad tap, but along the way I decided the WNCs are not the best choice for bandstacked and universal LNBs. That will be fodder for another thread, because I measured a lot of switches and amps, resulting in some interesting data. But the bottom line was I went looking for another 4x8 switch that could really pass 950-2150 MHz in its sleep.
I finally decided to try the Zinwell WB68, even though it is powered by receiver(s) rather than a wart. The prices are now much lower than a few months ago and it is designed for DirecTV's Ka/Ku system that requires 250-750, 950-1450 and 1650-2150 MHz IF. The first one arrived yesterday and after some quick performance checks I was sold.
Even though the WB68 is a 6x8 switch, only four inputs are readily accessible to FTA receivers. The other two input are called 'Flex Ports' and supposedly require special codes to be sent for their selection. I can't imagine it would be that hard to figure this out, but I couldn't find anything on the web indicating someone has done it. In the meantime this makes the WB68 an effective 4x8 switch, using 18/13V and 0/22 KHz for selection purposes. That was the fly in the ointment for me because I want all the inputs to send +18V to my LNBs. So out came the weapons and open came the WB68 to see what could be done about this.
The WB68 innards are pretty easy to access. One peels the sealant off the back, leaving four screws holding on the backplate. With that off, you get to see one side of the circuit board. That's all I needed. Each input has a 16 µH choke used to feed the LNB voltage. I removed the chokes on the Flex Ports for other purposes, and desoldered the choke lead connected to the circuit board for the four inputs I plan to use. These inductors are hot glued to the circuit board, and it seemed reasonable to leave them in that state. I next lifted the center pin of the F-connector for one of the Flex Ports and isolated it from the circuit board with a sliver of electrical tape. Finally I ran a wire from this contact to each of the open sides of the four chokes. The idea was to connect a +18V supply to the rewired Flex Port and have it provide power to each of the four LNBs connected to the WB68.
I wasn't entirely happy about this mod before testing it. There are SMT bypass capacitors in addition to the chokes isolating the IF from the LNB from the WB68's power feeds. Finding capacitors with resonances in or above L-band that could fit this mod wasn't going to be that easy, so I measured the switch performance using the chokes only. On the spectrum analyzer and employing some spot checks with receivers convinced me the caps weren't necessary.
So is this mod worth doing? The alternative would be to use external power inserters, but that's more cabling and parts. I found this took a only a few minutes and was cleaner. For either external or internal power insertion, this takes most of the powering load off the receiver and ensures each LNB gets the optimal voltage for best performance.
For anyone interested, I've attached before and after photos.
I ran into some bizarre problems the other day that ultimately turned out to be a bad tap, but along the way I decided the WNCs are not the best choice for bandstacked and universal LNBs. That will be fodder for another thread, because I measured a lot of switches and amps, resulting in some interesting data. But the bottom line was I went looking for another 4x8 switch that could really pass 950-2150 MHz in its sleep.
I finally decided to try the Zinwell WB68, even though it is powered by receiver(s) rather than a wart. The prices are now much lower than a few months ago and it is designed for DirecTV's Ka/Ku system that requires 250-750, 950-1450 and 1650-2150 MHz IF. The first one arrived yesterday and after some quick performance checks I was sold.
Even though the WB68 is a 6x8 switch, only four inputs are readily accessible to FTA receivers. The other two input are called 'Flex Ports' and supposedly require special codes to be sent for their selection. I can't imagine it would be that hard to figure this out, but I couldn't find anything on the web indicating someone has done it. In the meantime this makes the WB68 an effective 4x8 switch, using 18/13V and 0/22 KHz for selection purposes. That was the fly in the ointment for me because I want all the inputs to send +18V to my LNBs. So out came the weapons and open came the WB68 to see what could be done about this.
The WB68 innards are pretty easy to access. One peels the sealant off the back, leaving four screws holding on the backplate. With that off, you get to see one side of the circuit board. That's all I needed. Each input has a 16 µH choke used to feed the LNB voltage. I removed the chokes on the Flex Ports for other purposes, and desoldered the choke lead connected to the circuit board for the four inputs I plan to use. These inductors are hot glued to the circuit board, and it seemed reasonable to leave them in that state. I next lifted the center pin of the F-connector for one of the Flex Ports and isolated it from the circuit board with a sliver of electrical tape. Finally I ran a wire from this contact to each of the open sides of the four chokes. The idea was to connect a +18V supply to the rewired Flex Port and have it provide power to each of the four LNBs connected to the WB68.
I wasn't entirely happy about this mod before testing it. There are SMT bypass capacitors in addition to the chokes isolating the IF from the LNB from the WB68's power feeds. Finding capacitors with resonances in or above L-band that could fit this mod wasn't going to be that easy, so I measured the switch performance using the chokes only. On the spectrum analyzer and employing some spot checks with receivers convinced me the caps weren't necessary.
So is this mod worth doing? The alternative would be to use external power inserters, but that's more cabling and parts. I found this took a only a few minutes and was cleaner. For either external or internal power insertion, this takes most of the powering load off the receiver and ensures each LNB gets the optimal voltage for best performance.
For anyone interested, I've attached before and after photos.
Attachments
Last edited: