Hi! I’m using the GigaBlox SFP compact 6‑port Ethernet switch with an FS.com SFP transceiver (model 39143). The fibre link with this transceiver works fine when tested on its own, but when I plug it into the GigaBlox switch the link never comes up and the SFP indicator LED stays off. At the moment I’m powering the GigaBlox SFP via the 5 V connector next to the SFP port and have one white PicoClasp (port 1) connected to my Cube Orange Plus via RJ45. Could you help me troubleshoot why the SFP isn’t working?
Thanks!
Hi there! Thanks for reaching out. To help diagnose this, could you tell me a bit more about your setup? How are you testing the SFP link, and what’s on the other side of the fibre (for example, another switch or media converter)? Are any of the white Molex PicoClasp connectors (copper ports) in use on the GigaBlox SFP? Also, how are you powering the GigaBlox SFP board? With these details we can narrow down the issue.
Thanks for your response. On the other end of the fibre I have another unmanaged SFP with a media converter using the same FS.com transceiver, and the link works fine when the GigaBlox switch isn’t in the loop. When I add the GigaBlox SFP, the link drops and the SFP indicator on the back of the switch doesn’t light up. I’m only using the SFP port and port 1 via the white PicoClasp connector to my Cube Orange Plus; the other copper ports are empty. The switch is currently powered from a PM02‑V3 module at around 5 V. Do you have any suggestions on what to check next?
Thanks for the details. First, please ensure the SFP module is fully seated in the socket – you should feel or hear a click when it locks in place. The GigaBlox SFP board draws about 1.56 W; at 12 V that is roughly 0.13 A but at 5 V it needs around 0.31 A. 5 V is right at the minimum input voltage, so if your wiring is long or thin you can easily see a drop below 4.5 V, which will prevent the SFP port from working.
As a test, try powering the board from a bench supply at 12 V with the current limit set to about 0.5 A and see if the link comes up and the LED lights. Measure the current draw and the voltage actually delivered at the power terminal. If it works at 12 V but not at 5 V, then either the 5 V supply doesn’t have enough headroom or there is too much voltage drop in the cables. Shorter, thicker wires or a higher supply voltage are recommended.
Let me know what current you see at 12 V and the voltage at the board, and we can go from there.
I powered the GigaBlox SFP from a 12 V bench supply with the current limit set to 0.5 A and connected both red wires to 12 V and both black wires to ground. With 12 V the SFP link comes up and the network works fine. The board draws about 0.13 A at 12 V (around 1.56 W). Does this mean using a 5 V input isn’t possible?
0.5 A was just the current limit on your bench supply; the board itself only draws about 0.13 A at 12 V, which equates to roughly 1.56 W. When powered from 5 V it would therefore need around 1.56 W / 5 V ≈ 0.31 A. Using a 5 V input is possible provided your power module can deliver that current and, importantly, keep the voltage at the GigaBlox SFP above about 4.5 V under load.
If there are other devices on the same 5 V bus or if the wiring is long/thin, you may see the voltage drop below this threshold. Please check what else is connected to the 5 V rail and add up their current draw. I also recommend measuring the actual voltage at the GigaBlox power terminal while it’s running. If it’s below 5 V (for example, 4.4 V) then that would explain why the SFP port doesn’t work. In that case, either reduce the voltage drop by using shorter/heavier gauge wires or supply a higher voltage (e.g. 12 V) and let the onboard regulator do the work.
Let me know what else is on the 5 V bus and what voltage you measure at the board.
The GigaBlox SFP is the only thing on the 5 V rail. There’s another Holybro PM on the same LiPo battery powering a different device, but it isn’t sharing the 5 V rail with the GigaBlox. The LiPo is 3 S (50 C, 2200 mAh – about 11.1 V).
I measured the voltage at the PM feeding the switch and found it’s around 4.4 V, which would explain why the SFP port wasn’t coming up.
I’ll fix the power wiring to make sure the switch actually sees at least 5 V — I might just move it to 12 V to be safe.The GigaBlox SFP is the only thing on the 5 V rail. There’s another Holybro PM on the same LiPo battery powering a different device, but it isn’t sharing the 5 V rail with the GigaBlox. The battery is a 3S LiPo (50 C, 2200 mAh), so about 11.1 V. After measuring I see the voltage at the PM feeding the switch is about 4.4 V, which would explain why the SFP port wasn’t coming up.
I’ll fix the power wiring to make sure the switch actually sees at least 5 V — I might just move it to 12 V to be safe.
That would explain it — 4.4 V is below the absolute minimum input voltage for the GigaBlox SFP. Even 4.5 V is cutting it very close because voltage drop in the wiring can pull it down further. Running the board on a 12 V input is much safer since the onboard regulator will provide a stable 5 V.
If you want to stick with a 5 V rail, make sure the supply can deliver at least about 0.31 A for the switch and use short, thick wires so the voltage at the board stays above roughly 4.5 V under load. Let me know if you have any more questions.