Notices by F4GRX Sébastien (f4grx@chaos.social)

@benjojo @jaseg @azonenberg amazing project with amazing testing opportunities. I hope you manage to test the transatlantic toslink. For dialup you could just attempt SLIP over 430800 bauds uart. And cant wait to test this.

@marshray yeah. no reason not to open source that. I will put it somewhere. If you have ideas I can add them. the size is planned for a Farnell Multicomp MCREAS80 box https://fr.farnell.com/multicomp/mcreas80/case-aluminium-80x63-5x30mm/dp/1549965

hello @azonenberg random crazy question I think you know about, could a normal cheap 1G/100M SFP easily eat a 10 MHz signal source for fiber distribution of a clock without too much work? Like, I dont know, a LVDS driver chip right on the signal pair of the SFP? Would that work for reception too? with the SFP pair to a LVDS receiver feeding more CMOS circuits? I have no pretention to keep compatibility with anything ethernet.

the challenge is the slow speed here I guess. thanks!

@azonenberg yep, I'll attempt to do that, thanks for confirming it's worth trying!

@azonenberg I think I have a few old 850nm SFPs that were used for fiber channel so it's 4 Gbps max IIRC. I would not spend money on a pair of 10G SFP for such a hack lol

edit: ah I did not think about AGC. hmm.

edit2: also retimers will be a problem if they do clock recovery. We'll see. a breakout does not look too hard to build.

I think i found the one-chip-does-all: https://eu.mouser.com/datasheet/2/268/SY100EPT28L_3_point_3V_LVTTL_to_Diff_LVPECL_and_Di-1891475.pdf

@jaseg @azonenberg (oh hi, @benjojo )

@philpem @azonenberg @benjojo @jaseg @gsuberland @hennichodernich I've done a basic circuit, power not routed yet and I could add a ft232r (max speed 3mbps). It's my first differential circuit, so I would appreciate if someone could throw an eye on the differential signal area? would that work? caps should be integrated in the SFP so I did not add any. The receiver is MC100LVELT23 (at farnell) , seems to have integrated pull ups/down, so I only added a diff termination. Transmitter MC100LVELT22.

@marshray yep ofc I need decoupling on the transceivers.

power will come either from USB or the XT30 to a LF33 LDO.

Rout done

unused gates: are they really useful? OK, I can add test points.

other SFP would be hard to install, that would need more SMAs.

We can make a dual channel version on a 80x100 pcb later.

I also have plans for a kind of optical clock distribution with several outputs to serve different devices from a single clock signal.

@marshray Hello, I have not done all changes yet, but I have reorganized the board to keep all interesting circuits on the same side of the box. This can allow for having several modules on the same front plate. Decoupling is next. I still dont love the long VCC line to the sfp cage, IDK how "legal" it is to have a vcc line under the differential pairs.

Thanks @altomare for this very interesting antique oscillator. It's massive! It's from 1971, made in France by "Quartz et Electronique" a dead company that was acquired by Alcatel.

@altomare the insides are quite complex, two transistors heat the box that I understand as an outside oven. Inside there are circuits, includind two transformer-like units, and a big cylindrical plastic box that should be the inner oven. Interestingly the ground seems connected to the +24v line. Full reverse engineering and restoration will happen!

If it was used for a long time, this crystal should have aged like good champaign and will be quite stable.

edit: alt incoming - done

@altomare a few more pics while disassembling for cleanup.

@altomare reversing reversing. No idea what this crap does.

@altomare Calling all EE engineers, specifically ones that were designing stuff in the seventies. Does this look like a typical heat regulation circuit that would work? I suspect this is the power part, and the other PCB is a kind of oscillator to drive this. it has 3 transistors and I have not looked at it yet.

I'm not posting a bitmap as it's unreadable

I am working on page 3 of this PDF.

I imagined what a transformer could be but the second one is a mystery.

https://cloud.f4grx.net/index.php/s/JtSfTx4rmwZfS7W

@altomare starting analysis of the second board. As suspected, the coax shield is connected to the positive supply, so this oscillator is actually powered by -24VDC to ground... Or, output signal is referenced to +24V. Weird.

@discatte YIKES

This rust does not look very safe to me.

@discatte shin ramen or neoguri

@gorplop @TotalSonic the scroll wheel of the one I have at dayjob was indeed full of crap and hard to clean. It lasted long enough that i had to replace the microswitches twice.

@TotalSonic get a logitech trackball, you'll have to do the same in 2025.

@gorplop I see that, pretty hardware.