If you need assistance, please send an email to forum at 4hv dot org. To ensure your email is not marked as spam, please include the phrase "4hv help" in the subject line. You can also find assistance via IRC, at irc.shadowworld.net, room #hvcomm.
Support 4hv.org!
Donate:
4hv.org is hosted on a dedicated server. Unfortunately, this server costs and we rely on the help of site members to keep 4hv.org running. Please consider donating. We will place your name on the thanks list and you'll be helping to keep 4hv.org alive and free for everyone. Members whose names appear in red bold have donated recently. Green bold denotes those who have recently donated to keep the server carbon neutral.
Special Thanks To:
Aaron Holmes
Aaron Wheeler
Adam Horden
Alan Scrimgeour
Andre
Andrew Haynes
Anonymous000
asabase
Austin Weil
barney
Barry
Bert Hickman
Bill Kukowski
Blitzorn
Brandon Paradelas
Bruce Bowling
BubeeMike
Byong Park
Cesiumsponge
Chris F.
Chris Hooper
Corey Worthington
Derek Woodroffe
Dalus
Dan Strother
Daniel Davis
Daniel Uhrenholt
datasheetarchive
Dave Billington
Dave Marshall
David F.
Dennis Rogers
drelectrix
Dr. John Gudenas
Dr. Spark
E.TexasTesla
eastvoltresearch
Eirik Taylor
Erik Dyakov
Erlend^SE
Finn Hammer
Firebug24k
GalliumMan
Gary Peterson
George Slade
GhostNull
Gordon Mcknight
Graham Armitage
Grant
GreySoul
Henry H
IamSmooth
In memory of Leo Powning
Jacob Cash
James Howells
James Pawson
Jeff Greenfield
Jeff Thomas
Jesse Frost
Jim Mitchell
jlr134
Joe Mastroianni
John Forcina
John Oberg
John Willcutt
Jon Newcomb
klugesmith
Leslie Wright
Lutz Hoffman
Mads Barnkob
Martin King
Mats Karlsson
Matt Gibson
Matthew Guidry
mbd
Michael D'Angelo
Mikkel
mileswaldron
mister_rf
Neil Foster
Nick de Smith
Nick Soroka
nicklenorp
Nik
Norman Stanley
Patrick Coleman
Paul Brodie
Paul Jordan
Paul Montgomery
Ped
Peter Krogen
Peter Terren
PhilGood
Richard Feldman
Robert Bush
Royce Bailey
Scott Fusare
Scott Newman
smiffy
Stella
Steven Busic
Steve Conner
Steve Jones
Steve Ward
Sulaiman
Thomas Coyle
Thomas A. Wallace
Thomas W
Timo
Torch
Ulf Jonsson
vasil
Vaxian
vladi mazzilli
wastehl
Weston
William Kim
William N.
William Stehl
Wesley Venis
The aforementioned have contributed financially to the continuing triumph of 4hv.org. They are deserving of my most heartfelt thanks.
Registered Member #3324
Joined: Sun Oct 17 2010, 06:57PM
Location:
Posts: 1276
Steve Conner wrote ...
The SFP transceivers need a LVDS input. Not much use without something like a FPGA or the Dallas chips I mentioned.
You can get the SFP "cages" from Digikey.
Oh cool, i couldn't see that, thanks for the link..
I doubt i will get into that quite yet... but its cirtaintly interesting. Always wanted to get into this kind of stuff, the big issue is you cant just prototype it onto a breadboard or anything, Aha.
Do you have any good reccomendations for learning about this kind of high speed digital?
Registered Member #3324
Joined: Sun Oct 17 2010, 06:57PM
Location:
Posts: 1276
Steve Conner wrote ...
This book is pure gold.
Thanks, I shall go buy that book when I can (got to find £40-80 first it seems)
How hard is it generally to work with such things as these? calculating lengths of wires and complex mathmatics or is it just a matter of knowing the tips and tricks? This stuff does really interest me in general.
Registered Member #30656
Joined: Tue Jul 30 2013, 02:40AM
Location: UK
Posts: 208
If you're looking at using SFP modules then be careful of minimum clock rates when using a SERDES - some SFP+ modules I have specify a minimum of a ~10GBps, though they seem to run okay at 1.25GBps (Gigabit Ethernet). Fast LVDS I/Os from a FPGA really make sense here - if you're going to the effort of a 4 layer board anyway then you may as well do it properly and put one on, lets you ditch the 74/4000 series logic in the driver too.
That said, you may _just_ be able to get away with a 2 layer board using those SERDES chips Steve mentioned - they would need to be placed very close to the SFP connector, with a good ground plane and big fat traces to get as close to 50 ohms (100 differential) as you can on a 1.6mm PCB. Absolutely no guarantees though, and routing could be rather difficult.
As for the cages, you also need a surface mount board edge connector that goes inside them too - a digikey search for SFP should find them (super cheap too!). These are designed to be reflowed before the cages are put on and wave soldered, and are easily DIYable.
Registered Member #3324
Joined: Sun Oct 17 2010, 06:57PM
Location:
Posts: 1276
Hydron wrote ...
If you're looking at using SFP modules then be careful of minimum clock rates when using a SERDES - some SFP+ modules I have specify a minimum of a ~10GBps, though they seem to run okay at 1.25GBps (Gigabit Ethernet). Fast LVDS I/Os from a FPGA really make sense here - if you're going to the effort of a 4 layer board anyway then you may as well do it properly and put one on, lets you ditch the 74/4000 series logic in the driver too.
That said, you may _just_ be able to get away with a 2 layer board using those SERDES chips Steve mentioned - they would need to be placed very close to the SFP connector, with a good ground plane and big fat traces to get as close to 50 ohms (100 differential) as you can on a 1.6mm PCB. Absolutely no guarantees though, and routing could be rather difficult.
As for the cages, you also need a surface mount board edge connector that goes inside them too - a digikey search for SFP should find them (super cheap too!). These are designed to be reflowed before the cages are put on and wave soldered, and are easily DIYable.
Yeh, i noticed that.
I have no real experiance with HF stuff however i get some of the general ideas from reading a book called Electromagnetics Explained, which did mention a lot of this stuff.
This is the one ive been looking at, and i believe Steve may have too.
From the Datasheet here:
I believe it says you need a minimum of 1.0625Gb/s datarate... I am curious about how you go about designing this kind of thing really.
Registered Member #30
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706
I wouldn't even bother trying less than a 4 layer board for a project like this. I would probably build a module with the SFP cage and two serdes chips, make a batch of the things and sell the extra to recoup my costs.
SFPs come in 1Gb and 10Gb flavours. The 1Gb ones are available cheaply used on Ebay as people upgrade to 10G Ethernet.
The Dallas serdes chips should be capable of cranking out data fast enough to keep a 1Gb transceiver happy. I know someone who tried this and by all accounts it worked fine.
The biggest math in high-speed digital design is working out characteristic impedances of PCB traces.
Registered Member #3324
Joined: Sun Oct 17 2010, 06:57PM
Location:
Posts: 1276
Steve Conner wrote ...
I wouldn't even bother trying less than a 4 layer board for a project like this. I would probably build a module with the SFP cage and two serdes chips, make a batch of the things and sell the extra to recoup my costs.
SFPs come in 1Gb and 10Gb flavours. The 1Gb ones are available cheaply used on Ebay as people upgrade to 10G Ethernet.
The Dallas serdes chips should be capable of cranking out data fast enough to keep a 1Gb transceiver happy. I know someone who tried this and by all accounts it worked fine.
The biggest math in high-speed digital design is working out characteristic impedances of PCB traces.
Sounds like a plan! how much would you likely sell these boards for? i could be interested. Or at least in the schematics / artwork.
Registered Member #30656
Joined: Tue Jul 30 2013, 02:40AM
Location: UK
Posts: 208
Steve Conner wrote ...
I wouldn't even bother trying less than a 4 layer board for a project like this. I would probably build a module with the SFP cage and two serdes chips, make a batch of the things and sell the extra to recoup my costs.
SFPs come in 1Gb and 10Gb flavours. The 1Gb ones are available cheaply used on Ebay as people upgrade to 10G Ethernet.
The Dallas serdes chips should be capable of cranking out data fast enough to keep a 1Gb transceiver happy. I know someone who tried this and by all accounts it worked fine.
The biggest math in high-speed digital design is working out characteristic impedances of PCB traces.
Yeah, I'd personally stay clear of 2 layer for anything like this, just saying it may just be possible for masochists.
The SPF modules I have are of the 10GBit SFP+ variety, but you're right about surplus 1GBit ones - a friend of mine working in a datacentre has stacks of them from upgrades. Was mainly noting that the speed was something to keep in mind - one of those SERDES chips running full tilt will be easily fast enough, but don't try run it at minimum speed.
For PCB impedance calculations and more I'd suggest trying this: , it's rather handy!
Registered Member #3324
Joined: Sun Oct 17 2010, 06:57PM
Location:
Posts: 1276
Hydron wrote ...
Steve Conner wrote ...
I wouldn't even bother trying less than a 4 layer board for a project like this. I would probably build a module with the SFP cage and two serdes chips, make a batch of the things and sell the extra to recoup my costs.
SFPs come in 1Gb and 10Gb flavours. The 1Gb ones are available cheaply used on Ebay as people upgrade to 10G Ethernet.
The Dallas serdes chips should be capable of cranking out data fast enough to keep a 1Gb transceiver happy. I know someone who tried this and by all accounts it worked fine.
The biggest math in high-speed digital design is working out characteristic impedances of PCB traces.
Yeah, I'd personally stay clear of 2 layer for anything like this, just saying it may just be possible for masochists.
The SPF modules I have are of the 10GBit SFP+ variety, but you're right about surplus 1GBit ones - a friend of mine working in a datacentre has stacks of them from upgrades. Was mainly noting that the speed was something to keep in mind - one of those SERDES chips running full tilt will be easily fast enough, but don't try run it at minimum speed.
For PCB impedance calculations and more I'd suggest trying this: , it's rather handy!
Looking at those chips and all, do you think there would be some cheaper SERDES chips? those ones are around £10 each, and you need 2 per board and likely 2 boards meaning 4 chips / £40, kind of expensive. Just wondering if there are any cheaper.... or could you get a cheapish FPGA to run somthing similar to both chips, at the risk of added complexity. it could have an advantage in that you would be able to add more features.
Registered Member #30
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706
I don't know of any FPGA cheaper than about £10. Certainly none with the required hardware to pump out 1Gb/s serial data streams. You need something like a Spartan-6 before you get that.
One possible cost saving of a FPGA would be to use the same FPGA as both the fibre optic receiver and Tesla coil driver.
One way to look at this is that two second-hand gigabit SFPs for say £10 each, plus four serdes chips at £10 each, plus a duplex patch cord (another £10) give you 24 channels of digital IO in both directions for a total cost of £70. About £3 per channel. So if you actually need all those channels, this is a cost-effective way to get them. The traditional fibre link used with Tesla coils costs about £40 and delivers one channel.
The sampling rate on the digital IO is fast enough (25MHz IIRC) that you could use each digital IO line for almost anything. Interrupter pulses, MIDI, RS232, SPI, I2C, PWM or whatever.
If that is too expensive for you then try the Teslink project described here. Fewer channels but cheaper, assuming you already have a PIC programmer or are prepared to make your own JDM-style one. No fancy 4 layer board is needed, all the components are available in DIP packages, so you could probably build it on stripboard.
This site is powered by e107, which is released under the GNU GPL License. All work on this site, except where otherwise noted, is licensed under a Creative Commons Attribution-ShareAlike 2.5 License. By submitting any information to this site, you agree that anything submitted will be so licensed. Please read our Disclaimer and Policies page for information on your rights and responsibilities regarding this site.
"Teslink" multiplexed fibre optic link
