TransimpedenceAmp

TableOfContents

0m5802 Fiber Demo board
Purchase AN98082: OM5804 Receiver demoboard for 155/622/1250 Mbps from China. http://translate.google.co.za/translate?hl=en&sl=zh-CN&u=http://www.chinadz.com/html/file/09w/3w/pdf_34560_file_AN98082%2520-%2520OM5804%2520Receiver%2520demoboard%2520for%2520155-622-1250%2520Mbps.htm&ei=tpPAStOJFqTSjAfP55xC&sa=X&oi=translate&resnum=27&ct=result&prev=/search%3Fq%3Dom5804%26hl%3Den%26lr%3D%26num%3D100

Replace CAT-5 cable with FSO links
Replace short range indoors 1Gig Ethernet links CAT-5 cable with a laser diode, FSO combination. The TZA3023 with TZA3044 as optical front end provides 1Gig, the tza3033 and tza3034 are used in Ronja with 155Mbit throughput. The TZA3023 with TZA3044 combination will be used for indoor applications with a laser diode used for 1Gig bandwidth instead of a LED. Indoors a laser diode doesn't have the same atmospheric effects as outdoors.

FDDI LED driver

 * http://www.onsemi.com/PowerSolutions/product.do?id=MC10SX1130
 * http://www.onsemi.com/pub_link/Collateral/MC10SX1130-D.PDF

900 - 1300nm range cam
http://www.edmundoptics.com/onlinecatalog/displayproduct.cfm?productid=1402

photodiode with preamp
Useful for 1gig links, they are to expensive for 10meg and 100meg.

http://www.edmundoptics.com/onlinecatalog/displayproduct.cfm?productID=2795 Silicon Photodiodes With Built in Preamplifier * Available in Standard TO-5 and TO-8 Packages * Low Noise * High Sensitivity

These detectors consist of a silicon photodiode and a preamplifier chip integrated in the same package. The standardized packages allow for easy integration into optical systems. Both High Speed and High Gain detectors are highly sensitive over a wide spectral range. The close proximity of amplifier and detector minimizes stray capacitances, increases response speed and eliminates external noise pickup. High Speed detector applications include optical fiber communication, video signal transmission and optical disk pick up. The High Gain detectors can be used in low-light level applications such as analytical and measurement equipment. A variable gain can be achieved on the High Gain detectors by externally connecting a resistor in parallel with the built-in 1GO feedback resistor. These detectors are compatible with our C-Mount photodiode mounts. A power supply is required to operate the detectors.

http://www.edmundoptics.com/onlinecatalog/displayproduct.cfm?productid=2513 diode detector with preamp in box

data acquisition (DAQ) system. Get link for hacking.

Encoding schemes
http://www.freepatentsonline.com/EP1209841.html Manchester NRZ for optics

France telecom patent
Provide FULL FpGa verilog code: http://www.patentstorm.us/patents/pdfs/patent_id/5687193.html from http://www.patentstorm.us/patents/5687193/description.html Applied Optics, vol. 31, No. 12, Apr. 1992, pp. 2001-2004, A. J. Al-Sammak et al., "Manchester Encoder-Decoder for Optical Data Communication Links"

RONJA input stage
Url ot this page: http://bit.ly/10Mi8h The RONJA input analogue electronics replace with a TZA3033 and TZA3034 Phillips Transimpedance and Post-amplifier for PECL logic levels to Manchester encoding circuits. The TZA3033 is used in fiber SONET projects, get hold of a SONET Fiber switch to get PCB and reverse engineer for Gerber files. Mod the design then for FSO.

Design with tza3033
SONET equipment. Get a board somewhere and do Reverse engineering.

Purchase AN98082: OM5804 Receiver demoboard for 155/622/1250 Mbps from China. http://translate.google.co.za/translate?hl=en&sl=zh-CN&u=http://www.chinadz.com/html/file/09w/3w/pdf_34560_file_AN98082%2520-%2520OM5804%2520Receiver%2520demoboard%2520for%2520155-622-1250%2520Mbps.htm&ei=tpPAStOJFqTSjAfP55xC&sa=X&oi=translate&resnum=27&ct=result&prev=/search%3Fq%3Dom5804%26hl%3Den%26lr%3D%26num%3D100

http://www.nalanda.nitc.ac.in/industry/datasheets/philips/index.html

http://www.nalanda.nitc.ac.in/industry/datasheets/philips/Sec09/AN1443.pdf Tutorial on fiber/FSO signal conditioning

http://forum.allaboutcircuits.com/newsgroups/viewtopic.php?t=9060

http://www.nalanda.nitc.ac.in/industry/datasheets/philips/Sec09/an98082.pdf diagrams of 155mbit interface

http://www.lazerlink.ru/1/sirco/components/TZA3033.pdf OM5802 demo board for laser driver and OM5804 receiver

* laser driver TZA3041

http://www.nalanda.nitc.ac.in/industry/datasheets/philips/index.html optical receivers

Ronja mailing list on tza3033
* http://fsoandre.blogspot.com/ * http://pointless.net/pipermail/ronja/2008-February.txt * Ml6652, original design ( using PCB), IC?s from Maxim and Philips... * Ml4664, tza3033, tza3034, max 3263 max 3264, max3657, max3645 * MAX3766 replaces MAX3263

Max3657
* http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=MAX3657ETC%2B-ND $2.2.unit * http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3927 The MAX3657 is a transimpedance preamplifier for receivers operating up to 155Mbps. The low noise, high gain, and low-power dissipation make it ideal for Class-B and Class-C passive optical networks (PONs).

The phillips TZA3033 http://www.datasheetcatalog.com/datasheets_pdf/T/Z/A/3/TZA3033.shtml is an alternative to the discontinued Max3657.

http://www.icminer.com/datasheetsearch.php?search=CPHILIPS3209-TL replacement chips

buy tza3033
* http://www.digchip.com/datasheets/parts/datasheet/364/TZA3033T.php * http://www.sierraic.com/pnresults.asp?navpart=TZA3033

patent on TZA3033
http://www.freepatentsonline.com/y2002/0027689.html

cost
http://findarticles.com/p/articles/mi_m0EIN/is_2000_March_8/ai_60000355/ Samples of all products are available now and TZA3023 and TZA3043 are also in production. Volume production of the TZA3033 and TZA3013 will commence in second quarter 2000. Prices for the TZA3013, TZA3023, TZA3033 and TZA3043 are $16.50, $3.85, $2.65 and $3.50 for 10,000 pieces respectively.

Technical reports
* CERN-LHCC-96-42 * http://epublicns03.esc.rl.ac.uk/search?st=browse-by-orgunit&ou=316&mpp=20&so=ta&f=1 optics * http://home.saske.sk/~bruncko/img/scpaper/DBpreprints.html

phillips pre post amplifier
TZA3033

TZA3034

Transceiver
TZA3031 http://www.datasheetcatalog.com/datasheets_pdf/T/Z/A/3/TZA3031.shtml Transceiver. The TZA3005H SDH/SONET transceiver chip is a fully integrated serialization/deserialization STM1/OC3 (155.52 Mbits/s) and STM4/OC12 (622.08 Mbits/s) interface device. In conjunction with the data and clock recovery unit (TZA3004), optical front-end (TZA3023 with TZA3034/44)and a laser driver (TZA3001). A typical network application

Design with TZA3033
* http://content1.docstoc.com/flash/Tangs_Interface_paper.swf shows tza3033 implementation * http://hep.uchicago.edu/atlas/tilecal/interface.php Full drawings documentation available * http://hep.uchicago.edu/atlas/tilecal/notes/Tangs_Interface_paper.pdf Image better * http://www.lasermate.com RST-M13A306

Tangs board
VIIII. REFERENCES [1] Tile Calorimeter Technical Design Report, CERN/LHCC 96-42.

[2] http://ttc.web.cern.ch/TTC/intro.html

[3] Marcel J.M. Geurts OM5804 Receiver demoboard for 155/622/1250 Mbps

http://www.nikhef.nl/~n05/antares/Design_Docs/AN98082_Ph.pdf

Data Rate http://www-ppd.fnal.gov/tshaw.myweb/CMS/Standards/low_power_glink.pdf

[4] "Receiver ASIC for Timing, Trigger and Control Distribution in LHC Experiments", J. Christiansen, A. Marchioro, P. Moreira and A. Sancho, IEEE Trans. Nuclear Science, Vol. 43, June 1996, pp. 1773-1777

sdfv
http://en.wikipedia.org/wiki/Positive_emitter-coupled_logic * RST-M13A306 optical input -> noise filter -> TZA3033 -> noise filter -> TZA3034 ->LV-PECL to LVDS * http://search.datasheetcatalog.net/key/TZA3034 Post Amplifier * http://lhc-electronics-workshop.web.cern.ch/LHC-electronics-workshop/2002/DAQ/B31.pdf * Tile Calorimeter Technical Design Report, CERN/LHCC 96-42 * http://www-ppd.fnal.gov/tshaw.myweb/CMS/Standards/low_power_glink.pdf IC optic transmitter receiver * http://hep.uchicago.edu/atlas/tilecal/notes/Tangs_Interface_paper.pdf * http://hep.uchicago.edu/atlas/tilecal/interface/2468v33asm_list_2.pdf Assembly list

Fermionics
* http://www.fermionics.com/HS_Opt_Recvrs.htm Optical receiver * http://www.fermionics.com/

http://www.fermionics.com/opticalreceiver.html

MAXIM 3864E/D pre-amplifier coupled to PIN diode. Hack and reverse engineer the circuit for PCB layout.

Fermionics FD100 photodiodes are connected to a plurality of transimpedance amplifiers Maxim MAX 3657 All the plus outputs of all TIAs are connected together... Idea from the Optics patent page on scratchpad.

http://www.google.co.za/search?hl=en&lr=&num=100&ei=ij5YSvWVDeWMjAfRjLEb&sa=X&oi=spell&resnum=0&ct=result&cd=1&q=10+Mbps+%22Low-Noise+Transimpedance+Amplifier%22&spell=1

RonjaFpga, GmaxwellRonja

PIN-FET receiver
* http://www.wtd.com.cn/en/index.asp * 8-280 MB/s transimpedence amplifiers. * PFWM931 part number. 8meg transmipedence laser based won't work for FSO * IC with AGC. http://www.wtd.com.cn/en/UploadFile/PFWM93.pdf * http://www.wtd.com.cn/en/UploadFile/RTXM119.pdf uses the

20meg IC receiver
* http://www.imec.be/esscirc/ESSCIRC2002/presentations/Slides/C27.03.pdf * http://www.optoelettronica.com/prodottien.htm

companies
http://www.iptronics.com/ Creates http://en.wikipedia.org/wiki/Optical_interconnect

100meg Tx
jdb at lartmaker.nl

http://www.pointless.net/pipermail/ronja/2005-September/007454.html I have a prototype PCB for a 100MHz RX/TX unit but simply have had no time to test/tweak/enhance it. LED fall time is the issue so far; I know what to do about it, but (payed) work takes precedence over a GPLed project.

> > [just finished a 5ns IR pulse detector with associated 25W pulse >> laser driver for Work, looking into using a similar design for a >>  50..100Mbit/s Open Hardware optical link] > >That sounds great.

The work project was fun. The design can send and receive pulses over several hundred meters without any lenses. Of course, this uses the abovementioned 25W pulse laser ;-) Driving that one was...interesting, seeing that I needed to switch 30Amps with edge rates of a few nanoseconds. Inductance really sucks.

>Tell us more about it.

I intend to use the optical link for a local volunteer mesh-networking project. The plan is to have it rate-limited by the SNR of the link. Rough sketch:

* Tx: 100Base-T->PHY->FIFO->FEC->Luxeon/laser * Rx: PIN->Cascode amp->Detector->Sync/FEC->FIFO->PHY->100Base-T * Tx: 100Base-T->PHY->FIFO-> http://en.wikipedia.org/wiki/Forward_error_correction ->Luxeon/laser (TransimpedenceAmp) * Rx: PIN->Cascode amp->Detector->Sync/ http://en.wikipedia.org/wiki/Forward_error_correction ->FIFO->PHY->100Base-T

The FEC is an error correcting block code implemented in a CPLD and a ROM. This FEC is also responsible for creating state transitions on the link (to avoid baseline wander). The SNR determines the transmission of the FEC. EthernetTransceiver implements all this in a single IC commercially available from many vendors.

Target is to have all components SMD, but as solderable as possible (ie no BGA or finer pitch than necessary), and available from international distributors such as DigiKey. Schematics and PCB will be available as PS/PDF/Gerber, and ABEL/VHDL for the CPLD. I won't be using gschem/PCB, as I described in the other thread, sorry (although if I have enough time I may redo the schematics in gschem). As with all volunteer projects, this may take a while to materialize. Work is especially busy nowadays, so don't hold your breath.

mail 1 ronja
http://translate.google.co.za/translate?hl=en&sl=cs&u=http://www.pointless.net/pipermail/ronja/2003-December/001115.html&ei=U5ZsSqKjFoahjAemx9CZCw&sa=X&oi=translate&resnum=9&ct=result&prev=/search%3Fq%3DTZA3033%2Bsite:.pointless.net/pipermail/ronja%26hl%3Den%26lr%3D%26num%3D100

Where MAX 3964? Kde MAX3263? Where MAX3263?

mail 2 ronja
http://pointless.net/pipermail/ronja/2008-February.txt > I am working on a 100 mbps design based on ML6652 and LED driver is > MAX3263 and Reciver uses philips SA5211 as transimpedance Amplifier > .Will be releasing first shematics in a day or two. I am planning to > use VCSEL laser to begin with.

mail3 ronja
>    5. He already has a prototype MTL3 reciever at http://home.arcor.de/positiveelectron/files/MLT-3-receiver.png >    > >     > Meanwhile CD930 shows us he already has a 100 Mbps version, but >    alas the in >     > the clutches of proprietary license >    >     He has videos at >     >     http://www.youtube.com/watch?v=9h3hDm32kv4 >    >      http://www.youtube.com/watch?v=YcjaJhMHcNk >    > >     > and even better people like Jim Thompson are waiting to be >     convinced to >     > finance a 100 Mbps version.

I offer it for experiments to you. It's FCI-HR008 >    (http://www.datasheetarchive.com/pdf/1382205.pdf). It has capacitance >    0.8pF at 5V! Price: 1 000CZK (40?) per piece.

mail4 User cd930 HAS 100Meg design
CD930 napsal(a):       Sorry...no, maybe later. But...TX,RXis >    MAXIM chipset, >    mediaconvertor http://www.MARVELL.com 1108  (beforeused Micro-Linear), laser is >     870nm/10mW and >    photodiode is PDB-C102. >    Tested (tx lens 11cm, rx fresnellens 20cm) on 2500m. RSSI go to AD >    and Atmel >    CPU (LCD and RS232c). >    M.

Please consider the MAX3766 instead of the MAX3263, because the MAX3263 has reached end of life.

From: cd930 at centrum.cz (CD930) Sorry, this development was for one firm from our city. I must not publish circuit PCB :( But....some ideas i am here publish along the years. In future looking on 1Gpbs..... avalanche photodiode with 200V dc :)))  r

mail5 Ronja User: pkyaduvanshi
Alternative 10Meg transmitter circuit using three 78hc245 http://luxeonstar.blogspot.com/

ML6652 Ethernet 100Meg fiber coppper converter

mail6 twighbright
> > first is you would have to change the encoding of the MLT3 signal to 2-level signal which is suited for optical transmission. in easiest case manchester code.

I have discovered a bi-level DC-nulled modulation that has better efficiency than the formerly patented IBM 8B10B - http://en.wikipedia.org/wiki/8b/10b_encoding, which has only 80%. Namely 90% exactly. And doesn't have excessive requirements on the codec. Maximum run length of identical bits is 10 bits. That means 1Gbps optical Ethernet could be upgraded to 1.125Gbps probably without a change in the PHY or Ronja Manchester 10Mbps could be upgraded to 18Mbps without a change in the PHY. The same for all the other numerous technologies that use 8B10B, see the Wikipedia article on 8B10B. Does anything like this already exist? If not, I could try to make a contract with someonethat he can patent it, sell the patent to proprietary designs only, give irrevocable rights to all open source designs, and pay a fulltime salary to me for the rest of my life. Then I could go developing Ronja :) Otherwise I can write an implementation and publish it to constitute a prior art which makes sure it's unpatentable ;-)

mail7 Russian site translate
Look this switch based on RTL chipset   http://lan.neomontana-bg.com/CleverSwitch.php

http://translate.google.co.za/translate?hl=en&sl=bg&u=http://lan.neomontana-bg.com/CleverSwitch.php&ei=yqJsSrC8OtmMjAfT_pmRCw&sa=X&oi=translate&resnum=1&ct=result&prev=/search%3Fq%3Dhttp://lan.neomontana-bg.com/CleverSwitch.php%26hl%3Den%26lr%3D%26num%3D100

This switch have bandwith limit on each port (upload/download) 128Kbit, 256Kbit, 1MBit, 2Mbit, 4MBit, 8M

Price is only 13$

It is possibly RONJA 10MBit  work in 20 or 40 Mbit with this swich ?

SFH 495 as alternate IR transmitter? ent there looking for the Hsdl-4230. By the way Clock advised me that it might be a laser and will require a different drive circuit, any idea, how we might proceed with that.... and i wasn't able to find any safety Class specification of this led in the datasheet... anybody know how we might do that?

I'd like your opinion about the possibility of using a SFH 495 IR diode. http://pdf1.alldatasheet.com/datasheet-pdf/view/45680/SIEMENS/SFH495P.html It has a rise time of 7ns and a half angle of 30 degrees at 2.1 volts. Please let me know if this may be used as an alternate to HSDL-4230.

Sounds like it's yet another OpenRRCP compatible switch ASIC, and there's plenty of those around: http://openrrcp.org.ru/hardware

I'm wary of RRCP -- it seems that it's completely disabled on most new hardware, and even when it hasn't been fused out, soldering pullups to TQFP packages is a pain in the behind.

For more practical FSO-ISP use, needing another external microcontroller to do RRCP to SNMP in such a design, (for it *is* the standard), is also a pain in the behind, leaving aside the consideration of the time cost for writing software to do that.

Anyway, to get back to what clock said.

This is a store-and-forward switch ASIC. Realtek sheets state clearly the bandwidth throttling uses 802.3x flow control as its throttling mechanism rather than early drop.

802.3x PAUSE should "just work" over a RONJA link, but of course it assumes the guys at the other end of the link are in your Ethernet broadcast domain -- which might be too much network intimacy for some individuals.,

I believe this capability however is independent of its ability to generate 10MHz link beat, which is what RONJA requires, and this is what clock was getting at.

links
FreeSpaceOptics, FiberSwitches