Post by thread:[EE]: Opto-isolated Ethernet connection

Sorry to trouble you with this OT, but I feel there is an answer "Out There" that I have failed to find. I have several computer applications that are both connected to a patient (basically) and a network for storing data on a server. For patient safety reasons the PC is isolated (including measuring electronics), but the system is not isolated unless the ethernet connection is run through a fiber - or something. The optofiber interfaces we use are too bulky, and inteded to run optical fibers for a substancial length. Not a 2 meter fibre. My question to you is, is there a j45 (or-whatever-they-are-called)- connector with a optical isolator(/-coupler) in it? It really shouldn't have to more than a 1 " cube. OR a network-card, that has this built in (4kV isolation preffered). OR is this a neat PIC-project to put in que ? (SEE it isn't off topic!!). Opto for data, and a "high" frequency transformer for powering the isolated side would be what is in that cube.... Sven G Milton Milton Medicinteknik KB Herrestad Åkersberg 590 47 Vikingstad SWEDEN tel +46 13 83 202 fax +46 13 83 202 http://w1.138.telia.com/~u13802667 -- http://www.piclist.com hint: PICList Posts must start with ONE topic: "[PIC]:" PIC only "[EE]:" engineering "[OT]:" off topic "[AD]:" ad's

>My question to you is, is there a j45 (or-whatever-they-are-called)- connector with a optical isolator(/-coupler) in it? It really shouldn't have to more than a 1 " cube. >OR a network-card, that has this built in (4kV isolation preffered). >OR is this a neat PIC-project to put in que ? (SEE it isn't off topic!!). Opto for data, and a "high" frequency transformer for powering the isolated side would be what is in that cube.... It wouldn't be easier to use a simple in-line transformer? Although there are optocouplers that can be used. That ones in the Philips (magnavox?) TV Sets, who isolate video inputs (>5MHz) from the main chassis --------------8<-------Corte aqui-------8<-------------- All the best!!! Alexandre Souza spam_OUTxandinhoTakeThisOuTinterlink.com.br --------------8<-------Corte aqui-------8<-------------- -- http://www.piclist.com hint: PICList Posts must start with ONE topic: "[PIC]:" PIC only "[EE]:" engineering "[OT]:" off topic "[AD]:" ad's

>It wouldn't be easier to use a simple in-line transformer? >It wouldn't be easier to use a simple in-line transformer? >Although there are optocouplers that can be used. That ones >in the Philips (magnavox?) TV Sets, who isolate video >inputs (>5MHz) from the main chassis This would be a dangerous thing to do, as they will not meet medical requirements for ground isolation leakage currents. These are very stringent, as the electrodes have an absolutely excellent contact with the body, so any microamps of earth leakage go directly where you do not want it. If transformers were sufficient, then using Ethernet would be sufficient in itself as every connection to the coax is through a transformer. I have not dug inside base-T interfaces, but I think you will find the same here to stop earth current problems between two pieces of equipment which can be some distance apart. The easiest way to solve the problem may be to use optical Ethernet cards, but I do not know if you can get one for a laptop. Another alternative may be to use a serial link with optocouplers between the laptop and a PC, running PPP so it is directly on the network, if that is what is required. The PC would not need to be anything specially fast as it would be used as a router - could even be running a floppy version of Linux to do this. -- http://www.piclist.com hint: PICList Posts must start with ONE topic: "[PIC]:" PIC only "[EE]:" engineering "[OT]:" off topic "[AD]:" ad's

Best Peter !, (others answered privately) Thank you too, for answering my post ! This is interesting. You are the third indicating a transformer. I don't have any knowledge what the signals actually look like on an ethernet wire. To me it's a plug with bidirectional data at an alarming rate, both ways, at an unknown voltage. Is it just "spikes" pos. and neg. ?. It would be great not to have to supply the uninsulated side of a optical solution. The transformer sounds too simple to be true (or?). The leakage would easily be under control, and, should the in/output circuitry accept it, the ground wire not necessary (I doubt that "just" the transformer would be sufficinet for this). Further comments much appreciated. By mistake we "double-grounded" a system consisting of 3 computers. Ground current induced in this case was 80 mA. Enough to smoke a small signal cable - yes (I do feel sorry for you...). The magnetic field would be another problem in these applications. Finally: a suggestion for a "ethernet connectivity supplier" ? Kindest Regards Sven in Sweden -----Ursprungligt meddelande----- Från: Peter L. Peres <.....plpKILLspam@spam@ACTCOM.CO.IL> Till: PICLISTKILLspamMITVMA.MIT.EDU <.....PICLISTKILLspam.....MITVMA.MIT.EDU> Datum: den 20 september 2000 20:23 Ämne: SV: Re: [EE]: Opto-isolated Ethernet connection {Quote hidden}>Ethernet is inherently transformer coupled at some stage or other. There >must be transformers that satisfy your insulation requirements. They may >even be cubical and 1" ;-) I would look at ethernet connectivity supplier >company's catalogs. There are also hubs that insulate. > >Most people do not realize that one just does not wire all the grounds of >the equipment in a large building together without asking for troubles >(hint: ground current problems. bad ones). Just think about why the phone >company is so pesky about *its* wires not touching anything. They have >been there and learned... > >Trust me on this, there is nothing like a signal cable smoked by ground >current laid in a nice freshly built newly painted and decorated ceiling. >With no spare pair. 2 days before opening. > >Peter > >-- >http://www.piclist.com hint: The list server can filter out subtopics >(like ads or off topics) for you. See http://www.piclist.com/#topics > > -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

>The transformer sounds too simple to be true (or?). The leakage would easily be >under control, and, should the in/output circuitry accept it, the ground wire not >necessary (I doubt that "just" the transformer would be sufficinet for this). just make sure the transformer meets medical isolation requirements otherwise you will "smoke" your patient with ground leakage currents!!!!!!!!!!!!!!!!!!!!! This is a one really serious reason why medical electronic equipment is so expensive. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.