upsidedown@downunder.com wrote:> On Wed, 09 Nov 2011 15:12:55 -0800, Joerg <invalid@invalid.invalid> > wrote: > >> upsidedown@downunder.com wrote: >>> On Wed, 09 Nov 2011 14:26:20 -0800, Joerg <invalid@invalid.invalid> >>> wrote: >>> >>>>> Of course you might have to _power_ something across that isolation gap, >>>>> too. >>>>> >>>> Oh, that's easy. You let off a ton of press releases that you will use >>>> the super duper green-tech solar approach, then get some major loan >>>> guarantees, smile when the TV crews and the big brass from the hill is >>>> visiting. Then let the sun supply it. If it don't shine no more you just >>>> file for chapter 11 and start over :-) >>> Supply your own "sun" from a laser via an optical fiber to a solar >>> cell. Return the measurement through an other fiber. Such tricks are >>> used e.g. for phase current measurements on 400 kV (and above) HV >>> lines. >>> >> But they won't give you a fat loan guarantee for that :-) >> >> Anyhow, I always ferry power over magnetically. It is very cheap and you >> can pump tens of watts to the other side if needed. That's kind of tough >> to do with lasers and solar cells. > > > While a 10-30 cm ferrite rod with a well insulated primary at one end > and a secondary at the other end can transfer a lot of power, the > distance between the primary and secondary is quite small, if we have > to worry about lightning strikes. >Well, the discussion started at 5kV which isn't a whole lot. Now we are already at lightning hits :-) Many ways to get a lot of insulation. One method can be seen in things like electric toothbrushes. And those are without fancy magnetics, it's all rather simple in there and every penny in production counts. If you really engineer things out well, keep it resonant and all that, you can transfer a lot of power through several centimeters of glass, porcelain or other insulation materials. Think about it this way: There has to be some way those 380kV are transformed down to more manageable local voltage levels, you can buy transformers for that off the shelf. Ok, big shelves. They are expected to work at 50Hz or 60Hz and at very good efficiency, soemthing we don't have to be concerned about when transferring a few watts or tens of watts. Sometimes I wish there was an ISM frequency in the 2-3MHz range, or maybe even one just below the AM band. It would make life so much easier in that field. -- Regards, Joerg http://www.analogconsultants.com/
RF galvanic isolator ...
Started by ●November 8, 2011
Reply by ●November 9, 20112011-11-09
Reply by ●November 10, 20112011-11-10
In article <4e96de0d-ccc6-4084-a208-1612a2dc087a@n6g2000vbg.googlegroups.com>, halong <ccon67@netscape.net> wrote:> On Nov 8, 11:22�am, Tim Wescott <t...@seemywebsite.com> wrote: > > On Tue, 08 Nov 2011 10:08:48 -0700, Jim Thompson wrote: > > > On Tue, 08 Nov 2011 10:18:14 -0600, Tim Wescott <t...@seemywebsite.com> > > > wrote: > > > > >>On Tue, 08 Nov 2011 06:41:54 -0800, halong wrote: > > > > >>> Hi all, > > > > >>> I have a 200 MHz clock signal to send across the barrier of 5 KV I'm > > >>> thinking of some RF transformer - perhaps "air core" ... but can't > > >>> find any good candidate > > > > >>Have you tried widening your search to just specify a 200MHz signal > > >>frequency and galvanic isolation? �I think the 200MHz part isn't a big > > >>deal, but the galvanic isolation may be. > > > > >>If EMI is a concern then an air core transformer may not be a good > > >>choice: it's going to leak RF like a sieve, and the spacing you'll need > > >>for 5k isolation won't help. > > > > >>If your production volumes are low, anything that works right off the > > >>bat is way less expensive than you think -- engineering time costs > > >>money, after all. �So you may want to at least price out your fiber > > >>optics solution. > > > > > Fiber optics? �Definitely best for a clock! > > > > I hadn't mentioned anything, but presumably he wants to get data back > > from wherever his clock is going -- fiber would be good for that, too. > > > > --www.wescottdesign.com > > You bet :-) > > The clock has the timing reference information, that's why we need > it...otherwise we can populate a crystal on the other side :-) > > Hmmm, speaking out of this I used to dream about a wireless clock > reference, and it's available everywhere in a closed system :-)))))Just for fun, pull the timing information from the cell phone network (or GPS) and phase lock all your disparate local oscillators to that. Yeah, a totally impractical idea, but nearly implementable.
Reply by ●November 10, 20112011-11-10
On Nov 9, 10:05=A0pm, Ralph Barone <invalid@not_real.ca> wrote:> In article > <4e96de0d-ccc6-4084-a208-1612a2dc0...@n6g2000vbg.googlegroups.com>, > > > > =A0halong <cco...@netscape.net> wrote: > > On Nov 8, 11:22=A0am, Tim Wescott <t...@seemywebsite.com> wrote: > > > On Tue, 08 Nov 2011 10:08:48 -0700, Jim Thompson wrote: > > > > On Tue, 08 Nov 2011 10:18:14 -0600, Tim Wescott <t...@seemywebsite.=com>> > > > wrote: > > > > >>On Tue, 08 Nov 2011 06:41:54 -0800, halong wrote: > > > > >>> Hi all, > > > > >>> I have a 200 MHz clock signal to send across the barrier of 5 KV =I'm> > > >>> thinking of some RF transformer - perhaps "air core" ... but can'=t> > > >>> find any good candidate > > > > >>Have you tried widening your search to just specify a 200MHz signal > > > >>frequency and galvanic isolation? =A0I think the 200MHz part isn't =a big> > > >>deal, but the galvanic isolation may be. > > > > >>If EMI is a concern then an air core transformer may not be a good > > > >>choice: it's going to leak RF like a sieve, and the spacing you'll =need> > > >>for 5k isolation won't help. > > > > >>If your production volumes are low, anything that works right off t=he> > > >>bat is way less expensive than you think -- engineering time costs > > > >>money, after all. =A0So you may want to at least price out your fib=er> > > >>optics solution. > > > > > Fiber optics? =A0Definitely best for a clock! > > > > I hadn't mentioned anything, but presumably he wants to get data back > > > from wherever his clock is going -- fiber would be good for that, too=.> > > > --www.wescottdesign.com > > > You bet :-) > > > The clock has the timing reference information, that's why we need > > it...otherwise we can populate a crystal on the other side :-) > > > Hmmm, speaking out of this I used to dream about a wireless clock > > reference, and it's available everywhere in a closed system :-))))) > > Just for fun, pull the timing information from the cell phone network > (or GPS) and phase lock all your disparate local oscillators to that. > Yeah, a totally impractical idea, but nearly implementable.uh huh, why in the earth we have to spend tons of joules just to have information such as time reference moving from one place to another ? :-))) it remind my phone bill still in the in the inbox
Reply by ●November 10, 20112011-11-10
halong schrieb:> Hi all, > > I have a 200 MHz clock signal to send across the barrier of 5 KV > I'm thinking of some RF transformer - perhaps "air core" ... but can't > find any good candidateHello, I would send it over an optical fiber. If 200 MHz is to fast for a cheap modulator, the frequency is easily converted to 6.25 MHz with a binary divider 1:32. If you need the 200 MHz on the other side, a PLL could transform back to the 200 MHz. But to measure the 200 MHz clock, the 6.25 MHz signal will do, just multiply the result with 32. Bye
Reply by ●November 10, 20112011-11-10
Uwe Hercksen wrote:> > > halong schrieb: > >> Hi all, >> >> I have a 200 MHz clock signal to send across the barrier of 5 KV >> I'm thinking of some RF transformer - perhaps "air core" ... but can't >> find any good candidate > > Hello, > > I would send it over an optical fiber. If 200 MHz is to fast for a cheap > modulator, the frequency is easily converted to 6.25 MHz with a binary > divider 1:32. If you need the 200 MHz on the other side, a PLL could > transform back to the 200 MHz. But to measure the 200 MHz clock, the > 6.25 MHz signal will do, just multiply the result with 32. >That's almost like a client had it until I came in :-) Fiber-optics stuff is very expensive, much of it due to labor. Someone has to assemble the connectors onto the fiber, polish the front and all that. Or you have to buy a fiber link, curl it up and strap it down. But that'll be expensive as well. A custom transformer is much cheaper. Once production training has occurred those can be assembled in minutes. -- Regards, Joerg http://www.analogconsultants.com/
