Forums

Lighning protection

Started by Phil Hobbs January 27, 2016
On 01/28/2016 01:11 PM, Cydrome Leader wrote:
> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote: >> So I'm looking at lightning protection for the mezzanine units of the >> cotton spark detection system. >> >> I have a copy of Standtler's book, "Protection of electronic circuits >> from overvoltage", which is useful but a bit out of date (1989). (He >> doesn't know about HV depletion MOSFETs or polyfuses, for instance.) >> >> It looks as though I can protect the isolated RS485 pair and the >> power/common pair with a cascade consisting of >> >> (3-terminal spark gap) 500v 1a depl MOS >> 0-------*-----------* *--*-----*-----*--* *-------*-----(Iso RS485) >> A | | V | | | V | | >> | ---------- | --------- V >> | -------* | *------- /---/ bidirectional >> | | | | A TVS >> | *-----*-----* | >> V | >> -----GND | >> A | >> 0-------* .... similar.... .... * ... >> B >> >> Lighter-weight lines are easier to protect, of course--Ethernet just >> uses transformers. >> >> Any wisdom about other ways of proceeding? > > fiber media converter. > > They make them for serial as well as ethernet. Complete isolation between > end points, and distance limiations for serial and interference is no > longer a problem. The only issue is price and where to plug in the power > packs for each side although they do make rack mount concentrators as > well. >
Relying on having mains wiring up in the attic of a cotton mill well outside Dhaka is going to be problematic, I think. Relying on anything staying plugged in is also a problem. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
upsidedown@downunder.com wrote:


> I have been working with industrial communication systems for decades > also in tropics with daily thunderstorms. I have used the thumb of > rules: > > 1.) you can use non-isolated RS-232 for equipments in the same room > 2.) use galvanic isolation (2 kV) RS-422/485 in the same building > 3.) use fibres between buildings
Yes, that is a good rule. I did build a system to transmit bidirectional RS-232 between buildings quite some years ago. I had a separately-shielded twisted pair cable run between buildings, about 20 pair, I think. Then, I made up little boards that had a pair of op-amp circuits on them, one a differential transmitter, the other a differential receiver. The single-ended levels were RS-232 levels, the differential signals were something like 1 V differential, so the slew rate was less of a problem. This worked great at 9600 baud. it also withstood several close lightning strikes that blew out various gear in the buildings, and completely blew up a locally-built optoisolated system for the same purpose. I assume that system was popped from the RS-232 side. The resistances used in my differential converters plus the larger voltage range of the op-amps must have been part of why my design survived. Your rule #1 only works if all the gear in the room is bonded to the same ground. Lots of places have several power panels feeding different equipment in one room (such as terminals on one feed, and computers on a different feed.) Some CEs I knew long ago had a data center in downtown St. Louis that was fed from THREE different substations. The problems they had there were beyond belief. People sometimes reported sparks jumping between equipment and power panels, people being knocked on their rear ends when touching two pieces of nearby gear at the same time, and LOTS of crashes, where 5 computer systems from different vendors would crash at the same instant. One of these substatiosn was fed by a frequency converter station from Ontario Hydro, fed 25 Hz power to the substation that converted it to 60 Hz with giant roatary converters that were half a city block wide. Jon
On Thursday, January 28, 2016 at 12:40:49 PM UTC-8, Jon Elson wrote:
> upsidedown@downunder.com wrote:
> > rules: > > > > 1.) you can use non-isolated RS-232 for equipments in the same room > > 2.) use galvanic isolation (2 kV) RS-422/485 in the same building > > 3.) use fibres between buildings > > Yes, that is a good rule.
> Your rule #1 only works if all the gear in the room is bonded to the same > ground. Lots of places have several power panels feeding different > equipment in one room (such as terminals on one feed, and computers on a > different feed.)
It's a violation of US national electrical code to have 'different' grounds, of course; there's only one exception, for secondary power systems (like, isolated power in hospital operating rooms). Even labs with busbar secondary ground systems, have bonding to the power ground. I hope rule #1 will always work. Formerly, I hoped and trusted...
On Thu, 28 Jan 2016 10:34:32 -0500, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

>On 01/28/2016 07:55 AM, legg wrote: >> On Wed, 27 Jan 2016 21:25:14 -0500, Phil Hobbs >> <pcdhSpamMeSenseless@electrooptical.net> wrote: >>
<snip>
>> >> You might check the relevent sections of IEC61000 and IEEE/ANSI C62 to >> figure out just what level of transient or surge immunity that you >> intend and what exposure level is targeted. > >I wish. The design requirement is "It has to work through repeated >thunderstorms in an old wood/steel/cinderblock/mud/thatch/whatever >building in the hills of Bangladesh, with 200 metres of rusty steel >ductwork that may be carrying mains current, and very few grounds." >
Then the first thing to do is to include a 4-foot passivated ground rod in the instrument's kit, to be pounded into the ground at the duct port entry to the receiver's building location. This needs to couple into the wire bundle's ground/drain/earth conductor using braided grounding wire or strap. Supposedly the same arrangement can be expected at the TX end, using the same kit. This is actually the best location for any air-gapped suppressors on your bill of material, if access to data or power conductors is physically possible. Same principal as a domestic residence's antenna lead-in suppressor. Transient suppression works on the assumption that there's a route to redirect the current involved, that does not involve the sensitive circuitry, or other sensitive leads exiting the equipment container. The bulk energy entaileded is expected to dissipate in the tailored transient path, with any residual local voltage disturbances (shorter duration peak, lower amplitude tail) being handled by lower-powered limiters. Most circuits that you see published already expect that this basic protection level is present. Limiters at the service entrance levels can be expected to degrade with use and to require regular inspection/ replacement. The northern districts of the Indian Subcontinent aren't the worst place for lightning, being on par with the American gulf states, but an average of 1 strike a week per square kilometer isn't something to ignore. RL
whit3rd wrote:

> On Thursday, January 28, 2016 at 12:40:49 PM UTC-8, Jon Elson wrote: >> upsidedown@downunder.com wrote: > >> > rules: >> > >> > 1.) you can use non-isolated RS-232 for equipments in the same room >> > 2.) use galvanic isolation (2 kV) RS-422/485 in the same building >> > 3.) use fibres between buildings >> >> Yes, that is a good rule. > >> Your rule #1 only works if all the gear in the room is bonded to the same >> ground. Lots of places have several power panels feeding different >> equipment in one room (such as terminals on one feed, and computers on a >> different feed.) > > It's a violation of US national electrical code to have 'different' > grounds, of course; there's only one exception, for secondary power > systems (like, isolated power > in hospital operating rooms). Even labs with busbar secondary ground > systems, have bonding to the power ground. > > I hope rule #1 will always work. Formerly, I hoped and trusted...
I should have said bonded to the same PANEL. Yes, all the grounds typically sprout from one common point. But, when lightning hits nearby, the magnetic field can be enormous and easily penetrates the whole building. If you have a big loop, where one feed comes around the left side of the room, and one comes in the right side, maybe coming down through several panels along a hallway, you can have a loop that encloses a huge area. The induced voltage is proportional to the area of the loop. Now, you can have hundreds or thousands of Volts between the grounds on two outlets in the same room. This is, of course, not the best practice in wiring up a building, but is quite common. That will cause stuff to let out the magic smoke. Jon
On Thu, 28 Jan 2016 14:43:59 -0500, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

>On 01/28/2016 01:11 PM, Cydrome Leader wrote: >> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote: >>> So I'm looking at lightning protection for the mezzanine units of the >>> cotton spark detection system. >>> >>> I have a copy of Standtler's book, "Protection of electronic circuits >>> from overvoltage", which is useful but a bit out of date (1989). (He >>> doesn't know about HV depletion MOSFETs or polyfuses, for instance.) >>> >>> It looks as though I can protect the isolated RS485 pair and the >>> power/common pair with a cascade consisting of >>> >>> (3-terminal spark gap) 500v 1a depl MOS >>> 0-------*-----------* *--*-----*-----*--* *-------*-----(Iso RS485) >>> A | | V | | | V | | >>> | ---------- | --------- V >>> | -------* | *------- /---/ bidirectional >>> | | | | A TVS >>> | *-----*-----* | >>> V | >>> -----GND | >>> A | >>> 0-------* .... similar.... .... * ... >>> B >>> >>> Lighter-weight lines are easier to protect, of course--Ethernet just >>> uses transformers. >>> >>> Any wisdom about other ways of proceeding? >> >> fiber media converter. >> >> They make them for serial as well as ethernet. Complete isolation between >> end points, and distance limiations for serial and interference is no >> longer a problem. The only issue is price and where to plug in the power >> packs for each side although they do make rack mount concentrators as >> well. >> > >Relying on having mains wiring up in the attic of a cotton mill well >outside Dhaka is going to be problematic, I think. Relying on anything >staying plugged in is also a problem. > >Cheers > >Phil Hobbs
It would be cool if you could collect and transmit the light from the burning cotton with a fiber. Something big and tacky like Toslink. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
On 01/28/2016 05:38 PM, John Larkin wrote:
> On Thu, 28 Jan 2016 14:43:59 -0500, Phil Hobbs > <pcdhSpamMeSenseless@electrooptical.net> wrote: > >> On 01/28/2016 01:11 PM, Cydrome Leader wrote: >>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote: >>>> So I'm looking at lightning protection for the mezzanine units of the >>>> cotton spark detection system. >>>> >>>> I have a copy of Standtler's book, "Protection of electronic circuits >>>> from overvoltage", which is useful but a bit out of date (1989). (He >>>> doesn't know about HV depletion MOSFETs or polyfuses, for instance.) >>>> >>>> It looks as though I can protect the isolated RS485 pair and the >>>> power/common pair with a cascade consisting of >>>> >>>> (3-terminal spark gap) 500v 1a depl MOS >>>> 0-------*-----------* *--*-----*-----*--* *-------*-----(Iso RS485) >>>> A | | V | | | V | | >>>> | ---------- | --------- V >>>> | -------* | *------- /---/ bidirectional >>>> | | | | A TVS >>>> | *-----*-----* | >>>> V | >>>> -----GND | >>>> A | >>>> 0-------* .... similar.... .... * ... >>>> B >>>> >>>> Lighter-weight lines are easier to protect, of course--Ethernet just >>>> uses transformers. >>>> >>>> Any wisdom about other ways of proceeding? >>> >>> fiber media converter. >>> >>> They make them for serial as well as ethernet. Complete isolation between >>> end points, and distance limiations for serial and interference is no >>> longer a problem. The only issue is price and where to plug in the power >>> packs for each side although they do make rack mount concentrators as >>> well. >>> >> >> Relying on having mains wiring up in the attic of a cotton mill well >> outside Dhaka is going to be problematic, I think. Relying on anything >> staying plugged in is also a problem. >> >> Cheers >> >> Phil Hobbs > > It would be cool if you could collect and transmit the light from the > burning cotton with a fiber. Something big and tacky like Toslink. > >
Interesting idea--pipe the signal all the way to detection modules installed in the panel, and so avoid the whole mess. There's a bit of an etendue issue, though, and a lot of the signal is in the IR where plastic fibre isn't any good. My first official act on this project was to spend about a week and a half doing photon budgets for various scenarios. We wound up with a small InGaAs diode as the most cost-effective choice, interestingly. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
On 01/28/2016 04:31 PM, legg wrote:
> On Thu, 28 Jan 2016 10:34:32 -0500, Phil Hobbs > <pcdhSpamMeSenseless@electrooptical.net> wrote: > >> On 01/28/2016 07:55 AM, legg wrote: >>> On Wed, 27 Jan 2016 21:25:14 -0500, Phil Hobbs >>> <pcdhSpamMeSenseless@electrooptical.net> wrote: >>> > <snip> >>> >>> You might check the relevent sections of IEC61000 and IEEE/ANSI >>> C62 to figure out just what level of transient or surge immunity >>> that you intend and what exposure level is targeted. >> >> I wish. The design requirement is "It has to work through repeated >> thunderstorms in an old wood/steel/cinderblock/mud/thatch/whatever >> building in the hills of Bangladesh, with 200 metres of rusty steel >> ductwork that may be carrying mains current, and very few >> grounds." >> > > Then the first thing to do is to include a 4-foot passivated ground > rod in the instrument's kit, to be pounded into the ground at the > duct port entry to the receiver's building location.
Which very likely would instantly be stolen and sold for the metal value. In rural Bangladesh, folks are super poor. (You can tell a first-world country because the poor people are fat. That's far from true elsewhere.) And of course people steal copper wiring and plumbing even in Canada and the US.
> This needs to couple into the wire bundle's ground/drain/earth > conductor using braided grounding wire or strap. Supposedly the same > arrangement can be expected at the TX end, using the same kit. This > is actually the best location for any air-gapped suppressors on your > bill of material, if access to data or power conductors is > physically possible. Same principal as a domestic residence's antenna > lead-in suppressor.
> > Transient suppression works on the assumption that there's a route to > redirect the current involved, that does not involve the sensitive > circuitry, or other sensitive leads exiting the equipment container. > The bulk energy entaileded is expected to dissipate in the tailored > transient path, with any residual local voltage disturbances (shorter > duration peak, lower amplitude tail) being handled by lower-powered > limiters. > > Most circuits that you see published already expect that this basic > protection level is present. Limiters at the service entrance levels > can be expected to degrade with use and to require regular > inspection/ replacement. > > The northern districts of the Indian Subcontinent aren't the worst > place for lightning, being on par with the American gulf states, but > an average of 1 strike a week per square kilometer isn't something > to ignore. > > RL >
Part of the fun of this project is that I can't make a lot of assumptions of that sort. The customer has been in the business for quite awhile, and so can provide good guidance about what has and hasn't worked in the past, but this system's architecture is quite different from what they've done before. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
On Thu, 28 Jan 2016 19:05:11 -0500, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

>On 01/28/2016 05:38 PM, John Larkin wrote: >> On Thu, 28 Jan 2016 14:43:59 -0500, Phil Hobbs >> <pcdhSpamMeSenseless@electrooptical.net> wrote: >> >>> On 01/28/2016 01:11 PM, Cydrome Leader wrote: >>>> Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote: >>>>> So I'm looking at lightning protection for the mezzanine units of the >>>>> cotton spark detection system. >>>>> >>>>> I have a copy of Standtler's book, "Protection of electronic circuits >>>>> from overvoltage", which is useful but a bit out of date (1989). (He >>>>> doesn't know about HV depletion MOSFETs or polyfuses, for instance.) >>>>> >>>>> It looks as though I can protect the isolated RS485 pair and the >>>>> power/common pair with a cascade consisting of >>>>> >>>>> (3-terminal spark gap) 500v 1a depl MOS >>>>> 0-------*-----------* *--*-----*-----*--* *-------*-----(Iso RS485) >>>>> A | | V | | | V | | >>>>> | ---------- | --------- V >>>>> | -------* | *------- /---/ bidirectional >>>>> | | | | A TVS >>>>> | *-----*-----* | >>>>> V | >>>>> -----GND | >>>>> A | >>>>> 0-------* .... similar.... .... * ... >>>>> B >>>>> >>>>> Lighter-weight lines are easier to protect, of course--Ethernet just >>>>> uses transformers. >>>>> >>>>> Any wisdom about other ways of proceeding? >>>> >>>> fiber media converter. >>>> >>>> They make them for serial as well as ethernet. Complete isolation between >>>> end points, and distance limiations for serial and interference is no >>>> longer a problem. The only issue is price and where to plug in the power >>>> packs for each side although they do make rack mount concentrators as >>>> well. >>>> >>> >>> Relying on having mains wiring up in the attic of a cotton mill well >>> outside Dhaka is going to be problematic, I think. Relying on anything >>> staying plugged in is also a problem. >>> >>> Cheers >>> >>> Phil Hobbs >> >> It would be cool if you could collect and transmit the light from the >> burning cotton with a fiber. Something big and tacky like Toslink. >> >> >Interesting idea--pipe the signal all the way to detection modules >installed in the panel, and so avoid the whole mess. > >There's a bit of an etendue issue, though, and a lot of the signal is in >the IR where plastic fibre isn't any good. My first official act on >this project was to spend about a week and a half doing photon budgets >for various scenarios. We wound up with a small InGaAs diode as the >most cost-effective choice, interestingly.
I don't suppose the ducts are willing to be straight, to give you a long view. Stuffing wide-angle light into a tiny fiber is inefficient, so you'd need a lot of light for that to work. I wonder what burning cotton "sounds" like, namely converting the light to audio. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
On Thu, 28 Jan 2016 20:26:49 +0200, upsidedown@downunder.com wrote:

>On Thu, 28 Jan 2016 10:34:32 -0500, Phil Hobbs ><pcdhSpamMeSenseless@electrooptical.net> wrote: > >>> You might check the relevent sections of IEC61000 and IEEE/ANSI C62 to >>> figure out just what level of transient or surge immunity that you >>> intend and what exposure level is targeted. >> >>I wish. The design requirement is "It has to work through repeated >>thunderstorms in an old wood/steel/cinderblock/mud/thatch/whatever >>building in the hills of Bangladesh, with 200 metres of rusty steel >>ductwork that may be carrying mains current, and very few grounds." > >I have been working with industrial communication systems for decades >also in tropics with daily thunderstorms. I have used the thumb of >rules: > >1.) you can use non-isolated RS-232 for equipments in the same room >2.) use galvanic isolation (2 kV) RS-422/485 in the same building >3.) use fibres between buildings
How about variations of ISM radios (whether it be WiFi, Bluetooth, or some sort of mesh) available for cheap, these days? It's probably cheaper and more reliable than wires, especially for low bandwidth uses.
> >After my customers also understood these rules, I haven't heard >complaints by the end users in the tropics > >Rule #1 for lightning protection: There is no such thing as a ground >potential, the best that you can get is equipotential bonding !!