What's the best form of surge protection

[JohnW2]

In most cases it's really obvious - in some cases just by visual inspection ! Though I suppose you can argue that it makes little difference whether it's the phone line changing voltage relative to the mains, or vice versa. In one case, it was clear that the phone line was part of the circuit by the physical damage to the modem in the laptop and where it had arced over to the main board.

Yes, of course, if there is chronological association with lightning and visibly obvious physical damage, then the causation is obvious - but are we not then talking more of a 'lightning strike' (of overhead cables, nearby earth or whatever) rather than induced transients?

... this is bad practice as even a thick earth wire has considerable impedance when you throw a spike the rise time of a lightning strike at it. .......... They'd setup a demo with a small DC motor fed through a very expensive* filter earthed to the alli panel with about a foot of very stout earth braiding. The filter was ineffective as the earth braid was too high an impedance to the noise - as an amplifier attached to the DC supply demonstrated.

I have to say that that 'demo' rather surprises me. I would have thought that the impedance of "very stout earth braiding" at any frequency low enough for a human being to hear (which I presume is what you mean) would be extremely low.

Kind Regards, John

 

[SimonH2]

Interesting what SimonH2 says, I noted that BT master sockets have spark gaps, and other items to protect from non direct lighting voltages, however in my house there is no earth run to the master socket, so the protection devices will do very little.

Actually, only the spark gap is there for surge protection, and it only protects against differential mode surges (ie between the two wires). As you say, there's no earth and so nothing at all to protect against common mode.

However with my mothers house the GPO have fitted a steel earth rod and a bare copper wire from it to the first telephone box. As to if still used I don't know?

I know back in the 1950's when the phone was fitted there was a shortage of lines, so my dad was put on a party line
...
I am not sure if the earth stake was to do with the party line or to stop damage due to lighting strikes?

The earth rod was purely for the party line, nothing to do with protection. You'd probably also remember having to press a button on the phone to get dial tone. I remember we had a party line when we moved back in 1969.
With a party line, they put the ringer between one line and earth for one user, and between the other line and earth for the other user - that's how they ring the two lines independently. For dialling out, you press the button and that connects one line to earth (again via the rod) which pulls in a relay at the exchange and connects you to the appropriate port in the exchange.
AFAIK, the earth rod wasn't there for any form of protection, other than incidental - in the same way that equipotential bonding isn't there for earthing, but does so as a side effect of one pat already being earthed.

the SP1 device was the spark gap, R1 and C1 were to make the bell ring, but there seems to be no earth connection.

Indeed. As above, the spark gap is for differential mode protection, the cap is to split off the AC ringing signal, and the resistor is really only there for testing - so even with nothing plugged in, there's a minimum load on the line when a test is run (if it's not there, the line will test as faulty - been there ...)

The broad band filter

seems to isolate the phone, but not the modem, ups sorry should not be called a modem with broadband as it only swaps the voltage from 50 to 5 it does not modulate or demodulate the signal.

Actually, it is still a modem, it really does do a heck of a lot more than just level shift ! xDSL systems are what they call "discreet multitone" - in effect a lot of modems, each operating on a small bit of bandwidth (a "bin"), and handling a small amount of the overall bandwidth. Some modems/routers will show you the spectral use on the line (a graph of bits/bin) and it's usually clear to see the bit/bin drop off with frequency as the higher frequencies suffer higher attenuation and so can carry less information. The reason FTTC can do faster speeds than ADSL from the exchange is basically due to being able to use higher frequencies over the shorter cable, shorter cable = lower attenuation.

Also, the filter doesn't actually isolate the phone - it's just a low pass filter to keep the xDSL signals away from the phone and so stop the phone interfering with the broadband. Some filter do high-pass filter the broadband, others do nothing to it (as shown in your diagram). And contrary to popular belief (and if you believe some clueless BTOR people, the law ) you don't actually need a filter if there's no phone attached.

It does seem some master sockets do have an earth

...

Again, functional earth only - for systems that need it (primarily party lines if they still exist. IIRC it connects to terminal 4 of the internal wiring connectors - 2&5 being the voice pair, and 3 the ringer line (1&6 being completely unused in every system I've worked with).
As an aside, if you don't filter at the master socket (ie with a filtered faceplate) and feed all the internal wiring via the filtered port, it's generally advised to disconnect terminal 3 as it's rarely used by modern equipment and it creates a branched (ie in RF terms, quite complex) circuit for the ADSL signal.

They've moved on from party lines, and I found this page that summarises the options used.

I have to say that that 'demo' rather surprises me. I would have thought that the impedance of "very stout earth braiding" at any frequency low enough for a human being to hear (which I presume is what you mean) would be extremely low.

I think it surprised those of us who thought we knew a bit about such stuff - but that was the effect.

 

[deadshort]

I am going to purchase some high end AV equipment and would like to protect it against unexpected surges (e.g. lightning).

Linn as they are dead anyway

 

[SimonH2]

Just to add, while discussing earths on phone sockets ...
Some internal phone systems (ie PBXs) used a functional earth for signalling (Earth Loop Recall) well into the 90s, although by then I think most modern systems (if they used it at all) used "timed Break Recall" - the equivalent of tapping the hook switch and opening the loop for a defined period.

And the reason for the 3rd ringer wire when many countries (notably the USA) were already using 2 wire systems ? It's a hangover from mechanical bells and pulse dialling. Prior to the "modern" BT socket (IIRC designed by Vero Electronics, and available with different keying as the "W" form) phones were either hardwired or used a multipole jack plug - and there were a minimum of 4 wires. Bells were in a loop, the jack sockets had a switch contact to maintain the loop when a plug was inserted, and the phones were wired in such a way that when the dial wasn't against it's rest stop it disconnected the bell loop - all that was to stop the bells "tinkling" while dialling. When they switched to the modern sockets, the bells went to parallel connections - and phones were all (initially) required to short out the ringer line while dialling for the same reason.
These days it's hard to find a pulse dialling phone, and all are pretty well designed for 2 wire operation and include anti-tinkle in the electronics.

And when I realise I remember all that, the days of the phone stuck on the wall in the hallway, pressing the recall button to get dialtone (if the spinsters at No 7 weren't already using the line), bell tinkle when I jerry rigged my first modem (a fast one, it did 1200bps down, only 75bps up !) into the line, and converting the house from the old jack sockets to the new ones and playing dumb when the BT guy came to fix a fault
Then I suddenly realise how the years are adding up

And while looking for references for that, I came across this group and this site.

 

[JohnW2]

... my first modem (a fast one, it did 1200bps down, only 75bps up !) ....

Ah - like my second one. The first (which I came across in the cellar recently!) could only do 300 bps (up and down).

Kind Regards, John

 

[SimonH2]

Ah the days ... When you'd dial up, download the messages from the bulletin board, hangup. Then you'd read the messages, compose your reply in a text file, then dial back in to copy & paste the replies. Then my favourite BB went to a forum type arrangement that needed you to be online all the time, and more than a few of us complained that we couldn't afford to stay as members due to the dial up costs.
Then along came Demon Internet with their "tenner a month" internet - I remember the days when their user count was in hundreds, they pioneered some of the stuff others later took for granted, and for a while I could look at their graphs of subscriber numbers (which they had online) and go .... "that level bit there was when <some service> got over-subscribed, and the line goes up again when they fixed it".
And now people complain if they can't get on FarceBork while doing their shopping
At least back then, MS hadn't yet screwed up email by getting people to think that top posting is the "right" way to do it

 

[JohnW2]

Ah the days ... When you'd dial up, download the messages from the bulletin board, hangup. Then you'd read the messages, compose your reply in a text file, then dial back in to copy & paste the replies. Then my favourite BB went to a forum type arrangement that needed you to be online all the time, and more than a few of us complained that we couldn't afford to stay as members due to the dial up costs.

In the earliest (pre-Internet) days of CompuServe, in addition to the telephone costs of the dial-up, there was also a per-minute charge for being connected to CS and, if I recall correctly, an additional per-minute charge for participation in forums. There was also a 'per item' charge for both outgoing and incoming e-mails. I certainly recall having to seriously discipline myself at the time in order to avoid horrendous monthly CS bills!

Kind Regards, John

 

[ericmark]

Ah - like my second one. The first (which I came across in the cellar recently!) could only do 300 bps (up and down).

Kind Regards, John

I still have a BAYCOM well really AMICOM modem with a switch 1200 or 300 baud, I will guess now there are no BBS's to connect to? Slow is an under statement, talking to a guy in Aus I could put the kettle on and make a cuppa while waiting for reply. Hand Morse was quicker.

Seem to remember it was adapted from AX25?

But still no real way to protect domestic premises from lighting, and earthing is not good, I remember a friends long wire being reduced to copper balls along the lawn. Had he not earthed it likely it would have survived as likely it would not have been hit.

 

[bernardgreen]

The earth rod was purely for the party line, nothing to do with protection.

In the very early days the earth was used to protect telephones on open wire ( overhead ) from lightning strikes and contact with overhead power lines.

Protector Box & Fuses No1.2/2 ( http://www.telephonesuk.co.uk/connection_boxes.htm )

The terminal on the left was the Earth, voltage arresters were fitted between the spring clips
The telephone connected to the terminals in the middle and the iver head line to the terminals at the right ( under the fuses )

The extremely small number of incidents of lightning strikes led to these protectors being discontinued on almost all new lines. Fuse were replaced by links as and when a fuse failed.
.

 

[bernardgreen]

The filter was ineffective as the earth braid was too high an impedance to the noise

That can be used to help keep lightining strike energy out of a building. Bring the cable down below the point of entry and then U turn it back up to the hole with a U bend radius of about a foot. Fit a substantial earthed rod along side the descending side of the U and when the cable is struck by lightning the impedance of the U bend can cause the voltage spike to take the lower impedance capacitive route to ground by jumping off the cable and onto the earthed rod. Once the voltage spike has jumped and created a plasma route the majority of th currrent will take that path as well. It may not fully protect equipment in the building but it may well reduce the risk of fire and /or structural damage inside the building.

 

[SimonH2]

But still no real way to protect domestic premises from lighting, and earthing is not good, I remember a friends long wire being reduced to copper balls along the lawn. Had he not earthed it likely it would have survived as likely it would not have been hit.

I think the best way it can be described is "it's complicated"
I recall that in school physics, and at uni, we were taught that lightning conductors have sharp points at the top as that causes discharge without a strike. The sharp point intensifies the electric field strength and creates a corona discharge (or charge depending on which direction you're looking) around the point and this takes charge out of the air. Supposedly, as the theory goes, this avoids some strikes from even happening - though I imagine that's hard to prove one way or another.
On the other hand, I suspect that unless you live on top of a hill with nothing around you, then direct strikes are probably very rare - and if they do happen, not defendable on a domestic budget. So we're generally back to defending against the surges that come to us either through induction, or strikes on the lines some distance away, or due to ground potential changes from ground strikes (what I think did for all my serial kit). That I think is defendable, by creating an equipotential zone and using surge protectors (or for water & gas, just good bonding) on everything crossing the boundary. Ideally that means bringing all the services in from the outside at one point, fitting quality surge protectors on power and phone lines, and bonding everything together at the one "earthed" point. If (for example) there's an indirect strike on the mains, "earth" as measured at the MET may spike a few hundred/thousand volts relative to "earth" as in that mass under your feet - but everything will ride up and down with that so it's all common mode and apart from parasitic capacitance there's now path for differential mode currents to flow. It would pretty well avoid the problem where your fax machine/modem/cordless phone/answering machine/whatever is connected to the mains and to the phone line - and suddenly gets a few kV of differential voltage between the two sides and lets the magic smoke out.

As another aside, a friend is a keen walker and mountain climber. He tells of lightning being a particular danger up on the hills - and how you can feel the charge in your hair long before there's any other signs of a lightning storm coming. Advice is "get as far down the mountain as you can, as quick as you can, and when it gets to 'imminent' then crouch down into as small a ball as you can". He then went on to add ... " and don't shelter in a cave, you know how a spark plug has that pip sticking up in the middle ? Well you don't want to be that pip "

 

[bernardgreen]

I think the best way it can be described is "it's complicated"

That is an understatement

I recall that in school physics, and at uni, we were taught that lightning conductors have sharp points at the top as that causes discharge without a strike

The electric wind, electrons streaming off the high voltage point created a wind strong enough to deflect a candle flame. It also generated a force on the point that cpild be used to turn a rotor

https://www.physics.purdue.edu/demos/display_page.php?item=5A-23

A lightning to ground strike in Dagenham Essex did not cause any damage to equipment in a communications equipment room other than one modem ( 1200 / 75 baud ) which was fried around the telephone line terminals. This modem was connected directly to a modem in Lambeth by private wire ( GPO copper pair, street cabinet to street cabinet all the way ). All other lines to the room were routed via the local telephone exchange. The line to Lambeth stayed at "lambeth" potential while all around it in the comms room followed the local dagenham ground potential spike when the lightning hit.,