Telephone protection circuit needed

[Robert Scott]

In the recent ice storm, the 4800 volts from the power lines crossed over into
the phone lines. It burned out many circuits in our area. It also fried some
modems and phones in my house. Surprisingly, two connected modems survived.
These were both protected by whatever surge protection circuit they put in a
UPS. But it had the side effect of burning out that circuit in the UPS, so I
cannot use that again.

Rather than protect phone equipment piecemeal, I would like to protect all phone
equipement in my house using one circuit that I put at the entrance. I would
like to use a fuse. I gather that most commercial "surge protectors" do not
have fuses, but I think they might be a more reliable protection than varistors
or whatever they put in those commercial units. Has anyone done this? What
sort of current limit should I set? I guess the ringing with a maximum number
of phones connected would be the upper limit. But a fuse might not be enough.
What about a crowbar circuit? Is it possible to detect an overvoltage condition
with an active circuit and then short all lines to ground through some
fast-acting power FETs downstream of the fuses, just to make sure the fuses blow
right away? I don't mind replacing some power FETs and fuses in my own custom
circuit once in a while if it means I won't have to replace modems and phones.
I know ice storms and downed power lines are my focus right now, but I am also
looking toward spring with lightning will be the major risk.


Robert Scott
Ypsilanti, Michigan

 

[D from BC]

On Wed, 24 Jan 2007 18:45:31 GMT, ---@--- (Robert Scott) wrote:
[snip]

What about a crowbar circuit? Is it possible to detect an overvoltage condition
with an active circuit and then short all lines to ground through some
fast-acting power FETs downstream of the fuses, just to make sure the fuses blow
right away? I don't mind replacing some power FETs and fuses in my own custom
circuit once in a while if it means I won't have to replace modems and phones.
I know ice storms and downed power lines are my focus right now, but I am also
looking toward spring with lightning will be the major risk.


Robert Scott
Ypsilanti, Michigan

I haven't done any surge suppression/isolation active clamping
circuits yet...
I'm guessing some electronic process like:
1)Slow down the voltage surge risetime (If not slow already from line
inductance) to give a slow crowbar cct a chance to react
2)After the fuse blows from the crowbar cct., there's now a big spark
gap for cct isolation
3)Find some place to dump surge energy ...another spark gap?

Without looking, I'll guess this is a well beaten topic.
Check achieves.
D from BC

 

[Robert Scott]

From past experience, in a protective configuration where you have a
'fuse' protecting active components it is the silcon/germanium/whatever
that protects the 'fuse' and fails first...

Well, I had some melted 22-gauge wire. A 500 ma. fuse would have at least
protected that wire.



Robert Scott
Ypsilanti, Michigan

 

[Stanislaw Flatto]

In the recent ice storm, the 4800 volts from the power lines crossed over into
the phone lines. It burned out many circuits in our area. It also fried some
modems and phones in my house. Surprisingly, two connected modems survived.
These were both protected by whatever surge protection circuit they put in a
UPS. But it had the side effect of burning out that circuit in the UPS, so I
cannot use that again.

Rather than protect phone equipment piecemeal, I would like to protect all phone
equipement in my house using one circuit that I put at the entrance. I would
like to use a fuse. I gather that most commercial "surge protectors" do not
have fuses, but I think they might be a more reliable protection than varistors
or whatever they put in those commercial units. Has anyone done this? What
sort of current limit should I set? I guess the ringing with a maximum number
of phones connected would be the upper limit. But a fuse might not be enough.
What about a crowbar circuit? Is it possible to detect an overvoltage condition
with an active circuit and then short all lines to ground through some
fast-acting power FETs downstream of the fuses, just to make sure the fuses blow
right away? I don't mind replacing some power FETs and fuses in my own custom
circuit once in a while if it means I won't have to replace modems and phones.
I know ice storms and downed power lines are my focus right now, but I am also
looking toward spring with lightning will be the major risk.


Robert Scott
Ypsilanti, Michigan

From past experience, in a protective configuration where you have a
'fuse' protecting active components it is the silcon/germanium/whatever
that protects the 'fuse' and fails first.
Read what is time response of fast fuse against overcurrent percentage.
Interesting!

Stanislaw

 

[Jim Thompson]

From past experience, in a protective configuration where you have a
'fuse' protecting active components it is the silcon/germanium/whatever
that protects the 'fuse' and fails first.
Read what is time response of fast fuse against overcurrent percentage.
Interesting!

Stanislaw

I've designed some crowbar fuse blower circuits that circulated >100A
thru a 5A fuse... now you see it, now you don't ;-)

You have to be very careful with the foil pattern on the PCB to avoid
it being the fuse ;-)

...Jim Thompson

 

[Joerg]

Well, I had some melted 22-gauge wire. A 500 ma. fuse would have at least
protected that wire.

4800V is a whole lotta volts. If the fuse is long enough that plus a
huge MOV could have saved the day. Long in order to avoid arcing over.
Of course the fuse would have to be mounted in a way that things don't
arc over to any place that could ignite and burn down the house.

 

[Robert Scott]

You have to be very careful with the foil pattern on the PCB to avoid
it being the fuse ;-)

This isn't going to be a PCB. It is a one-off home-made thing, hand-wired. I
was going to use wood for the base, but with all this talk about arcing over,
maybe I ought to use plastic.


Robert Scott
Ypsilanti, Michigan

 

[mpm]

In the recent ice storm, the 4800 volts from the power lines crossed overinto
the phone lines.  

Part of the problem is the voltages involved in normal phone operation.
You have to protect against faults, but still allow ring voltage,
reverse battery, etc...

There are many providers of Telco protection equip.
A great many of them rely on MOV's or Sidactors.
The problem with MOV is they have a limited lifetime.
They work perfectly for a long time, show no signs of wear, and then
"poof!": One day they're "Dead & Gone". No warning whatsoever, so
they're hard to RELY on.

Gas discharge tubes also work, but they are slow (and probably would
not have helped you.) Actually, I wouldn't be surprised to see the
phone company replacing a bunch of these!

Active circuits are going to void any chance of continued compliance
with FCC Rules Part-68, unless you want to go through the testing &
certification procedures.

You can always just wind the tip & ring (22 ga) around a pencil. Cheap
& dirty but it does help.
10 turns or so ought to work wonders. (For lightning frequencies, not
too sure about 60 cycles)

From what we see on TV, if "dead" modems are your biggest worry right

now -- I'd consider myself "lucky"!

-mpm

 

[w_tom]

What happened to that protector installed free by your telco? A
protector that is defined by its earth ground connection. Was that
telco 'provided for free' protector earthed by same electrode also used
by all other incoming utilities? If not, then interior damage would
result; appliances inside building might instead become an electrical
return path - destructively.

 

[Jan Panteltje]

Well, I had some melted 22-gauge wire. A 500 ma. fuse would have at least
protected that wire.

Maybe not.
the _length_ of the fuse, that is how many volts it will stop before flash-over
is also an issue.
Lightning will jump a across meter distance if iit feels like it, seen that.

 

[Robert Scott]

You can always just wind the tip & ring (22 ga) around a pencil. Cheap
& dirty but it does help...

Actually I did have something like that. It was made of enamelled wire wound
around a ferrite torriod. I don't know if the inductance helped, but a short
did develop between tip and ring through the enamel, which welded the copper
wire together. But that short may have helped to protect a modem that was
downstream.


Robert Scott
Ypsilanti, Michigan

 

[Harry Dellamano]

I've designed some crowbar fuse blower circuits that circulated >100A
thru a 5A fuse... now you see it, now you don't ;-)

You have to be very careful with the foil pattern on the PCB to avoid
it being the fuse ;-)

...Jim Thompson
--

I've designed a 4A SMD fuse blower that circulated 150A and blew some SMD
fuses in 15uS. The others reformed into resistors and limited the current to
<4A. Go figure.
Harry

 

[Robert Scott]

On a related topic, I know gas discharge tubes exist for the purpose of
overvoltage protection, but I was wondering what could be done with an ordinary
air gap. I seem to recall hearing somewhere that no matter how close you place
electrodes in air, you cannot reduce the arcing voltage below a certain
theoretical limit that has something to do with the voltage required to ionize
air. (which is why they use gas discharge tubes). Can anyone validate that
concept for me?


Robert Scott
Ypsilanti, Michigan

 

[Andy I.]

Robert Scott wrote:
....

Rather than protect phone equipment piecemeal, I would like to protect all phone
equipement in my house using one circuit that I put at the entrance.
....

Robert Scott
Ypsilanti, Michigan

I've posted a schematics of a phone line protector in a.b.s.e,

Subject "phone line protector", Message-ID:
<[email protected]>


I've designed this to produce a very low impact on the ADSL signal.

The (constructive) feedback / remarks are very welcome.

Thanks.

-- Andy

 

[Robert Scott]

Robert Scott wrote:
...

I've posted a schematics of a phone line protector in a.b.s.e,

Subject "phone line protector", Message-ID:
<[email protected]>

Excuse my ignorance, but what is a.b.s.e. and how can I look at your schematic?


Robert Scott
Ypsilanti, Michigan

 

[Robert Scott]

sounds like a binaries newsgroups a.b. would be alt.binaries.,
perhaps the s.e. are sci.electronics?

OK, I found it. It is in alt.binaries.schematics.electronic

Thanks for the hint.


Robert Scott
Ypsilanti, Michigan

 

[AZ Nomad]

Excuse my ignorance, but what is a.b.s.e. and how can I look at your schematic?

sounds like a binaries newsgroups a.b. would be alt.binaries.,
perhaps the s.e. are sci.electronics?

 

[Andy I.]

Excuse my ignorance, but what is a.b.s.e. and how can I look at your schematic?


Robert Scott
Ypsilanti, Michigan

I should have expanded, sorry about that, I meant the
"alt.binaries.schematics.electronic" newsgroup.

A few comments:

- As with any such circuit, a good ground connection is fundamental, it
goes to the pin 2 of the CN1 connector. The other two pins of CN1 go to
the phone line, the home phone equipment is connected to CN2.

- One should probably aim at a higher voltage rating of the fuses F1 and F2.

-- Andy

 

[joseph2k]

In the recent ice storm, the 4800 volts from the power lines crossed over
into
the phone lines. It burned out many circuits in our area. It also fried
some
modems and phones in my house. Surprisingly, two connected modems
survived.
These were both protected by whatever surge protection circuit they put
in a
UPS. But it had the side effect of burning out that circuit in the UPS,
so I cannot use that again.

Rather than protect phone equipment piecemeal, I would like to protect all
phone
equipement in my house using one circuit that I put at the entrance. I
would
like to use a fuse. I gather that most commercial "surge protectors" do
not have fuses, but I think they might be a more reliable protection than
varistors
or whatever they put in those commercial units. Has anyone done this?
What
sort of current limit should I set? I guess the ringing with a maximum
number
of phones connected would be the upper limit. But a fuse might not be
enough.
What about a crowbar circuit? Is it possible to detect an overvoltage
condition with an active circuit and then short all lines to ground
through some fast-acting power FETs downstream of the fuses, just to make
sure the fuses blow
right away? I don't mind replacing some power FETs and fuses in my own
custom circuit once in a while if it means I won't have to replace modems
and phones. I know ice storms and downed power lines are my focus right
now, but I am also looking toward spring with lightning will be the major
risk.


Robert Scott
Ypsilanti, Michigan

Given your location you need at least two stage protection. First stage is
lightning type telephone line protectors. The second stage should be what
is called a 5-pin (replaceable) protector. You should be able to find them
easily with Google.

 

[w_tom]

Given your location you need at least two stage protection. First stage is
lightning type telephone line protectors. The second stage should be what
is called a 5-pin (replaceable) protector. You should be able to find them
easily with Google.

Protectors are not protection. Shunt mode protectors become
conductive only during a transient. They do not stop or absorb
surges. They make a temporary connection (a shunt) to protection.

Reviewing the schemtic posted by Andy, those fuses also do nothing
useful for a 4800 volt surge. Fuses are only rated for 600 volts -
remain conductive when trying to stop many thousands of volts. So
what provides protection? A protector that connects to earth; a shunt
to earth so that a surge need not find earthing connection via
household appliances.

What is two stage protection? A system that has layers of
earthing. Same exists with AC electric. 'Whole house' protector
installed in a mains box makes a short connection to building's
earthing electrode. That is secondary protection. Primary protector
provided by the utility is demonstrated in pictures in:
http://www.tvtower.com/fpl.html

Meanwhile, the telco installs a 'whole house' type protector on all
subscriber lines where their wires meet a homeowner's. Again, this
protector is only as effective as its short ('less than 10 foot')
connection to a single point earthing electrode. A protector
installed for free because it is so effective and so inexpensive. A
protector often unknown to those who somehow want to stop or block
surges (ie those fuses). Surges are not stopped or blocked. Anything
that would do that stopping is already inside phone appliances.
Surges must be diverted to what surges seek. Either a surge is
earthed before it can enter a building OR surge will seek earth ground
destructively via household appliances. So that protection already
inside appliances is not overwhelmed, the effective protector makes
that short connection to earth.

Telco install that device, for free, in the NID. A protector that
is only as effective as earthing provided by the homeowner.