Astraphobe DSL lightning protection - has anyone tried it?

Out of interest, we had access via the SA Weather Service, from the weather radar located in Irene. The lab location was just 4000 metres from the radar site, I can try post some pics then you can see what some typical storms look like from a birds eye view. Its quite scary / impressive.

yes please :)
 
OK added some pics to the product website, http://www.astraphobe.com/pics.html. First pic shows for interest where the development lab was relative to the radar station. Just under 4km away meant we had great, real time radar imagery available. The radar images are from SA Weather Service (SAWS). For interest, these pictures are normally used by pilots, this is part of the aviation weather service offered by SAWS. The 2nd pic down shows storms extending to the south west for some 200km in April 2013. Then for fun I added another 3 pics, showing the movement of storms together with the forecast position.
 
Thanks, Jorgen. Welcome. Your clarifications are most welcome.

Are you planning any multi-line models?

I have three phone lines coming in to the property. One is dsl, but the other two have cordless phones and other electronics attached.
 
This is seemingly impressive; I was looking at this just the other day when ordering some new networking equipment R3k worth, but there wasn't enough reviews for me to justify splashing out on it. More comparative information could be useful.

Maybe Jorgen can highlight the difference between the Astraphobe and the Ellies Telephone Surge Protector Combo ? I'm really interested in knowing whether these two products basically do the same thing ? (My Ellies Telephone Surge Protector does add noise to my line attenuation goes up and SNR goes down when plugged in, so there's that I'll concede).
 
My understanding (or presumption, rather) is that Jorgen's Astraphobe actually performs a physical circuit disconnect through some sort of mechanical contraption inside the box ... in other words, the cable is effectively unplugged physically. I imagine some sort of solenoid-activated switch...

Is this so?
 
@ Arthur: Yes on multi line, we actually designed it in a way that in the future we can switch more than a single pair of wires. Adds a bit of internal complexity, and we do have this in the pipeline, but figured that most demand will be for typical home type setup. The biggest issue with multiline will be the test & licensing, I do not foresee any tech issues other than getting the engineering done.

@ Alexcon and Arthur: The Ellies product you linked and the Astraphobe take a very different approach. I am not going to comment specifically on the actual product you mention, but in general terms. Surge protectors work by having a Metal Oxide Varistor (MOV)connected to the line, and to ground (via 3 pin plug). At low voltages, they conduct but poorly, so allow a small current through. Alexcon, that is why you notice attenuation and SNR change. At high voltages (surge) they conduct well allowing higher currents through. You can read http://en.wikipedia.org/wiki/Varistor for more info, as it does get somewhat more technical.

The disadvantage with using MOVs, is that the response time may not be fast enough, and it could still allow a damaging surge through, or it could work 100% in the case of the first surge, but the surge actually degrades or destroys it. After that you no longer protected, as the active component inside is no longer working. This can all happen in one storm, and most surge protectors do not provide indication that the MOV is no longer 100%. The ones that do can cost thousands of Rand, and are a lot more complex.

The Astraphobe takes a completely different approach - it actively monitors the environment, and if a storm gets close to you, it disconnects the line, thereby preventing any surges from passing through. It is the same as unplugging the line. Arthur your guess is pretty close, but there is no solenoid as they do not open far enough and lightning could still arc across. Instead we use a motor to drive a custom designed internal switch that opens the contacts about 5 inches. Unfortunately this approach does cost more, because instead of a few passive components, we have computers inside, radio, LCD display, motors etc. And its all quality stuff, for example we use gold plated spring loaded pins on switch contacts. These are MIL SPEC imported from the USA, rated for 1,000,000 cycles.

Hope that clears up the differences!
 
pretty close, but there is no solenoid as they do not open far enough and lightning could still arc across. Instead we use a motor to drive a custom designed internal switch that opens the contacts about 5 inches. Unfortunately this approach does cost more,

5 inches... ah... I was wondering about that
 
5 inches... ah... I was wondering about that
Other products exist that do something similar - with some key differences.

First, let's back up a bit. Phone line protectors must not use MOVs. MOVs have too much capacitance and other problems. Protectors for phone lines (for example, found standard in all telco switching centers where damage does not happen) use other technologies. BTW, all protectors (even the slowest GDT type) are more than fast enough.

Second, surges that damage DSL equipment more often are incoming on AC mains. Remember, this is electricity. That means both an incoming and an outgoing path must exist ... simultaneously. The most common path is incoming on AC mains. Then outgoing to earth via the DSL (phone) line. Damage is often on the outgoing path - the DSL and ethernet connections.

Third, others do same with a relay that disconnects long enough to avert the surge but short enough that phone conversations are not interrupted. These feature something always so essential for surge protection - a low impedance connection to earth. Jorge means this. Improper earthing means any one wire in any utility wire without that proper earthing compromises the entire protection 'system'. In the competitor's system, it does not just disconnect phone line. It must also earth the incoming phone line. Only then does a typically surge not use other insulators (that are also electrical conductors) to get into the house. And again, any ongoing phone call is not disconnected.

Finally, this 5 inch disconnection for many minutes is only one part of what must be a complete protection system. Telcos use the other technology proven by over 100 years of science and experience. Key to both solutions is every incoming wire with some low impedance (ie 'less than 3 meter') connection to single point earth ground. Best connection is hardwire (ie from a coax cable) to
 
Last edited:
Westom agree on with you on your post with technical statements, but not on your comment regarding surges being mostly incoming on AC mains.

Practical experience across many installations, my own experience over 20+ years with dialup then ISDN then ADSL, and over 2 full storm seasons on the Highveld while we were doing R&D for the Astraphobe suggest that most damage occurs from surges originating on the DSL line. My personal experience living in Centurion confirms this, my house had 3 phase power, with surge protectors in the main DB board outside the house. The expensive stuff that shows you its working. Still lots of problems with surges, every time without fail, the entry point would be the DSL line. Most of the time the damage would be the router and a switch, but other times there was extensive damage. On one occasion Telkom had to replace the cabling to my house, the wires were burnt. That was fun (not), as the pipe was blocked and had to dig up the $%^$%^& garden! The last event fried my R4k router, 3 Ethernet switches, a 2 bay NAS drive, MVIX media player, wife notebook, my PC motherboard, sons motherboard and other sundry bits. By unplugging the DSL line, and later with the Astraphobe installed, I never had a problem, even with massive lightning storms right overhead, neighbours losing equipment etc.

I also agree that if it does come in via mains and the outgoing path is the DSL line, then the router is toast. In this scenario however, unplugging the DSL line manually or having the Astraphobe do it for you would prevent damage.

An interesting sidenote: Telkom Labs in Pretoria did all the testing of the Astraphobe as the precursor for ICASA licensing, they also test all the telephone / DSL surge protectors there. Out of interest I asked the Telkom engineers if they have DSL and what do they do in a storm, they all told me they unplug their lines!
 
Out of interest I asked the Telkom engineers if they have DSL and what do they do in a storm, they all told me they unplug their lines!
Then those Telkom engineers have no idea how their switching computers suffer, on average, 100 surges with each storm and no damage. Do they disconnect phone service from all over town before each storm? Of course not. Then the town must be without phone service for four days while they replace that computer? Of course not. Disconnecting is unnecessary when using the solution proven more than 100 years ago. And disconnecting only works when a surge can be predicted - ie an approaching storm. Not all surges are predictable as in that example. So the surge protection always found in every phone service must be effective even without disconnecting.

Your fried DSL line is classic of a surge incoming on AC mains. It used a phone line as the outgoing connection to earth. Damage is often on the outgoing path.

Long before DSL (a 1981 technology), modems were damaged. Classic damaged semiconductor in a modem due to surges was incoming on AC mains via a PNP transistor that controlled an off-hook relay. Incoming on AC mains. Outgoing to earth via a modem via its PNP transistor, from relay coil to wiper, and then to earth via the phone line. We did not just replace defective parts. We literally had to explain each surge - in some cases a review board. Replace defective semiconductors to make things work. And solved on an incoming surge path so that no more damage could result. Most damage was due to human failure - defective earthing.

Telcos all over the world protect their computers and other equipment by earthing. If any one wire enters without that required earthing, the all protection is compromised. Supplemental protection can be achieved by filters (ie a bulkhead plate) and disconnecting. But these work by encouraging a surge to use an earthing connection that exists before a filter or disconnecting point. Surges will easily cross a 12 centimeter gap if it does not have some other and better path to earth.

Early 1900 Hams learned this. They would disconnect their antenna and put the antenna lead inside a mason jar. Still suffer damage. Damage stopped when the antenna lead was both disconnected and then connected to earth.

Most important in all protection is that a surge must not even enter the structure. That means a surge must be connected to earth and harmlessly dissipated BEFORE it enters. Because buildings inside have numerous overlooked electrical conductors. Even linoleum tile is an electrical conductor.
 
Last edited:
Actually those Telkom engineers I am sure understand what goes on in their switching centres, but they don't live there.
 
Actually those Telkom engineers I am sure understand what goes on in their switching centres, but they don't live there.
Telecom engineers all over the world understand this stuff. As also true of rocket launching facilities and skyscrapers. An AT&T forum discusses how surge protection for DSL (and everything else) works:
Surge protection for DSL and dialup service.

Surge protection takes on many forms, but always involves the following components: Grounding bonding and surge protectors. ...

Grounding is required to provide the surge protector with a path to dump the excess energy to earth. A proper ground system is a mandatory requirement of surge protection. Without a proper ground, a surge protector has no way to disburse the excess energy and will fail to protect downstream equipment.

Bonding is required to electrically connect together the various grounds of the services entering the premises. Without bonding, a surge may still enter a premise after firing over a surge protector, which will attempt to pass the excess energy to its ground with any additional energy that the services surge protector ground cannot instantly handle, traveling into and through protected equipment, damaging that equipment in the process. ...

Now, if all the various service entrance grounds are bonded together there are no additional paths to ground through the premise. Even if all of the grounds cannot instantly absorb the energy, the lack of additional paths to ground through the premise prevents the excess energy from seeking out any additional grounds through that premise and the electronic equipment within. As such, the excess energy remains in the ground system until dissipated, sparing the protected equipment from damage. ...

By far, the whole house hardwired surge protectors provide the best protection. When a whole house primary surge protector is installed at the service entrance, it will provide a solid first line of defense against surges which enter from the power company's service entrance feed. These types of protectors can absorb/pass considerably more energy than any other type of protector, and if one does catastrophically fail, it will not typically be in a living space. ...

Plug in strip protectors are, at best, a compromise. At worst, they may cause more damage than they prevent. While they may do an acceptable job of handling hot to neutral surges, they do a poor job of handling any surge that must be passed to ground. ...

Then, to add insult to injury, some strip protectors add Telco and/or LAN surge protection within the same device, trying to be an all-in-one sale. Remember bonding? When Telco or LAN protection is added to a strip protector, if the premise ground, which is not designed to handle surges, cannot handle all of the energy, guess where that excess energy seeks out the additional grounds? You got it! The Telco and LAN connections now becomes the path, with disastrous results to those devices. ...
 
Hi i purchased this unit 22 december 2015. I have not disconected the phone line since and each time a storm comes near the unit disconects the line its self no problems yet.
 
Unfortunately the area where i live will not get fibre for quite a while 2meg adsl is the fastest in the area.

I lost modem, router, cordless phone and AVR last night :( strikes out of nowhere.
Fibre coming next month.
 
Where does the Astrophobe get its information and know when to (dis)connect? (I'm interested in knowing it's AI/reliability to disconnect/reconnect at appropriate times). Does it discover ambient static / noise caused by the lightning as light crackle induced into the DSL line, or does it connect to a weather service for radar information, via the network?

Something tells me a fibre isolator between the modem and switch, is still a better (and probably cheaper, if done properly) method to protect your network gear from storms striking the DSL line... then you probably won't even need to disconnect at all! The only challenge I've had with this, is getting power to the router without providing it with another path to earth. Thought of solar power, but haven't tried this yet. Would be cool if DSL modems pulled a low enough current that you could then tap the necessary power from the telco line using a "PoE-injector"-like passthrough to bridge rectifier/regulator... but I think that's too wishful to be a practical solution to the problem, and might even be illegal (lol)
 
Last edited:
Top
Sign up to the MyBroadband newsletter
X