atcomsystems.ca/forum Forums General General Telephone & Systems Information Grounding of KSU's directly to the cabinet

Yes it is. Read my explanation above as to why the ground screw is there. If the receptacle is properly grounded the ground screw is only needed if the plug is pulled. Grounding the ground screw isn't going to be any better than you already have and can actually cause problems if not done the way I said- which isn't always the way the manufacturer recommends.

-Hal

Thanks Hal!

Well, yes and no. If the power cord from the AC supply goes directly into the KSU AND the grounding conductor is extended within the KSU to a metal KSU ground plane, then the system is "effectually" grounded via the electrical cord.

There are plenty of KSUs that have the power cord enter the KSU cabinet and then go to a transformer and the grounding conductor is connected to a piece of copper or aluminum "foil" square gluded to the KSU plastic housing. There is no way this is an effective ground.

Then there are the system with external power supplies that require a ground be externally connected.

Here where it really comes down to a point. The CO telephone cable head ground and the KSU ground should be exactly the same, provided the building ground at the main service panel and the telco cable head ground are less than .1 volt difference.

IT people are much too cavalier when it comes to grounds. They think a cabinet ground is sufficient if the cabinet is grounded to the "floating" rack. This is ridiculous.

When I was at Bell, it was manditory that the ground for the cable head, PBX and all ancillary equipment be identical. When the 1978 NEC came out, we were required to make sure the building ground and our ground were identical and it has been that way ever since.

But, of course, the Bell system never actually adopted the NEC as a standard. The IEEE National Electrical Safety Code was the actual "Standard" that the Bell System adopted as the official code. This may be recognized by all the OSP people as the "Code Bible." Within that code, it is manditory that all sheath, ground lugs and grounding "devices" be bonded together and extended to all equipment served by such devices. Here is where it gets a little complicated. In large buildings where the primary electrical supply may be 480VAC or higher and individual floors are supplied through transformers, bonding becomes a nightmare, both for the electrician and the telephone company. Here's where impedance mismatches can cause more harm than not having a ground at all. That is why one may see a 1/0 ground cable running to all floors from the teleco cable head in the large buildings to maintain a good cable ground.

Rcaman

I agree that that doesn't constitute an effective ground but it's really poor design. What does the ground screw connect to and why is it not tied to the power cord ground like it's supposed to be? Same with external power supplies- there either should be a ground conductor in the interconnecting cable or the power cord ground gets connected to the negative of the supply voltages along with all metal parts. Are the two grounds not connected together because the manufacturer found out that there may be a damaging difference of potential due to how they (improperly) recommend the external ground be provided? I do understand what you are saying but I'm not sure this separated ground problem exists today with newer systems- at least it shouldn't. Easy way to find out is a simple continuity test between the ground prong and ground screw.

-Hal

Hal,

The ESI uses a 2 wire power plug from the electrical receptacle to an external transformer and then a two wire coax power connector to the KSU. There is NO physical connection between the electrical receptacle and KSU. The ground stud of the KSU is internally connected to the metal hardware of the KSU and the ground conductor of the CO ports. I am sure ESI is not the only system like this.

Rcaman

I posted this because plenty of Partner proccessors, some Merlin KSU,s AT&T Spirit, Vodavi DHS, Vodavi STS, Panasonic KX-T etc. etc. seem to get hit by some type of surge, then go belly up. I thought this grounding redundancy may help alleviate such misfortune for the customer.

Grounding (as some have described) is essential and defines hardware protection. But what some have posted does not completely describe what you must understand to have a useful answer.

For example, do not ground equipment to earth a surge. That simply makes equipment a best and destructive current path. Equipment must be bonded. But a surge current path, to avert damage, must be on a path that goes to earth AND not on a path into or near equipment. That is one reason why electronic ground and chassis ground remain electricaly separate (therefore dfferent) and only interconnected at a single point.

Do not assume both ends of a wire are electrically same. For example, safety ground in a mains breaker box is electrically different from earth ground. Earth ground is essential for protecting electronics. Safety ground in a wall receptacle will not properly connect to earth (ie too much impedance) and can even make damage to nearby equipment easier.

Do not confuse bonding with a low impedance connection to earth. That low impedance path must connect destructive currents to earth so that current does not enter a building. This rule applies to every incoming wire - not just telco cables.

Damage apparently exists because a current has connected to earth destructiveely via a KSU.

There is another fact that may be overlooked: There is no device, ground or method of grounding that can withstand a direct lightning strike. With a potential for a billion volts, the puny grounds and surge devices are incapable of containing a direct strike. Telephone equipment is exposed three ways. 1 through the electrical service, 2 through the telephone CO trunks and 3 via static through the telephone instrument cabeling. Many of the "fried" systems may be a victim of static electric which traveling through the telephone instrument wiring, in winter months, is extremely likely and can have potentials high enough to cause lightning like damage, but on a much smaller and less dramatic scale.

The bottom line here is make absolutely sure the telephone equipment is bonded and NOT the "best" ground in the facility. Take accurate readings and make sure the telephone equipment bonds are not mis-matched impedance wise from the good ground. If you are in a smaller facility, take the time and make sure there is a solid metallic ground with a bonding jumper across the water meter. In large, transformer fed facilities, make sure your telephone ground is the building main ground and not a high impedance electrical ground being derrived from a step down service transformer. The NEC and the NESC both require large buildings to maintain structural steel bonding. Don't be fooled. Take readings. Not only is this an equipment protection situation, it is also and, more importantly, a life safety issue.

Rcaman

ATT TP -76416 has some usefull stuff in it.

The many who are educated by hearsay and advertising only assume. Fewer who actually do this stuff use real world numbers. And learn from over 100 years of well proven experience and science.

Your telco's switching computer is connected to wires all over town. It suffers about 100 surges with each thunderstorm. How often is your town without phone service for four days while they replace their $multi-million computer? Never? Because direct lightning strikes without damage is routine ... when someone does not use wild speculation. And when someone does not use plug-in (adjacent) protectors that can sometimes make damage easier.

The engineer who did this stuf warns all of the naive who make 'woe is me' claims. Direct lightning strikes to electornics atop the Empire State Building without damage - 23 times annually. Direct strikes to electronics upon the WTC were 40 per year without damage. Only those who learn science(and not entertain hearsay) know how to spend less money on superior solutions ... so that even direct lightning strikes need not cause damage.

Relevant numbers means voltage (ie billions) is irrelevant and mythical. A lightning strike is about current - typically 20,000 amps per surge. A less expensive and supieror 'whole house' protector start at 50,000 amps. So that a direct lightning strike does not even damage the protector. As demonstrrated by telcos that suffer about 100 surges per storm - without damage. As demosnstrated by 23 direct strikes to Empire State Building electronics without damage.

Anyone who posts by first learning well proven science would know these numbers. The naive, educated by hearsay, only speculate about billions of volts ... wildly and without responsiblity.

This is about why scams work and that informed consumers learn from engineers who did this stuff. Direct strikes to cell phone towners without damage is routine. Using princples demonstrated even by Ben Fralnklin over 250 years ago. Informed homeowners are strongly advised to learn from engineers who recommend better earthing and a 'whole house' protector rated at least 50,000 amps. A superior solution even sold in Lowes and Home Depot. The superior solution is that easy.

Protecting a KSU is about how every wire insiide every incomign cable enters the building. And about the single point earth ground. Not about bionding. Protection is about a low impedance (e 'less than 10 foot') connection from each incoming wire to that essential (and best) earth ground.