paging system problem

Hi Lenny...

I suspect that you're confusing resistance with impedance... :)

The dc resistance that you're reading on your speaker lines is just the resistance of the wiring to those speakers (transformers), because after all it is a transformer, so the dc resistance should be virtually zero ohms, right?

The math is indeed simple, divide your 70 watts by the total of all of the tap wattages - add a little margin - and if it fits you should be OK.

If that works out, then afraid you're gonna have to check each and every transformer primary individually - perhaps by sub'ing - or maybe just disconnecting the primary looking for a large change in level - until you find one that's gone shorted, or until you find shoddy workmanship at a transformer or a splice that's a dead short.

Wear your oldest clothes, cause you're gonna be crawling around in an awful lot of really dirty dusty crawl spaces. Did it myself 20 years ago in a hospital, and wouldn't like to do it again.

Take care.

Ken

Has either amp got a slaving input ? If so, run the low resistance leg with a lot of speakers on it, off one of them on its own. You could also try calling the manufacturers of the new amp just to make sure that you are interpreting the specs correctly. Also, if all of the speaker / transformer combinations are pretty much the same, you could try disconnecting a couple of examples on each leg and measuring their primaries, then counting up how many individual stations there are on each leg, and just working out the math to see if the readings you are getting back at the amp are about right for that number of primaries in parallel. That way you would know if you have a bad one for sure, but in any case, if you did, from what I can remember, it having been some years since I was last involved in this stuff, a bad station usually produces low output, whilst all of the others continue to function.

The primary of a 70v line transformer should not have a resistance of close to zero. One that I just looked up, for instance, was a 100v line type for 2 watts into 8 ohms, and its quoted primary resistance (not impedance) was 320 ohms. I would expect 70v line trannies to be broadly similar. That would lead me to believe, as the primaries of these TXs are quoted by resistance and not impedance, that the figure of ~10 ohms on the back of the amp, is a DC load resistance figure. With a primary resistance figure of 320 ohms, that would allow for 30 of these TXs in parallel, to get down to your 10 ohms, and with each one 'stealing' 2 watts off the line for its speaker, that would take about 70 watts off the amplifier.

Now I know that's very rough 'back of a cigarette packet' math that I'm talking there, with no TX efficiency or line losses factored in, and that we're mixing 70v and 100v systems in the discussion, but that sounds like a pretty typical installation to me. I wonder if your low resistance "pretty much the rest of the building" leg at ~ 5 ohms, has around 50 - 60 speakers on it ? If it does, then it's probably about right. If not, then you need to do the primary disconnect and measure that I said above.

Just as a matter of interest, did the original system *ever* work right, or has it just gone faulty ? If the original amp was only 35 watts, sounds like it was under-rated for the job, or has become so as a result of more and more stations being added over the years.

Arfa

It did work at one point. They couldhave added things. They say they didn't but who knows. I'm going to look at it again on Monday. Thanks to everyone for all the advice. i really appreciate it. Lenny.

text -

Since the one leg won't work by itself, I would look for a short somewhere on that leg. Probably break it at midpoint, and see what happens, or maybe run a new wire pair to the midpoint and then try feeding each half to see what happens., With a long run, and a short at at unknown location, it will be a problem, but I bet a short is what the problem is.

H. R. (Bob) Hofmann

That doesn't mean that some employee didn't tamper with it because it was too loud, or not loud enough where they worked. The so called 'maintenance man where I worked kept screwing around, till the system was useless. I started at the amp, on the leg with the most problems, finding lots of miswired speakers. The volume was low, so the idiot changed the taps from .625 W to 5 W on speaker after speaker, making the problems worse. I was almost to the end of that line when i hit the production floor. Someone had moved a speaker, and extended the wire. They had shorted out the trunk line. When I pulled the tape off to take a look, the volume went way up, and one speaker cabinet caught on fire. Some idiot had removed the line transformer and connected the trunk line across the 20 ohm fader control. It went up in flames, and set the voice coil on fire.

You never know what you'll find. I had the same problem in a high school. The intercom worked fine, except to two speakers. but the all call was low and distorted. Someone had removed the transformer form one of the speakers, and that upset the entire system. The other speaker had an open pair back to the office, and the building had been wired with very early shielded pair cable, with no outer jacket. It was old and so hard that over half the building would have had to be rewired to fix the problem.

--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

When I was installing new systems, I tried to put no more than five speakers on each run. A 'Bud minibox' with a row of toggle switches mounted near the amp, with a small monitor speaker and fader control allowed isolating trouble to a bad zone in a few minutes, then you could troubleshoot the bad zone while the rest of the system was in use. If you go this route, you need a layout of the building to mark each zone, and the speaker locations, for future reference. I always did that step when giving a bid.

shows the aluminum boxes I used.

--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

Thats a really good idea for the future. The guy who installed this nightmare did something like that. There is a home made relay box on the wall near the amp. I'm guessing that at one time they did some kind of zone paging however no one remembers ever doing this or where the control point for this ever was.In theory then there should be only one wire feeding the amp from this box, I mean thats what I would have done but who knows what other people could have added or done over the years. Sometimes we can get lucky by pulling relays thereby opening the line at various points and finding the bad area but not so this time. I think I'm going to go in at various points in the line with my ESR meter and look for what I'm guessing, (hoping) is a dead short somewhere. Hopefully I can jump in somewhere and work my way toward the lowest resistance. It's going to be hot and messy and I'm really not looking forward to this one week before my vacation starts but I just know that I will feel so much better if we can get this cluster &&&& behind us and then I can sip my margaritas on the beach without thinking about it. Lenny..

Hmmmm.... cool drinks on the beach... lovely ladies in bikini's...

Now I have one more suggestion, if I may...

In the interests of efficiency, I propose that you put off fixing this problem until after your vacation. And that you should invite all of us along to discuss the repair, thereby turning it into a business meeting with all the attendant tax advantages. I think that after a couple of weeks of cool drinks, lovely ladies, and sunny beach we'll have come up with a consensus, and you can get back to work.

All in favor? :)

Take care.

Ken

What you need if your working on distributed sounds systems on a regular basis is a impedance meter like a TOA ZM-104.

It will tell you the AC load to the amp in ohms as well as put a 1000 HZ tone on the line so you can audibly hear if the speakers are working or not.

Bob

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

The #1 Newsgroup Service in the World! 120,000+ Newsgroups

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

I don't know as an ESR meter is going to help you too much. The fact that these usually run at a number of tens of kHz to make their measurement, may well lead to some pretty misleading readings in a system full of so much lumped and distributed L and C. Better to use the low ohms range of a 'standard' multimeter, I think.

Arfa

That makes a lot of sense. I'll bring the DVM. Thanks, Lenny

Actually, the ESR meter will show a short better than an ohm meter, The inductance in the transformer primaries at 100 KHz will present a high impedance to the meter. The capacitance between parallel conductors can cause some problems, but you are well under a wavelength in any normal sound system. 100 KHz would have a wavelength of 3 kilometers. Any system that large would have to be broken into transformer isolated zones, and use multiple amplifiers.

If you can find both ends of a run and check both ends you can decide which end is closer to the problem. Another thing to watch out for is a trunk line shorted to the building's ground somewhere. This happens when the wire is run across drop tile ceilings and the insulation is damaged by the metal suspension grid. Another cause is using bare steel wire to hang bundles of cable, and some idiot twisting the steel wire so tight that it cut into the bundle of wires.

If you use DPDT toggle switches with center off in the box I described, you can create a "Test buss" to connect your meter, or even a second amp for troubleshooting. I like to use a terminal block to terminate all wiring at the amp, so changes can be made quickly. When possible, I used surplus 66 series telephone blocks and staked the wires. The bridging clips work as well as the switches, but are not as easy to use.

--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

The few building sound systems upon which I have worked all used telecom punch blocks; perhaps it was because the sound system was also integrated into the music-on-hold features of the pbx.

Regards,

Michael

A lot of the systems I serviced were installed before 1950, when the WE was still using molded bakelite blocks with solid brass hardware. Back them, they didn't sell to anyone outside their trade.

I modified some of the school intercom systems to 50 pin 'Amphenol blue ribbon' plugs, so each bank of 25 switches plugged into the wall. A pair of mating nine pin Molex plugs allowed you to remove and exchange a panel by removing four screw, and three plugs. Then you could use solder wick on the switch lugs to clean off the old solder, clip all eight buss wires on one side of the switch, and pull out the bad switch. It made repairs that normally took hours something that took under 15 minutes.

--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

Like it tried to tell you before, a DVM or regular ohm meter is of limited use on 70 volt systems. If your doing school or business distributed sound systems, get a audio impedance meter. It uses a 1000 hertz or so tone and translates that into a ohms on the line reading. With this, you can determine the actual amp load in watts and see if a line is actually shorted. A DVM will just show the resistance of the line transformers primary and is of limited use.

Toa makes one. MCM electronics did have one. Goldline sells one too.

Bob

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

The #1 Newsgroup Service in the World! 120,000+ Newsgroups

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Your theory is sound Michael, but I have been mislead in the past looking for low resistances on a 'dynamic' system using my ESR meter. Just as an experiment, I just tried measuring a roll of power cable that I had here, using my Bob Parker / Dick Smith ESR meter. There is about 80m of cable on the reel. I measured the open circuit capacitance, and came up with 11pf. I then shorted the far end, and measured the inductance, as it was on a roll. This came up at 0.032uH. So neither figure is particularly huge. However, when I measured the cable with a low ohms range on a standard digital multimeter, far-end short still in place, I got a figure of about 1 ohm. When I repeated with the ESR meter, I got 55 ohms. I know that this isn't really a 'fair' test with the cable being on a roll and so on, but I think that it does illustrate that the readings you get on an ESR meter when you are not right on top of the short, may not be what you are expecting. Just as a matter of interest, the 'test' voltage from the Bob Parker appears to be pulses of about 10uS duration, occuring every 5mS or so.

Arfa

messagenews: snipped-for-privacy@earthlink.net...

Well we made the dreaded service call today to try to resolve this problem. We brought a DVM, an ESR meter, and a used 100W amp which I checked on the bench last night. I confirmed that it produced a clean

90W RMS into an 8 ohm load. We were loaded for bear. We decided to forego all the troubleshooting and just connect up the replacement amp to all three trunks in parallel and see what happens. The total DC resistance of 20, 16, and 5.5 ohms in parallel without calculating it must have been down below 3 ohms. I remembered that we are dealing with impedance. I just never thought that the resistive component could be that low. to our surprise the system worked perfectly! I brought both amps, their original 35W amp as well as the 60W unit that I installed last week back to the shop for evaluation. I determined that their original 35W amp produced 21W into 8 ohms before clipping. My 70W amp could not even squeeze out 1 measly watt! Their amp became weak and something happened to mine at some point that I was not aware of. It seems that after all this effort I'm embarrased to say that there was apparently nothing wrong with the distribution system. It was normal component failure after all. I never suspected my replacement, which was on the shelf marked "good" to be no good. But confirming power output on the 100W amp last night was the ticket. But anyway now I'm really ready for that vacation. Thanks to everyone who took part in this discussion. It was very helpful and enjoyable and I really appreciated it. Lenny Stein, Barlen Electronics.

messagenews: snipped-for-privacy@earthlink.net...

The thing you have to remember about some distributed systems is they can be in a dynamic environment. Especially in a factory or such with temperature extremes or vibration. One day they can be fine, the next day the line could be shorted. A shorted line could overheat the amp and cause a component failure.

The nice thing about a audio z meter is you can read the load and there is a chart to tell you how many watts that load is into 25/70 volt line. There is on digital unit that will tell you the wattage load directly

Bob

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

The #1 Newsgroup Service in the World! 120,000+ Newsgroups

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Good that you seem to have gotten to the bottom of it. There's nothing worse than having a job like that hanging over you, is there, particularly when you have some event like a vacation coming up ? Have a good 'un !

Arfa