Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Jerry Turnbow]

Another thing to consider, and this may be ancient history by now, but some of the older solid-state power amps didn't deal very well with the fact that a constant-voltage system is a more reactive load, with all those transformer primaries in parallel.

Bob Lee might be a good resource on this, but I remember in the old days blowing up some solid-state amps because they would go into HF oscillation on such loads.

It might be worthwhile to throw a scope on the line and watch for any spurious oscillation that't not related to the signal.

I suspect that the newer Class D stuff would be immune from this sort of thing, though.

Just a thought.

[Jerry Turnbow]

Jeff -

I spent 17 years with a large systems contractor here in St. Louis, and we did a lot of constant-voltage sound system installations - 25 volt systems for educational and healthcare, 70 volt systems for commercial/industrial/retail, and even 200 volt systems for large places like racetracks and airplane factories.  That being said, I'd be happy to share my $ .02 with you.  I'll apologize up front if I'm telling you things you already know.

In a constant-voltage system, the rated voltage is the maximum voltage that the amplifier can cleanly produce at the rated load.  Bear in mind, "clean" audio for PA/background music probably is around 2-5% THD, and the dynamic range of this material isn't nearly as wide as concert audio or any type of live performance.  If you're feeding this system with live music, you're going to want a good quality limiter feeding the input to ensure that you don't overdrive the system beyond the 25/70/200 volts that it can produce.

Headroom is achieved by tapping speakers (selecting the appropriate wattage secondary winding on the speaker) to provide a good level when nominal audio signals are played, and then ensuring that you never exceed the rated voltage on the primary sides of the speakers/amplifier output. In no case do you want to put more than the rated voltage into the primaries of the transformers;  it may or may not hurt the speakers, but would most likely damage the transformers themselves.  (I've seen many of them fried over the years.)

Any of today's commercial ampflifiers should be able to run the full rated load with no problem;  there is no derating for transformer loss.  The transformer just causes some of that power to be turned into heat instead of sound.  Of course, it never hurts to run any amp below it's rated capacity.

One common problem with constant-voltage sound systems is when someone has improperly wired an L-Pad to control a local speaker.  The most common error is that it's wired so that it puts a dead short on the secondary of the transformer, which reflects right back to the primary/amplifier output side.  On some occasions, I've seen them improperly wired across the primary side itself. These can often be difficult to locate, as the users may change the setting before you come out to troubleshoot, so check them for proper operation and wiring.

Also, check for non-transformer coupled speakers that someone may have added to the system - they can be loads of fun.

If you're going to do much work with constant-voltage systems, I'd highly recommend you invest in a commercial Z-meter, which measures the impedance of the load and is calibrated directly in wattage ratings for ease of use.  The "easy" math for 70.7 volt systems is to remember that power = voltage squared/impedance, and conveniently, 70.7 squared is 5000.  So a 500 watt 70 volt amplifier doesn't want to see less than a 10 ohm load.  The Z-meter will "see" through the transformers and show you the true load, helping to find those shorts, bad speakers, and miswired controls, as well as knowing the real load on the system.

I know the McMartin and Toa both used to make these;  I've been out of that business for 15 years, so I'm sure there are others by now.

Hope this helps!

[Bob Lee (QSC)]

Jerry Turnbow wrote on Tue, 18 September 2007 12:04

The "easy" math for 70.7 volt systems is to remember that power = voltage squared/impedance, and conveniently, 70.7 squared is 5000.

I learned that same shortcut from Ted Uzzle.

[Bob Lee (QSC)]

Perhaps even more important than load reactance is that if the speaker transformers saturate, they may put a voltage spike on the line when the cores un-saturate.

This phenomenon tends to happen mostly on lightly loaded lines, and it used to frequently take out some older direct-drive 70-volt solid-state amps that either didn't have flyback diodes to absorb the spike or had inadequate ones.

[Jerry Turnbow]

Bob Lee (QSC) wrote on Tue, 18 September 2007 22:44

Perhaps even more important than load reactance is that if the speaker transformers saturate, they may put a voltage spike on the line when the cores un-saturate. This phenomenon tends to happen mostly on lightly loaded lines, and it used to frequently take out some older direct-drive 70-volt solid-state amps that either didn't have flyback diodes to absorb the spike or had inadequate ones.

That's why the old tube-type paging/intercom systems had a relay contact that actually shunted the output transformer of the all-call amp when the system was idle.  That flyback voltage could actually damage those amps!  Not to mention give the service guy a good jolt if he happened to be across the terminals when the page was completed.  (I blew up several early JBL and McMartin solid state amps before I found that shunting relay!)

(Yeah, I'm that old!)

[Jeff Heart]

That definitely helped some.  Thanks.  Also, I just noticed after revisiting the specs for the Crown CH4, that it is not rated at 500 watts at 70v.  Rather it is rated for 70volt mode to deliver 1200watts per channel.  4.16ohms minimum based on the calculation method you gave me.
 The channel the speakers are on which had the problem is loaded down to approx. 11.9ohm (9 speakers - 5 at 60watts and 4 at 30 watts = 420 watts)
  The amp is ran after a DBX Driverack with limiting in place that is successfully protecting the amp from going into clipping and a crossover is in place with a high pass set to somewhere above 90hz.  I wired the system so I can personally assure proper cabling (with 16 awg wire and no more than 75 feet) and can rest assured there are no other speakers on the line.
 This leads me to believe it is something other than the amp reaching it's limit.  Hmmm, maybe I tapped one of the speakers at 8ohms and didnt even notice it.  It has a selector switch so it's possible I turned it too far.  Very doubtful but I will check.  They havent had the issue since that first time but I won't assume that it is not going to happen again.  It's bound to resurface sooner or later unless they just have no clue what they were talking about.  As far as I know, it could have even been a problem in the DJ's crappy MP3 burnt discs.  Maybe some bad recording in the one channel.

[Jeff Heart]

I was at this venue the other day when the 70v speakers did indeed reveal an issue just as the owner said.  It wasn't quite like he described but what is happening is that all the speakers on channel 1 lose all the top end, sounds very muffled, and they get real low in volume.  Every few seconds, they exhibit a popping sound as they cut in and out, sounding like static as if someone was taking either the plus or minus connection and repeatedly rubbing it against the speaker contact and then removing it.    It happened twice the other day at low volume so I now know it wasn't the amp clipping.  It's not the dj's recordings as it happened on a a few input sources.

One thing I did notice was that the driverack 260 had the outputs to the 70v amp crossed over at 44Hz - 18Khz.  I had it set at 70+ Hz when the system was first tuned so apparently someone messed with it.  Could that be the culprit?  Does the problem described above sound like a transformer saturation problem?  I know they don't like frequencies below 70hz too much.

Also, on Channel 1 there are 7 speakers.  5 of which are fed directly from the amps channel 1 output and the other 2 are fed from the same channel but are first fed through a 100w Atlas AT-100 attenuator.  Those two speakers are tapped the 30watts each.

My best guess at this point is that it was saturation from the lower frequencies.  My second guess is that perhaps there is a shorting issue with the attenuator or perhaps on the feed to the 5 other speakers. The issue effects all speakers on the channel.

Only thing that wouldn't make sense to me as far as the saturation issue is concerned, is that channel two was fine.  Both channels were receiving the 44Hz - 18khz passband so I would  think all the speakers should have had the problem if saturation were the case?   I don't have enough experience with 70v speakers to know what saturation issues sound like.

Since it is intermittent, it's hard to trouble shoot as it doesn't last long enough for me to run tests.  It did it for about 5 minutes the other day and then once again for less than a minute a few hours later.

Since the 70v system runs 80% of the bar/club, it is the primary sound source and if the amp dies, it's not going to be pretty.  So I am a bit concerend about finding the issue instead of sitting back and saying, "well, it seems to be gone now".  I'm wondering if the problem is a pre-warning sign to some approaching full failure. I'm thinking about removing the x-over and letting all frequencies through at low volume to see if it recreates the issue.  If it does, then I know it was only the low frequency issue and I can rest easy.  However, can that test damage the amp or speakers for using intentional saturation or is that an OK test as long as I do for a few seconds only at low volume?

I haven't heard the issue since I changed the highpass crossover back to 70Hz and that was three days ago.

[Ken Freeman]

Bob Lee (QSC) wrote on Tue, 18 September 2007 14:44

Perhaps even more important than load reactance is that if the speaker transformers saturate, they may put a voltage spike on the line when the cores un-saturate. This phenomenon tends to happen mostly on lightly loaded lines, and it used to frequently take out some older direct-drive 70-volt solid-state amps that either didn't have flyback diodes to absorb the spike or had inadequate ones.

Yes, as Master Bob has noted, it can get a little squirrelly if you don't give the amplifier something to drive.  I have always found that putting a 1000 feet of 14 gauge cable (about a 4 ohm load) seems to settle things down.  Due to the odd venues that I end up in, I often build these systems while they are hot so we can check them as we go...Then just remove the dummy load and continue to the next system or branch.

Ken

[Tim McCulloch]

Lock down the DriveRack.  I had a club install where they kept blowing up 15s.  The rig was for syndicated foreground music.  They kept it pretty loud, but not pushing it hard enough to clip, or even make it sweat hard...  it turned out the cleaning crew would crank the system far louder than the bar crowd would tolerate.  All the way up...

They shouldn't have been anywhere the controls were located, but they did anyway.  We put in a locking rack and disabled the remotes behind the bar... the maintainence issues went away.

Someone thinks those little speakers can do car stereo bass.  Lock down the DR or call a clergyman for an exorcism if the lockout doesn't work.

Have fun, good luck.

Tim Mc

[Jeff Heart]

Ken Freeman wrote on Sun, 23 September 2007 19:38

Yes, as Master Bob has noted, it can get a little squirrelly if you don't give the amplifier something to drive.  I have always found that putting a 1000 feet of 14 gauge cable (about a 4 ohm load) seems to settle things down.  Due to the odd venues that I end up in, I often build these systems while they are hot so we can check them as we go...Then just remove the dummy load and continue to the next system or branch.

I'm a bit confused by your response.  Is "yes" answering my question about damaging the amp if I do the saturation test, or is it "yes" about the 40 - 70 Hz bands being able to cause the problem,  OR about something earlier in the post?
  The amp is definitely driving something....7 speakers on channel 1 which is perhaps a light load? additively 360 watts.
The cable is 16 gauge with a total length of approx. 250ft to all speakers on the channel (250 ft. total, NOT 250 between each speaker).  So, does it seem possbible to exhibit the problem mentioned with that scenerio due to saturation?  THe amp is rated at 1200watts per channel in 70v.    360watts would indeed seem light.  IF that were the case, I could move over some speakers from channel 2 but then channel 2 would be light.  Both channels are already "light" so I am assuming that is not the issue.
Ken