BillFitzmaurice.info

sine143 wrote:shouldnt be using sine sweeps into the tweeter range. they take much less power than the woofer, so if you were too sweep with anything higher than 50 watts with a sine I wouldnt be surprisded by distortion due to its low crest factor.

Doubt I was putting anything like 50W into them... probably more like 1. Maybe with sine sweeps, you hit a few resonant freqs, as I did when testing the Lab12 with anything above 12V. The speakers sound good, but I never tested the tweets before soldering them up as I didn't really know what I was looking for (and I was getting impatient to hear my creations!) May well test with the good array and turn them up a little higher, to see if I can get a similar noise out of them. If so, then it is probably just the nature of the tweets.

I've just done the sine sweep through my other new cab array and that has resonances. Only at low power too. It doesn't appear to affect the musicality, again at low volumes, so am gonna run with it for now. I think the sine waves are a harsh test, very pure, doesn't need much power to be ear splitting, will exite anything with a resonance. There is probably a very real problem that would be obvious if it was bad enough to affect the music. Similar to your conclusion, thats my best guess for now.

Thanks Charles. Very much appreciated for testing on your rig

As I said before, I did notice a similar thing on the lab 12s when running them just over the 10V and close to Fs (I think that is the T-S name for it (still learning!!) - I mean the natural resonance of the speaker). At first I thought I had a dud... Needn't have worried .

Oh and an update on my digital impedance matcher (SPDIF to AES). I had a few clicks and pops, and traced it down to a couple of XLR cables which attach the shield of the cable to both pin 1 and the shell/chassis connector. I guess the shield was then acting like an rf receiver.. It didn't help that I copied the pinouts from the XLR onto my patch panel connectors. Snipped the connections so that chassis was unconnected and it's all good.

escapemcp wrote: I had a few clicks and pops, and traced it down to a couple of XLR cables which attach the shield of the cable to both pin 1 and the shell/chassis connector. .

I've seen that on some of the crappy, cheap cables. Build your own - you'll get better quality for less money.

I've just changed my mind escape. I taped my SPL meter to a piece of tube and went through the array element by element, pointing the tube right in the centre of each piezo. They all read within 0.2db of one another except one which is 15 to 20 db down and fluctuating. I bet thats my culprit. Try that method on yours. I wasn't just getting resonances at specific frequencies, but lower harmonics of the driving frequency, so thought id check deeper.

I've checked the offending element/s. There appears to be nothing wrong with them in a structural sense, no broken wires or detached piezo disc. They are both in the same vertical row (opposite each other, right and left) and are both the free element type, where many of mine are glued on elements. I'm stumped.

I will try to check mine tonight... Don't know if I'll manage it as the rig is being loaded up tonight for a weekend gig. If not I'll check it the next time that I have a chance... Will probably just replace offending pesos and see if I can get it going without the distortion.

And on a final thought, is running a >4k sine wave (so it is only going into the tweeters) OK for the amp, as running pesos by themselves causes oscillation apparently. Does the 4 ohm resistor just before the tweeters take care of this instability???

By the sound of it though, this tweeter problem isn't a 'biggie' as you have been using yours for a while without any noticeable distortion during normal playback, yes?

escapemcp wrote:By the sound of it though, this tweeter problem isn't a 'biggie' as you have been using yours for a while without any noticeable distortion during normal playback, yes?

No, not quite. I'm just finishing up building them, and so am in the comissioning phase.

I've rehashed my comments from last night and going to post them on my build thread, for consistency (gonna miss my build thread when this is over ). But briefly, I think I have 2 piezos not working but after taking backs off all seems well, can't figure it out. The tube and taped-on SPL meter works well.

Bruce Weldy wrote:

escapemcp wrote: I had a few clicks and pops, and traced it down to a couple of XLR cables which attach the shield of the cable to both pin 1 and the shell/chassis connector. .

I've seen that on some of the crappy, cheap cables. Build your own - you'll get better quality for less money.

+1 ....Cheap cables will cost you a show quicker than anything.

Got a proper test this weekend. Have managed to get a 225W into 8 ohm amp and am using that to run the tops. Am running the signal into the DEQ and then using the aux outs (set to be the same as the inputs - i.e. no processing) for the sub and the monitor (the iNuke's DSP will do all the processing) - the main outs can then run to this borrowed amp to run the tops.

I was going to use the DEQ's EQ to make my Otops flat, but last night realised that it needs a HPF I was then thinking/hoping that I could bodge it using several high shelving filters overlaid on each other in the parametric EQ (I know, not ideal, but I just needed to get the majority of <100Hz content away from the tops). I needn't have worried.. the DEQ has a "LC" (low cut) option. I tried using one, and the slope didn't look steep enough when I compared it to the one on my iNuke... I then laid another one on top of the first and bingo! The slopes looked about the same. I then A/B'd it and there was absoutely no difference in the sound. Therefore one of those LC filters must be 12dB/oct slope, so that when I overlaid the 2nd, I got the 24dB/oct that I was looking for. Add in the limiter that it also has, and I am wondering why the DEQ/DCX combo is needed, as the DEQ seems to do everything that BFM speakers needs. Using all 10 parametric EQ's (let's call them PMEQ), you could run one channel of subs (which would need 8 PMEQ's to create a LPF and HPF with 48dB slopes), and then 2 PMEQ's on the tops to provide the recommended 24dB slopes. You can then use the graphic EQ for RTA and you have a fully 'BFM compliant' system.

Thought I'd mention this as I haven't seen it anywhere else. If I had known this when I started, I may well have used a DEQ with a bog standard amp to get going for a reasonably cheap price - instead I chose the iNuke DSP as this had all I needed on it except RTA. Still, it's nice to have options now that both are capable of doing the job.

Charles Jenkinson wrote: But briefly, I think I have 2 piezos not working but after taking backs off all seems well, can't figure it out. The tube and taped-on SPL meter works well.

I've had piezos that weren't right, but after I pulled the back off and reinstalled them they were OK. Probably a parts alignment issue.

Many thanks Bill (as always). When I saw you had responded to my post, I thought it was going to be a on the DEQ as an HPF!

I will look at the piezos when I get the rig back next week and update

The two piezos that weren't working are now working. I took the array out to check for resonances by touching things to damp it out. Managed to stop one significant resonance - it was just a loose back cover - a tweak on the screws had it sorted. Then found 4 screws were not tight, so I went along the back joints near those screws with hot glue to try to stop any vibration. I don't think I have any mechanical vibration now, but there are weird noises. I can only imagine they are not sealed and it is air pulsating through tiny gaps. The noises are audible because they are underneath a pure loud sine wave sweep, but obviously no musical signal is a pure sine wave so I think these artefact sounds will be sufficiently masked. The only way to attempt to eliminate it would be glue up all joints and seal all solder tabs - not a small undertaking.