I've been trying to find a Labhorn's peak power handling edge by slamming a drum recording through it.
The edge seems to be when I get a loud clunk/knock out of the Lab 12s
I have a fast min/max voltmeter saying this happens at around 148v PEAK.
A little below that, no problem...
So I know where to back off...
But what I don't get is what makes the clunk/knock.
I have a lab12 torn apart and the suspension doesn't seem like it could ever move in or out enough to let the voice coil or anything else, hit anything.....
I guess I don't have a clue what happens at xlim.....(22mm according to eminence)
Thanks in advance, Mark
The edge seems to be when I get a loud clunk/knock out of the Lab 12s
I have a fast min/max voltmeter saying this happens at around 148v PEAK.
A little below that, no problem...
So I know where to back off...
But what I don't get is what makes the clunk/knock.
I have a lab12 torn apart and the suspension doesn't seem like it could ever move in or out enough to let the voice coil or anything else, hit anything.....
I guess I don't have a clue what happens at xlim.....(22mm according to eminence)
Thanks in advance, Mark
In all designs i've ever seen the end of xlim (near published value or not) is when the coil former slams into the inside of the T plate. May as well be advised that running just below that effect will likely lead to highly accelerated suspension wear. Better to design for at least 3db below "clunk", not just a couple volts. If you don't mind the severe compression, distortion, and continuous driver replacement you'll be fine at 148vp at whatever frequency.
Last edited:
hmmm.....$ilence not golden !
"In all designs i've ever seen the end of xlim (near published value or not) is when the coil former slams into the inside of the T plate. May as well be advised that running just below that effect will likely lead to highly accelerated suspension wear. Better to design for at least 3db below "clunk", not just a couple volts. If you don't mind the severe compression, distortion, and continuous driver replacement you'll be fine at 148vp at whatever frequency."
Thx Andrew. Makes sense.
What gets me, is in measuring how the coil sits in the magnet, it looks like the coil would have to go about 5mm beyond the given 22mm xlim, to strike the plate!
148vpeak must be so crazy hard on the drivers. Can't afford this...
But it certainly doesn't sound compressed right below clunk....in fact a single kick drum stroke sounds about as real as I've heard (when the mains are matched right).
It sounds Wow right until clunk
Since I can't really hear distortion or compression on an impulse signal like this, is their a way to know when, or measure, that I'm pushing past xmax?
Will REW or Smaart show this?
"In all designs i've ever seen the end of xlim (near published value or not) is when the coil former slams into the inside of the T plate. May as well be advised that running just below that effect will likely lead to highly accelerated suspension wear. Better to design for at least 3db below "clunk", not just a couple volts. If you don't mind the severe compression, distortion, and continuous driver replacement you'll be fine at 148vp at whatever frequency."
Thx Andrew. Makes sense.
What gets me, is in measuring how the coil sits in the magnet, it looks like the coil would have to go about 5mm beyond the given 22mm xlim, to strike the plate!
148vpeak must be so crazy hard on the drivers. Can't afford this...
But it certainly doesn't sound compressed right below clunk....in fact a single kick drum stroke sounds about as real as I've heard (when the mains are matched right).
It sounds Wow right until clunk
Since I can't really hear distortion or compression on an impulse signal like this, is their a way to know when, or measure, that I'm pushing past xmax?
Will REW or Smaart show this?
Good. It would be a rotten deal for interference to start 5mm below the datasheet limit. Tends to indicate that if you design below Xmax you're gonna have a good time, and that you're getting the driver you're buying. Not all drivers are actually linear out to the Xmax edges, but you Should be able to demand continuous running to that point without unreasonable wear. For this reason, Xmax spec is very handy for design. Xlim /Xmech just tells you when you're going to be buying new drivers right away.
Last edited:
drivers are in parallel, or with an individual amp on each driver.
in either case, the amp(s) are plx 3402 in bridge mode, so I know i'm pushing the heck out of the drivers.
I'm really just trying to learn how to get the most that I can from the labhorns.
I'm willing to destroy one set of drivers if I have to; but i have 4 labhorns and can't afford not to know where the common sense (and not so common sense
impulse edge is before I wail on all four.Continuous stress seems easier to hear, certainly to measure, than a drum type impulse stress....
Last edited:
You've been hitting Lab12s with 148v peaks?
Ouch. That's nearly 2kW (taking rms voltage) per driver.
I'd set your peak limiter for 140v peak out of the amplifier, which would allow some clearance for differences between drivers. Then, if possible, an rms limiter at 6-8dB below that, which should mean the drivers will stand long sustained bass notes.
Something to consider, though: "clunk" voltage at with one cabinet isn't the same as when there's 4 cabinets - FLHs load each other.. Kinda.
Chris
Ouch. That's nearly 2kW (taking rms voltage) per driver.
I'd set your peak limiter for 140v peak out of the amplifier, which would allow some clearance for differences between drivers. Then, if possible, an rms limiter at 6-8dB below that, which should mean the drivers will stand long sustained bass notes.
Something to consider, though: "clunk" voltage at with one cabinet isn't the same as when there's 4 cabinets - FLHs load each other.. Kinda.
Chris
I set the speaker cloth grills just a little beyond Xmax on my ported Lab 12s so when they started to flap, I knew they were exceeding Xmax.148vpeak must be so crazy hard on the drivers. Can't afford this...
But it certainly doesn't sound compressed right below clunk....in fact a single kick drum stroke sounds about as real as I've heard (when the mains are matched right).
It sounds Wow right until clunk
Since I can't really hear distortion or compression on an impulse signal like this, is their a way to know when, or measure, that I'm pushing past xmax?
Will REW or Smaart show this?
You can measure distortion at various voltage levels with Smaart, the method is shown in the Keystone thread.
http://www.diyaudio.com/forums/subwoofers/185588-keystone-sub-using-18-15-12-inch-speakers.html
Distortion results are in post #12, there is a typo: The * should read: third (odd order)harmonic louder than second.
Hornresp inputs are available in post #96. The link explaining how distortion was measured expired, an explanation is in post #315, and a correction in #316.
You can also measure excursion pretty easily by eye looking at a white dot painted on the edge of the cone.
I've pushed as hard as I could by hand and not made the Lab 12 voice coil hit the back plate, the clunk (which I've never heard) may be the spider hitting the magnet structure.
At any rate, the speaker is only rated for 400 watts AES2-1984, most people burn them up before they tear them.
Art
Last edited:
Art, I stole that idea from you on that vented box I made... set the grill height at 14mm off the top of the surround... we'll see if I can tap the grille
If you do like the OP and hit it with 148V, you will find out which is stronger, the grill or the surround
.Mine were set probably more like 16mm, all the surrounds show grill marks, but no damage. I would not try this with steel grills, I use speaker cloth.
Art, with the drivers sealed up the way they are in the labhorn (air leaks are death), there's no real way for visual observation without a cam or something.
Yes, even with the spider cut apart full round at its perimeter, I still have to push hard to get the coil down to the plate......that's why I started this thread....to make sure I wasn't missing something else .
The spider may be hitting the magnet structure, but the clunk/knock is so sharp it's hard to believe the sound could come from spider collision.
Amazing really, the lab12's take a lickin' and keep on tickin'
I'll study the thread you pointed me to, but will smaart work with a single stroke impulse like a kick drum? for determining peak limiting ?
BTW, the d-fend set at 140v peak with 19ms attack, stops all clunk, even at mixer levels 4db higher than what produced 148v amp peak ....
....but who knows what damage I'm doing to the drivers even with the d-fend limiting....
Yes, even with the spider cut apart full round at its perimeter, I still have to push hard to get the coil down to the plate......that's why I started this thread....to make sure I wasn't missing something else .
The spider may be hitting the magnet structure, but the clunk/knock is so sharp it's hard to believe the sound could come from spider collision.
Amazing really, the lab12's take a lickin' and keep on tickin'
I'll study the thread you pointed me to, but will smaart work with a single stroke impulse like a kick drum? for determining peak limiting ?
BTW, the d-fend set at 140v peak with 19ms attack, stops all clunk, even at mixer levels 4db higher than what produced 148v amp peak ....
....but who knows what damage I'm doing to the drivers even with the d-fend limiting....
Yea, ouch is right.
I guess it's all about safe thermal ....or ....safe excursion....
Right now, I'm focusing on excursion, and I guess a kick drum seems to be the biggest test I've found...
I know 140v peak limiting will stop the clunk. But is that OK for the driver long term? Maybe it is for what I play. I know of very few recorded songs that can produce peaks like this....
Oh btw, I'll get back to thermal after i figure out peak excursion. I've got some BBQ guru temp probes routed inside the test labhorn. If I can't trash it through excursion, maybe i can cook it to death
Yes, I've been pondering what effect on peak that 4 together will have.
I've measured the effects of coupling on freq response and impedance...it's not trivial is it ...
Your drivers will get warm - if you limit to 70v peaks, and then mash into the limiters, you're dissipating a lot of power.
I learnt that the hard way and blew a rather nice RCF driver. These days, I match amplifier power to speakers better, and avoid hitting limiters as much as possible.
Chris
how many labhorns do you have? give it 30 minutes of this on loop (no intro/outro) at more than 50 volts and Id say you'll be good to bbq
Kaiju ft. Flowdan - Hunter [OSMUK029] - YouTube
I'd assume hitting them with this track at 148 volts would result in near instadeath for a lab12.
may have to cop a hq version of it though, bass is lacking in the youtube vid.
Kaiju ft. Flowdan - Hunter [OSMUK029] - YouTube
I'd assume hitting them with this track at 148 volts would result in near instadeath for a lab12.
may have to cop a hq version of it though, bass is lacking in the youtube vid.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.