Driver break-in

marec

Member
2006-05-28 8:34 am
I've never been a voodoo believer since the excesses of the 70's. However, recently I decided to build MJK's latest open baffle project and ordered a pair of 108EZs. One of these turned out to be faulty and had to be returned. I thought I might as well "break in" the driver I had, and fed it a signal of ~ 20Hz at 2.8 volts (thanks Scottmoose...). After a couple of hours, the driver started bottoming. I reduced the signal level. After another hour, the driver was bottoming again. When the replacement driver arrives, I'll measure it's Fo before and after break in. At the moment, it appears that something is definately happening, but whether the driver changes will be apparent throughout the audible range, I don't know. I wish I'd bothered listening before and after, but i always think that comparisons separated by minutes, never mind hours, can be deceptive... However, with a "before" and "after" driver, there's opportunity for all sorts of entertainment!
Wonder if the same thing will happen with the 15 alphas that have been lying around for months?
Rgds
 
Probably.

Break in isn't a myth, or not entirely anyway. It can be over-rated, and applied to ludicrous things (wire anyone?). In the case of FR drivers, it certainly exists. You've got a driver with very stiff suspension (relative to, say, your average mid-bass), a light cone & powerful motor, fresh out of the box. It will take a while for that suspension to loosen up at least, and the fibres used in paper cones also mature over time. How significant it is depends on the driver, and also on the manner of loading. Got it in a horn, where excursion is tightly controlled, and break-in takes longer than if in a cabinet where the suspension gets more of a work-out. Etc.
 

marec

Member
2006-05-28 8:34 am
Thanks for that...
I tried with a 15a.
Fs (Fd, F0?), out of the box =46.5 Hz.
I fed it 8 volts at 20Hz for 48 hrs. Result, Fs = 45Hz.
I seem to remember a post recommending break in at Xmax. How do I do this? Do I just drive until the voice coil hits the stops and then back off a bit?
I'm still waiting for my second 108ez to arrive so that I can measure & (and listen!) before & after....

Incidentally, I've got some bamboo flooring strips that I was going to create the MJK OB with. However, the stuff rings like a bell. I was thinking of damping this with a layer of MDF attached with 'No more Nails' or something similar. Is there any experience of using this material i can tap into please?

IIRC Planet 10 has built enclosures from bamboo. Any comments?
 
Ah. The old resonance issue.

Something that people often forget is that it's often not how severely a material resonances so much as where it resonantes. If a material's (primary) resonant frequency occurs at a point where a box / whatever is working hardest & putting out a lot of energy, you're in trouble. If it occurs elsewhere, where comparatively little energy, which might excite it, is produced, then it's basically not much of an issue. A light, stiff material also releases energy quickly rather than storing it & slowly letting it go.

Go easy on the 108s. Just play some rock music or similar through them at moderate volumes for a few days to loosen things up.
 

marec

Member
2006-05-28 8:34 am
Re-reading my post and thinking about my casual use of 'ringing like a bell' set me thinking.
If it is bell like, does this imply high Q, low frequency spread? If so. notch filter?
I got out a couple of the strips. It's not yet a board! Holding the strips between two fingers on the edges and rapping with a knuckle ( Oh, this is soooo scientific...), reveals a musical instrument. There is the loud noise of the initial impact, which isn't an identifiable note (at least, not on my piano). I imagine, then, that this is a result of many frequencies. It is of relatively short duration, but longer than the same impact on an MDF strip.
There is also a much longer duration, lower frequency and purer note which isn't generated when the strip is suspended at it's end. Several strips tested have different short duration notes, but similar long duration notes. I guess this is related to the length of the strip. Strong harmonics are generated when supporting the strips at various positions on it's edges.
The amplitude of these concerns me. What, I wonder will happen when I glue them together as a board? Should I perhaps only glue them at an end? Or both ends, suppressing the observed reonance?
Isn't building even the simplest of loudspeakers fun when you are so ignorant????
 

marec

Member
2006-05-28 8:34 am
How sad is someone who posts to their own thread when no-one else does?????????
Scottmoose: Thanks, If low storage = low mass and if, in the case of an open baffle, preventing the backwave from reaching the front wave is a small issue (given the distance from a reflecting boundary...), should the baffle material for an OB be as light as possible commensurate with supporting the drive units? One could effectively decouple the drive units from the baffle, reducing the direct transmission of vibration from the chassis.
 
I've used cheap 12"x18" bamboo cutting boards for OBs at my desk with great success. I've been looking at the bamboo flooring as a method to make larger baffles from locally available materials (think I could get 2 pretty darned large ones from a single $70 carton of the thick, solid bamboo stuff, available in natural or caramelized). I'm of the opinion that if I can't get it done in bamboo, somehow, I'm not likely to build it (and that applies to anything). The resonance of the material should change significantly as you glue pieces together, from the increased width and the mechanical interfaces between the pieces. Good luck and let us know how it turns out.

Kensai
 

marec

Member
2006-05-28 8:34 am
Well,an interesting few days....
Having poured 1/2glass of perfectly good Crianza into my keyboard, the laptop started typing even more rubbish than usual.I've never had a laptop apart before, so started with every visible screw on the base.
This revealed all sorts of interesting bits,but after several hours I seemed to be no closer to the keyboard! Fortunately a kindly soul borrowed a kitchen knife from our drawer, flicked off a bit of plastic trim from the top and voila, keyboard falls out! A good wash and blow dry and (nearly) as good as new. I wonder where those four screws on my desk came from?

Anyway, the replacement 108 turned up, so I stuck them in a couple of small OB's.

OMG.... What have I got here? I really can't tell much difference between the one that's had the abuse and the brand spanker. In fact I can't get past the banshee like noises they make to even make a comparison. 10 second headache!

Out for lunch I think and see what happens after a couple of hours!

Lowthers were never like this.....
 

Bob Brines

Member
2003-01-31 10:11 pm
I have been preaching this for a long time and have been met with general derision here. My experience is that Fostex and Lowther drivers break in to usable degrees within a hour or two. While there are indeed small changes in detail, etc for maybe 100 hours or so, it takes A/B testing to detect this.

How do I know? I have built dozens of identical speakers. I take a new pair of drivers out of the boxes and install them. I run the new speakers while I build the packaging and boxes. The new speakers now sound the same as the speakers in my listening room that have 1000's of hours on them.

Here's the thing. When you change your system, the more different the new component from the old, the longer it takes your EARS to adjust. When your EARS accept the new sound as normal, the system is broken in. The 108's are completely different from what you had before and it took a long time for your ears to break in. When you got the new 108, it sounds just like the old one. There was virtually no break in.

QED

Bob
 
Agreed, & certainly no derision from me Bob. About the only caveat I'd add is the point I noted above -the time can greatly depend on the cabinet / whatever -something with high[er] excursion requirements will usually break the bottom end of a driver in more rapidly than one with lesser. For the rest, as you say, the final few percent usually seems to come in over 100 - 300 hours or use.

Something that I suspect is often forgotten about is the warming of the voice-coil & its effect on the electrical properties of the driver. Fostex drivers in my moderate experience tend to take about 15 - 20 minutes to fully come on-song. YMMV of course. At an audio-meet last summer, many people were surprised at the difference between hearing a pair of speakers immediately on firing up, wombling away, & coming back half an hour later.
 

marec

Member
2006-05-28 8:34 am
Another (and in this case important) issue is level. I came back from lunch in a more mellow frame of mind to find that I could hear the 108s clearly through a quite solid internal door. This is not the usual case with the normal loudspeakers used in this room (Delphics). My conclusion: Playing them far too loud. I remember this happening before with some of the smaller redehkos. For some reason they seem much quieter than they actually are. I suppose the usual story is low distortion making it sound cleaner. Anyway, after turning them down by a considerable margin, they are much more acceptable.
I still can't claim that there is any glaring difference between the two units, and as they stand, in modest OBs, they don't have a balanced presentation, male voice is too dry and I can't hear much of the cello in the quartet currently playing on R3. But then, that's what the 15As are for...

I tend to agree that most of the issues are acclimatisation of ears rather than drivers. Unless there are major anomalies,you tend to get used to them (and quickly) in the same way as you get used to a room acoustic. Notable exception being opera in Westminster abbey, which I never got used to!!
 

Bob Brines

Member
2003-01-31 10:11 pm
There is another issue here. I typed up my response above because it addresses the tread title and one of the issues addressed in post #10. I almost didn't send the post though, because I believe the real issue here is balance.

The FE108ES, particularly mounted on an OB, is going to have no bass at all. Also presuming no filtering at all, this driver has a rising frequency response. This heavily loads the top end of the spectrum -- ear-bleeding screechy and in-your-face. You need a woofer to give you a bottom end. IMO, a small FR driver like the 108 should be used as a mid-tweeter rolled off no higher than 300Hz to a 10" or bigger woofer.

I've tried many times to design a single-driver speaker around a 10-12cm driver and never come up with anything that satisfies me. The problem is that cannot produce any useful bass and still provide a balanced frequency response. I guess that "useful bass" has a different meaning to me than to some others on this forum.

One of my design requirements is that the speaker F3 must be no higher than 40Hz. This reproduces all of the normal orchestral instruments down to the double bass. I can get there and have listening chair SLP to 95dB or better with 16cm drivers and larger. With the 10-12cm drivers, the best I can get is F3=70Hz or even higher. This would be alright for voice backed by guitar, solo guitar, string quartet, maybe even piano jazz. For full orchestra or anything that uses a bass guitar, let alone a 5- or 6-string bass, this is unacceptable.

Is it possible to get to 40Hz with a 10cm driver? Well, yes it is. But size becomes an issue. Shipping costs skyrocket once the size exceeds what UPS/FedEx will accept. Also, I perceive that this is the normal upper limit of WAF. There is also a cost issue. I cannot build a good back horn design at a price point that I feel is commercially viable. There are two markets for a one-off boutique shops like me, very low price and very high price. I chose the former. I can't compete with the big name brands at their price point.

Enough.

Bob
 

marec

Member
2006-05-28 8:34 am
Thanks for that Bob:
I don't really expect a balanced result from a driver of this size. I was just a bit shocked by how awful it seemed after the almost unanimous praise Fostex units in general, and this one on particular, seem to have received.

These units were purchased to build MJK's project 7 and will have bass support from an Eminence 15a.

I opened the thread because i was surprised at the impact feeding this unit with 20 hz seemed to have. However I cannot relate the apparent changes in amplitude to anything audible. No Golden Ears I'm afraid...

Mark
 

marec

Member
2006-05-28 8:34 am
Scottmoose:
Thanks for your interest in this. I have seen the enclosures made from expanded polystyrene. Pig ugly???

Without any measurement support whatsoever, the OB's I liked the most were (after Briggs) 10mm MDF, 50mm Polystyrene, 6mm hardboard.

However, when I was thinking of light, non-resonant baffles, I had clingfilm over 10mm ^2 frame in mind.

I really don't understand what's needed to separate backwave from front. It (intuitively) can't be clingfilm, but why could it be cardboard or polystyrene? Why doesn't it have to be something inert and massy? Or is it just a matter of degree?

There are days when I feel stupid...

Then there are days when I am stupid. The rest of the time I'm dreaming.

Mark