They were measured in same session within minutes of each other. I took two sweeps to confirm consistent and kept second one.
Was this a blind test then? I.e. nobody knew what was being tested (if anything)?
Actually, getting the experiment set up and run properly is the expensive part. Once you have data, you take a couple of days to type it up and submit it to AES. Typing is cheap. Time consuming, yes, but cheap.
I would definitely be interested in comparing the same driver in the same enclosure with/without treatment. Keep everything the same and only change the EnABLing. That would be neat. I also do understand that you have a business to run and may/may not have time to set that up. It's all good.
Tom
Double blind, self-administered. A Mac with iTunes + PureMusic, a DAC, an amplifier with a speaker A/B switch hooked to 2 matched speaker pairs, 1 treated, 1 stock, speakers behind an opaque screen.
dave
About increased sensitivity: Of course adding mass in and of itself should reduce the system sensitivity.
However, if there is significant flexure, deformation, bending, whatever, of the cone while in use before treatment, stiffening the cone by treating it could increase sensitivity by reducing its in-use (dynamic) deformation. Perhaps this isn't significant, but it is a possible mechanism.
How much of the surround is treated? You could also increase sensitivity by increasing the 'effective' cone.
However, if there is significant flexure, deformation, bending, whatever, of the cone while in use before treatment, stiffening the cone by treating it could increase sensitivity by reducing its in-use (dynamic) deformation. Perhaps this isn't significant, but it is a possible mechanism.
How much of the surround is treated? You could also increase sensitivity by increasing the 'effective' cone.
I was under wrong impression that EnABL process is consisting only of painting those silly dots. I admit I didn't pay attention that during the process the whole loudspeaker cone is treated with coating, plus some treatment on cone/surround termination. Of course these modifications (without EnABL dots!) will change the frequency response of the loudspeaker and will be easy to measure and to hear. But, I remain very skeptical about the influence of the EnABL dots after the coating done before. IMHO whole process of EnABL before applying dots is responsible for ALL of measured/audible difference.
The surround treatment is not part of EnABL.
As i said earlier, i have been doing the other mods for a long time. When we 1st started doing this we evaluated the old mods vrs EnABL and old mods + EnABL. Our view is that EnABL added more to the mix than the old mods. Since then EnABL has been improved (EnABL2) and the surround treatment added (only on some drivers). This last also used on the lip of whizzers. It should be noted that the bigger FExx6 Fostex has something already at the surround attachment, and the whizzer on the TB W8-1808 has something on the back top edge.
dave
As i said earlier, i have been doing the other mods for a long time. When we 1st started doing this we evaluated the old mods vrs EnABL and old mods + EnABL. Our view is that EnABL added more to the mix than the old mods. Since then EnABL has been improved (EnABL2) and the surround treatment added (only on some drivers). This last also used on the lip of whizzers. It should be noted that the bigger FExx6 Fostex has something already at the surround attachment, and the whizzer on the TB W8-1808 has something on the back top edge.
dave
Here is a sound clip of the Dagger and RS225 FAST speaker (passive XO) so you can get a sense of what they sound like. Recorded with UMM-6 mic at 0.5m away (same position and same mic as measurement). Change .asc extension to .mp3 to hear. Recording made with Audacity to 44.1kHz wave file, then converted to 320kbps MP3. I should add that the source was only playing through the right channel (so any info on left channel is lost).
Last edited:
Cool! That sounds like a good experiment. How many participants and how many percent preferred the EnABL?
Tom
We didn't keep track. Most, if asked, preferred the EnABLed drivers. We have done this a couple times.
dave
dave
I disagree with you on many points. First off, I doubt the effect of the EnABL is significant and easy to measure and hear. Some of the measurements here indicate a 1.5 dB difference across a pretty narrow frequency range. That's audible, but is it easy to pick out? That's a good question. It is probably subtle and for some it'll be worth the expense of EnABLing the drivers. This is entirely my own personal opinion. Dave has his followers and for them the EnABL is undoubtedly worth it.
Secondly, the dots themselves have to make a difference. I've seen EnABLed drivers in person and the dots appear to be some form of paint, probably applied with a stencil or silk screen. They're nicely done and look pretty. The dots themselves have to have some amount of mass, hence, they'll change the moving mass of the diaphragm. Whether it's enough to cause a shift in the T&S parameters beyond what you'll see from part-to-part variation, weather changes, and whatnot is undetermined, or at least not known to me.
Tom
Last edited:
Bummer that the results weren't recorded. Should you set up an experiment like that in the future, feel free to toss me the results. I'll need the total number of participants and the number of participants preferring A, B, and no preference (if that's recorded). I'll run the statistical math on them to see if they're statistically significant or if you'd be as likely to get the same results by chance.
I'll encourage you to let people decide on their favourite in private or by anonymous voting (by dropping a note in a sealed container) to discourage social conformity.
If you're sneaky, you don't tell me what A and B are until after I've run the analysis. 🙂
If the results show a significant preference for the EnABLed drivers, that'd be a pretty neat story for your website and business.
Tom
Forget the FR. It has been done over & over again and little changes. The real differences are 20-40 dB down.
What i listen for is how the image/soundstage changes. If on the recording the EnABL drivers will throw a much more 3D image. I can pick it out in a couple switches.
Model train paint applied with a (specific) caligraphy pen and a template. I have a bunch of pictures i need to process & caption that show the process.
The shift is within unit-to-unit variation so an EnABLed driver works in the same box as a stock driver.
dave
Member
Joined 2009
Paid Member
I've also seen some FR comparison plots posted before and differences were rather small; it's something else that needs to be measured. Perhaps a single tone signal input. Bud talked about the dots helping surface waves leave the cone before they crash (my word choice) into the surround. How about some close-mic measurements at a few different single frequencies?
Fair enough.
That says THD or IMD to me. The THD plot a few pages back didn't show much difference, though.
Wow. I admire your skill and patience.
Tom
That doesn't make sense. Surface waves don't leave, at least not that I know of. I do suspect that the dots change the cone break-up somehow. Cone break-up is when standing waves develop across the cone. If the dots do change the breakup, it should be visible in the frequency response and THD as well, though.
We could probably figure out what's going on with one of those laser vibration measurement rigs. 🙂
Tom
Last edited:
That is stuff added (i tend towards Geddes view, these measures don't mean much). What i am talking about is the driver letting more information thru.
dave
I have no idea how EnABL works, but i suspect that what is happening is more about the air above the cone and not the cone. A laser vibrometer won't help if it is.
There is a tool that uses an array of microphones for measuring sound -- http://www.microflown.com/-- that might be appropriate. I believe this is the tool Geddes said detected his HOMs.
Both tools would be useful i suspect.
dave
- Status
- Not open for further replies.