Here is a distortion measurement of my 3" ETON at 77db measured through the XO grouped 2 ways. I used a dayton audio emm-6 + scarlett 2i2.
If I group it by SPL it looks less extreme than if I group it by percent.
I'd be curious to see yours grouped by the %.
If I group it by SPL it looks less extreme than if I group it by percent.
I'd be curious to see yours grouped by the %.
Attachments
Okay, new measurement including Noise floor, h2 and h3, with no hp or lp filters applied.
96db taken at 1/2 meter and then placed 10cm from from phase plug to lower noise floor.
110db/10cm
96db taken at 1/2 meter and then placed 10cm from from phase plug to lower noise floor.
110db/10cm
Attachments
Last edited:
Even taking into consideration the 3" driver, it looks like you have some baffle step happening. What distance are you measuring at?
12" is very close, you should use spl reading 100-110dB to see distortion over noise better. In rew you can select what lines it draws, jus unclick those you don't want to see (D4, D5, THD etc.) to make graph more readable. UMIK-1 starts distorting itself above 115-120dB!
Then, when you see a peak in distortion, use REW's generetor to play only that freq in sine and touch the speaker and frame to try if it's mechanical resonance. You will hear the change and even feel it in your hands!
There are different versions of REW, they have different options for analysis and presentation. Yestereday I couldn't find toggle to show noise level in some old measurements from 2013!
JBL LSR305 with added in-box support, measured at 1m indoors. I show noise only in 96dB graph, to make others more readable!
Then, when you see a peak in distortion, use REW's generetor to play only that freq in sine and touch the speaker and frame to try if it's mechanical resonance. You will hear the change and even feel it in your hands!
There are different versions of REW, they have different options for analysis and presentation. Yestereday I couldn't find toggle to show noise level in some old measurements from 2013!
JBL LSR305 with added in-box support, measured at 1m indoors. I show noise only in 96dB graph, to make others more readable!
Attachments
Last edited:
It was only at 12".
I was hoping you were just seeing the noise floor like I was. Some air filters were messing mine up.
You should use the AINO gradient for this! Why use a commercial product when you have a diy system you are very proud of?
My diy speaker system is an active 5 way using Peerless HDS exclusive drivers with 700 watts per side. Loudness vs distortion is not an issue I need to worry about!🙂
You need to quit yelling at me and stand behind your creation!🙂
(I am making fun of all your exclamations!)
My diy speaker system is an active 5 way using Peerless HDS exclusive drivers with 700 watts per side. Loudness vs distortion is not an issue I need to worry about!🙂
You need to quit yelling at me and stand behind your creation!🙂
(I am making fun of all your exclamations!)
Attachments
Last edited:
Juhazi
96db/1m
System hd stays below .5% through entire usable frequency band
I was just joking about comparing it with your diy system, but if you feel the need, it would be fun!
96db/1m
System hd stays below .5% through entire usable frequency band
I was just joking about comparing it with your diy system, but if you feel the need, it would be fun!
Attachments
Last edited:
Looots of distortion measurements in my AINO thread! Please check it! No need to copy here.
Individual driver measurements without eq and xo are rare and done mostly with low spl. I didn't save all my test sweeps, but believe me I have hundreds of them saved!
Concerning CLD damping/preventing distortion/resonances - most likely it does so, but making reliable and valid measurements to prove it is not easy!
p.s. Here is my thread about damping panel resonances of JBL LSR305 JBL LSR305 tweaking
Individual driver measurements without eq and xo are rare and done mostly with low spl. I didn't save all my test sweeps, but believe me I have hundreds of them saved!
Concerning CLD damping/preventing distortion/resonances - most likely it does so, but making reliable and valid measurements to prove it is not easy!
p.s. Here is my thread about damping panel resonances of JBL LSR305 JBL LSR305 tweaking
Last edited:
The point I’m trying to make, is if you have hundreds of measurements saved of your diy project, why would you use a measurement of the JBL’s, when posting your diy system distortion would be just as easy?
I am very familiar with the open baffle concept, and am good friends with Igor Levitsky. I helped him with beta testing headphones and speakers while he did a stint for Oppo Digital. As you can see, I'm using Neo 10's and Neo 3's in this open baffle project, from 350hz and up. This picture was from about 10 years ago.
Last edited:
Because i didn't find appropriate saved measurements of AINOs for this purpose! Fullrange sweeps were done at lower spl or not saved at all. I can't do new measurements now right away!
Here some distortion graphs of AINOs, but not the latest dsp-version. With LR4 xo I get lower distortion than with LR2, but I prefer the sound of LR2!
Attachments
Last edited:
Attachments
Constrained layer baffle, from top to bottom..
1. leather
2. 1/8" aluminum plate
3. 1/8" sorbothane
4. 1/4" aluminum plate
5. 1/8" sorbothane
6. 1/4" aluminum plate
7. 1/16" neoprene
8. 1/8" sorbothane baffle/box gasket
9. rubber washers on driver bolts, to decouple the driver from the back plate.
10. Tweeter is mechanically isolated from the midrange by being mounted on a different aluminum plate.
11. The mid/tweeter module is also sitting on sorbothane to help decouple it from the bass modules beneath.
In order for sorbothane to work it's best, it needs to be compressed. Baffle mounting bolt tightening provides the needed compression.
1. leather
2. 1/8" aluminum plate
3. 1/8" sorbothane
4. 1/4" aluminum plate
5. 1/8" sorbothane
6. 1/4" aluminum plate
7. 1/16" neoprene
8. 1/8" sorbothane baffle/box gasket
9. rubber washers on driver bolts, to decouple the driver from the back plate.
10. Tweeter is mechanically isolated from the midrange by being mounted on a different aluminum plate.
11. The mid/tweeter module is also sitting on sorbothane to help decouple it from the bass modules beneath.
In order for sorbothane to work it's best, it needs to be compressed. Baffle mounting bolt tightening provides the needed compression.
Last edited:
Remlab - first of all, about your 8 layers of CLD baffle: you, sir, are insane... in a good way. 🙂
Second: are you now confident that you have a valid distortion measurement of your 4 inch driver? I don't think I followed the last couple of posts very well...
J.
Second: are you now confident that you have a valid distortion measurement of your 4 inch driver? I don't think I followed the last couple of posts very well...
J.
I absolutely think the nonlinear distortion readings within the driver's passband(200hz to 2khz) are correct, but the low frequency readings are not, most likely due to the measurement distance being used, which is 10cm. Vance Dickason measures a driver with pink noise at around 94db/1m and then places the mic at 10cm to lower the noise floor and room effects and then uses a sweep. I'm using this technique. I think it works well from about 200hz and up, but it doesn't represent the low frequency nonlinear distortion/fundamental ratio well compared to normal listening distances.
Last edited:
I have to admit, It does look a little insane🙂
This idea was kicking around in the back of my mind for years, and I finally got some balls and actually did it.