OP
- Thread Starter
- #421
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
ErinsAudioCorner
- Thread starter hardisj
- Start date
OP
- Thread Starter
- #422
I have been busting my tail the past week, overhauling all my distortion testing of raw speaker drivers. I don't plan to change anything for the loudspeaker testing (which is mostly what everyone on this forum cares about). However, to those who care about raw speaker driver measurements, here's a question...
I am considering providing my HD plots as a gif. That way it is easier to see the relative changes as the output is increased. Start with standard 2.83v RMS, then +6dB, then +12dB. Stick with this, or just provide them separately? I was thinking this because it saves real estate and you don't have to scroll up/down to compare. But, I'm cool with providing them separately if that's what ya'll prefer. LMK.
And, yes, I can change the timing. This is with 2 seconds between each image. I can bump that up to 3 or higher if it helps.
(If you're wondering what's going on in the HF, I can't say for sure. I believe it may be more a function of SNR as the woofer under test drops like a rock above about 200Hz and all I'm left with is background noise. As the signal increases the HF distortion components went down which supports my theory. So don't sweat that. )
I am considering providing my HD plots as a gif. That way it is easier to see the relative changes as the output is increased. Start with standard 2.83v RMS, then +6dB, then +12dB. Stick with this, or just provide them separately? I was thinking this because it saves real estate and you don't have to scroll up/down to compare. But, I'm cool with providing them separately if that's what ya'll prefer. LMK.
And, yes, I can change the timing. This is with 2 seconds between each image. I can bump that up to 3 or higher if it helps.
(If you're wondering what's going on in the HF, I can't say for sure. I believe it may be more a function of SNR as the woofer under test drops like a rock above about 200Hz and all I'm left with is background noise. As the signal increases the HF distortion components went down which supports my theory. So don't sweat that. )
- Joined
- May 21, 2019
- Messages
- 4,036
- Likes
- 6,827
I think separately just for the user-level control.
For space/compactness maybe some additional formatting would be necessary. Thumbnail pics that could be enlarged, for example.
You already have to do quite a bit of scrolling. I don't think it would significantly worsen that problem either way, but I might prefer a slow-switching animation as a way to save space. More interested in the traces in a coarse way.
Those harmonic traces are a puzzle.
Those harmonic traces are a puzzle.
I doubt I would ever have picked up audio if a large and thorough set of measurements had existed for speakers. But now I can barely rouse my interest in finished speakers. Strange how these things work out.
OP
- Thread Starter
- #425
HD script is written and this is what I am going with, after some feedback here and on my FB group page.
OP
- Thread Starter
- #426
IMD = Intermodulated Distortion = Distortion caused by playing more than one tone at a time. Typically tested with just 2 tones: a low frequency tone near Fs and a higher frequency.
For example a speaker driver plays 30Hz at the same time it plays 200Hz. Any distortion artifacts created by the sum and/or difference of the speaker playing both tones at the same time is a result of IMD. For example, the speaker is supposed to play ONLY 30Hz & 200Hz at the same time. Thanks to IMD, it plays 200+/- 30. Second order IMD would be 200+/- 30Hz = 170Hz & 230Hz. Third order IMD would be 200+/- 60Hz = 140Hz & 260Hz. If these 'side bands' are high enough in output then they are heard as distortion. If not, they are not. Simple version. I'll create a video on this later that will hopefully help those who don't quite understand the concept.
With that in mind, I am using Klippel's IMD module to test mid/woofers in two ways:
The above is tested at 3 voltages each: 2.83v, 5.92v, 9v. This makes my life easier and gives us an idea of what happens as output increases.
The purpose of me sweeping voltages is to show what happens as the drive unit is pushed for more output.
The purpose of me testing with two methods (different bass tones) is to see the difference between what happens when the driver plays with high excursion vs when a typical HPF is used.
With all of that said, I have a few options for presenting the results.
Examples are attached. #2 is the first gif. #3 is the second gif.
Remember, the multitone testing is what I am using to determine max SPL through a bandpass region. The IMD test is really more for further understanding but isn't intended to provide a final number of SPL. Just to give us an idea of the performance of a speaker in regards to what happens when you push the excursion envelope.
For example a speaker driver plays 30Hz at the same time it plays 200Hz. Any distortion artifacts created by the sum and/or difference of the speaker playing both tones at the same time is a result of IMD. For example, the speaker is supposed to play ONLY 30Hz & 200Hz at the same time. Thanks to IMD, it plays 200+/- 30. Second order IMD would be 200+/- 30Hz = 170Hz & 230Hz. Third order IMD would be 200+/- 60Hz = 140Hz & 260Hz. If these 'side bands' are high enough in output then they are heard as distortion. If not, they are not. Simple version. I'll create a video on this later that will hopefully help those who don't quite understand the concept.
With that in mind, I am using Klippel's IMD module to test mid/woofers in two ways:
- "Bass tone" fixed at the driver's Fs, with a "voice sweep" where the 200Hz - 6kHz region is played, one tone at a time (over 50 individual tones).
- The same as above, but the "bass tone" fixed at 80Hz.
The above is tested at 3 voltages each: 2.83v, 5.92v, 9v. This makes my life easier and gives us an idea of what happens as output increases.
The purpose of me sweeping voltages is to show what happens as the drive unit is pushed for more output.
The purpose of me testing with two methods (different bass tones) is to see the difference between what happens when the driver plays with high excursion vs when a typical HPF is used.
With all of that said, I have a few options for presenting the results.
- "EVERYTHING, SEPARATELY". I can just dump all the charts (6 total) and you guys can figure out the differences.
- "COMPARE BASS TONE CHANGE". I can provide a gif that compares the difference of the Bass Tone's effect directly to each other. What happens when you emulate a HPF of 80Hz (with a brick wall). 3 graphics: one for each voltage (SPL output).
- "VOLTAGE STEPS COMPARED WITH SAME BASS TONE". I can provide a gif of each bass tone method. 2 total graphics. One for each bass tone, showing the difference as voltage (SPL output) is increased.
Examples are attached. #2 is the first gif. #3 is the second gif.
Remember, the multitone testing is what I am using to determine max SPL through a bandpass region. The IMD test is really more for further understanding but isn't intended to provide a final number of SPL. Just to give us an idea of the performance of a speaker in regards to what happens when you push the excursion envelope.
OP
- Thread Starter
- #427
About two weeks ago I decided I wanted to add IMD testing back in to my portfolio (I used to conduct IMD years back but quit when I no longer used Klippel products). I also figured while I was at it I would do my harmonic distortion and multitone testing a bit different.
After countless hours, I finally have completed and updated the first of my re-tests. I am sharing it here to see if you guys are cool with the GIFs I am using for HD and IMD. I also have readjusted my parameters for the maximum SPL testing; I used to test a much broader bandpass region so I was lax on the thresholds but now that I am testing a more typical bandpass region I have tightened the thresholds up (from 10% distortion to 3% distortion and from 3dB compression to 2dB compression).
Go check it out. LMK if the gifs are cool or if you'd rather have the still images. I prefer the gifs, myself. I don't really want to provide 6 images for HD and 6 more for IMD. The gifs make things simpler. If we are cool with it then I'll work on getting the other drivers' new data posted and I can finally get back on track with other tests.
https://www.erinsaudiocorner.com/driveunits/dyn_mw172/
here's an example of *one* of my IMD tests
After countless hours, I finally have completed and updated the first of my re-tests. I am sharing it here to see if you guys are cool with the GIFs I am using for HD and IMD. I also have readjusted my parameters for the maximum SPL testing; I used to test a much broader bandpass region so I was lax on the thresholds but now that I am testing a more typical bandpass region I have tightened the thresholds up (from 10% distortion to 3% distortion and from 3dB compression to 2dB compression).
Go check it out. LMK if the gifs are cool or if you'd rather have the still images. I prefer the gifs, myself. I don't really want to provide 6 images for HD and 6 more for IMD. The gifs make things simpler. If we are cool with it then I'll work on getting the other drivers' new data posted and I can finally get back on track with other tests.
https://www.erinsaudiocorner.com/driveunits/dyn_mw172/
here's an example of *one* of my IMD tests
Last edited:
OP
- Thread Starter
- #429
I don't think I do. Seems I've been there. But if I have an account there, I don't know about it.
OP
- Thread Starter
- #431
I’m just looking for feedback regarding using the gifs or not. The overall layout of the site is to my liking and I don’t have the time to overhaul it even if I wanted to.
Looks like I can't freeze the gif in Chrome, that would be nice (without copying it to some photo editor) There will always be someone who wants slower or faster flash rate...
About HTGuide diy, I think it's a very small community nowdays, almost one man's show. DIYAudio and ASR are enough in my mind. PE TechTalk forum perhaps too.
About HTGuide diy, I think it's a very small community nowdays, almost one man's show. DIYAudio and ASR are enough in my mind. PE TechTalk forum perhaps too.
GIF better in every way IMHO
I was referring to the gifs not your site. I should have been more clear. My mistake.
ernestcarl
Major Contributor
An alternative to GIFs when just comparing two images is putting the two side-by-side. But it will decrease the overall viewing size.
Another one would be partially transparent overlays of two or more images... but things can get confusing if you have too much stuff going on.
*maybe altering/inverting the colors and contrast in one of the images.
Many dedicated photo viewers can pause/play or skip forward or backward frames in a GIF... but within the web browser itself, it's something that can't be controlled -- or at least as far as I know -- without using some kind of viewer plugin/extension.
Another one would be partially transparent overlays of two or more images... but things can get confusing if you have too much stuff going on.
*maybe altering/inverting the colors and contrast in one of the images.
Many dedicated photo viewers can pause/play or skip forward or backward frames in a GIF... but within the web browser itself, it's something that can't be controlled -- or at least as far as I know -- without using some kind of viewer plugin/extension.
OP
- Thread Starter
- #436
Yea, having separate images would be helpful for direct comparisons. But then I'd have 12 images where 3 gifs would do the same thing. So, it's cleaner and takes up less space (storage and bandwidth). I can toggle the time to be 3 seconds each frame. Or, I try creating a video and uploading that to YouTube where people can pause/start whenever they want. But that's a lengthy process to have to go through each time.
OP
- Thread Starter
- #437
I just tried embedding a youtube video. Not sure I like that. There's no real workaround (anymore) for hiding the 'related' videos and the way YT frames the video with all their buttons is kind of annoying. And, when you pause the video the image gets blocked by all their crap buttons/banners. So, it defeats the purpose of having the video in the first place unless you go full screen.
You can see the video in action in my link above, but here's the direct copy:
You can see the video in action in my link above, but here's the direct copy:
ernestcarl
Major Contributor
In the case of youtube, it sort of gets annoying too that you have to manually click the reload button once the short animation ends. Again there's a plugin extension which fixes that -- but it's a workaround.
Enhancer for Youtube (Chrome)
I agree that the simplest method would simply be a gif animation with some reasonably set time frame change -- e.g. 1500-2500ms is probably good.
OP
- Thread Starter
- #439
Nerd status, but...
For the IMD testing, I have already provided test results at varying SPL with the goal to provide enough variance to be able to compare to at least 96dB. But, what is important with IMD is the physical excursion of the woofer when under test. That is why I test at 2 levels (one with a bass tone at Fs and the other with a bass tone at 80Hz).
I thought it would be as useful for people to know not just the SPL but what the excursion is for these tests. So, I found the equation that provides excursion based on SPL and Sd. I have implemented that in to my scripts to now provide the excursion for the bass tone. These excursion values are calculated but should be within 0.5mm of the actual excursion at the bass frequency I used for the test. Examples are attached.
For the IMD testing, I have already provided test results at varying SPL with the goal to provide enough variance to be able to compare to at least 96dB. But, what is important with IMD is the physical excursion of the woofer when under test. That is why I test at 2 levels (one with a bass tone at Fs and the other with a bass tone at 80Hz).
I thought it would be as useful for people to know not just the SPL but what the excursion is for these tests. So, I found the equation that provides excursion based on SPL and Sd. I have implemented that in to my scripts to now provide the excursion for the bass tone. These excursion values are calculated but should be within 0.5mm of the actual excursion at the bass frequency I used for the test. Examples are attached.
Thank you for the in depth review. It contains more information than just FR and Directivity, exactly what I want to see more.
Hope you get more great speaker to review, like ATC, then maybe we will have better understanding of what people arguing about.
Hope you get more great speaker to review, like ATC, then maybe we will have better understanding of what people arguing about.
Similar threads
