Hello Steve,
Yes 3k is visible (the yellow horizontal line). It is the file opened with Audacity.
With the spectrum tool still visible but there are other peaks
This is an other portion
Here is an other time. The 3k is lower
Could you remind which side is when? Sorry I can't understand it from the record (pink noise masks your English to my French hear!)
Have you tried some modifications like an hole at the exciter position to see if the peak disappear?
Christian
Have you made some modification on this panel li
Hello Eric,
Updating the history file, I read again your post. This makes a connection with an hypothesis I made post 6639 : is there a limit in the 1st mode under which some echos appear? The link with your post is : those XPS or EPS panels you tested, is the difference in the material or in their 1st mode (low 1st mode with plywood need large surface, easier to get with XPS/EPS?)
Christian
PS : it seems fitting with hkguy6 experience. see #6348
@Eucyblues99 , @BurntCoil
Hello
In parallel updating the history file I have already posted, I have opened a github repository (new tool for me, usually dedicated for code) to store the files posted here. The intention is everybody can find them easily, starting by the history file of this thread. I have an history file of an other thread also.
Thinking about documents already posted, I first thought to yours : Tall blonde story, Eucy's dome. Would you mind I store them under this github link? For the posterity and the benefit of all
Christian
Hello
In parallel updating the history file I have already posted, I have opened a github repository (new tool for me, usually dedicated for code) to store the files posted here. The intention is everybody can find them easily, starting by the history file of this thread. I have an history file of an other thread also.
Thinking about documents already posted, I first thought to yours : Tall blonde story, Eucy's dome. Would you mind I store them under this github link? For the posterity and the benefit of all
Christian
Hi Christian. Thank you for asking. I have no objection@Eucyblues99 , @BurntCoil
Hello
In parallel updating the history file I have already posted, I have opened a github repository (new tool for me, usually dedicated for code) to store the files posted here. The intention is everybody can find them easily, starting by the history file of this thread. I have an history file of an other thread also.
Thinking about documents already posted, I first thought to yours : Tall blonde story, Eucy's dome. Would you mind I store them under this github link? For the posterity and the benefit of all
Christian
Burnt
Hi Christian.
The left microphone was pointing towards the panel and the right microphone pointing away.
I was crouching below the panel trying to keep out of the way of the microphone.
I was using one hand to turn the panel around which was suspended on fabric ( not very scientific).
This was only meant to be listened to, so you could hear the offending sound.
As this is DML , as I turn the panel the peaks and troughs will appear and disappear from one cm to the next.
But the 3k peak will stay constant from the rear.
The 3k peak cannot be seen from the front as it has dropped below the pink noise in my plots.
The recording starts from the front of the panel, the 8 vertical lines in your plots show where the panel has been turned around .
I made the recording because I wanted everyone to hear the offending noise for themselves, not just see a peak on a graph.
The panel is just strips of crate ply glued together.
There is a 2mm hole I drilled in the centre to remove the 10k peak which I would normally fill with a very small amount of bluetack ,on ply.
I did not finish this experiment or take pictures so I have no before and after photos.
The 10k peak might have been low on this panel to start with ?
Steve.
The left microphone was pointing towards the panel and the right microphone pointing away.
I was crouching below the panel trying to keep out of the way of the microphone.
I was using one hand to turn the panel around which was suspended on fabric ( not very scientific).
This was only meant to be listened to, so you could hear the offending sound.
As this is DML , as I turn the panel the peaks and troughs will appear and disappear from one cm to the next.
But the 3k peak will stay constant from the rear.
The 3k peak cannot be seen from the front as it has dropped below the pink noise in my plots.
The recording starts from the front of the panel, the 8 vertical lines in your plots show where the panel has been turned around .
I made the recording because I wanted everyone to hear the offending noise for themselves, not just see a peak on a graph.
The panel is just strips of crate ply glued together.
There is a 2mm hole I drilled in the centre to remove the 10k peak which I would normally fill with a very small amount of bluetack ,on ply.
I did not finish this experiment or take pictures so I have no before and after photos.
The 10k peak might have been low on this panel to start with ?
Steve.
Hi Christian@Eucyblues99 , @BurntCoil
Hello
In parallel updating the history file I have already posted, I have opened a github repository (new tool for me, usually dedicated for code) to store the files posted here. The intention is everybody can find them easily, starting by the history file of this thread. I have an history file of an other thread also.
Thinking about documents already posted, I first thought to yours : Tall blonde story, Eucy's dome. Would you mind I store them under this github link? For the posterity and the benefit of all
Christian
All fine by me
Hoping someone will try a dome sometime and report back ......
Cheers
Eucy
Eucy.
Further back on this forum I have made various types of domes, showing measurements.
I went through my egg shell phase, but also plastic, various aluminium foil thicknesses, paper, and thin card.
I demonstrated how I made them.
The picture is of an old eps panel with a concave eggshell dome.
I usually try to attach the dome to the coil foot for better hf.
Also someone further back used the bottom of beer cans, I believe, to make a cone shape tweeter which was glued to the exciter coil area.
Steve.
Further back on this forum I have made various types of domes, showing measurements.
I went through my egg shell phase, but also plastic, various aluminium foil thicknesses, paper, and thin card.
I demonstrated how I made them.
The picture is of an old eps panel with a concave eggshell dome.
I usually try to attach the dome to the coil foot for better hf.
Also someone further back used the bottom of beer cans, I believe, to make a cone shape tweeter which was glued to the exciter coil area.
Steve.
Attachments
Hello Eric
Clear
The 3k is visible from the rear. I had the same measuring the rear of my plywood panel. A bit lower in frequency. I think the source is the air between the panel and the spider (a kind of flat horn?). In this panel, the exciter is on a 10cm x 10cm piece of plywood. I came to the idea to change this spine to a more conventional one. It is also the reason to propose a "spacer" between the coil and the panel to increase the distance, lower the frequency and hopefully its level. None of those ideas have been tested for now. An other idea is why not putting some absorbent (a piece of Magic sponge).
One information to collect is to understand if exciter with spider like the DAEX25HFE is more sensible than one with "arms" as spring. I think hkguy6 wrote something close to that.
Christian
Thank you Eucy. The idea is in good position of my "do list" but I don't have much time for testing. Be patient, sure somebody will test it.
Christian
Last edited:
Anything that moves on the exciter can cause noise.
I think you will find that the origin of the 2k to 3k noise is from the coil former cavity.
A lot of pressure can build up in this area.
Different panel materials make a lot of difference, more reflective materials will make it worse.
When I use my light fabric material dome, which allows the coil cavity to breath , the 3k noise drops below the pink noise .
This is handy as it removes the 10k approx peak as well.
I sometimes see other noise peaks which could be spider noise or coil resonance.
The other types of spider are pretty stiff and used to support the heavier exciters, these too will buzz as they are attached to the coil former.
I really liked the idea of the concave eggshell tweeter , but this reduced the cavity area and increased the pitch above 10k and was also quite reflective.
This gave the concave eggshell tweeter a harshness that I did not like.
The dome eggshell tweeters were harder to make but still sounded a little hard to me.
The soft mesh domes that I use are not high output tweeters,depending on the efficiency of the panel of course. but more to do with balancing between the output and the coil area breathing.
I did make a phase plug as an alternative, which works as well , but you loose hf output from the coil surface area.
I want my cake, and eat it 😃
Steve.
I think you will find that the origin of the 2k to 3k noise is from the coil former cavity.
A lot of pressure can build up in this area.
Different panel materials make a lot of difference, more reflective materials will make it worse.
When I use my light fabric material dome, which allows the coil cavity to breath , the 3k noise drops below the pink noise .
This is handy as it removes the 10k approx peak as well.
I sometimes see other noise peaks which could be spider noise or coil resonance.
The other types of spider are pretty stiff and used to support the heavier exciters, these too will buzz as they are attached to the coil former.
I really liked the idea of the concave eggshell tweeter , but this reduced the cavity area and increased the pitch above 10k and was also quite reflective.
This gave the concave eggshell tweeter a harshness that I did not like.
The dome eggshell tweeters were harder to make but still sounded a little hard to me.
The soft mesh domes that I use are not high output tweeters,depending on the efficiency of the panel of course. but more to do with balancing between the output and the coil area breathing.
I did make a phase plug as an alternative, which works as well , but you loose hf output from the coil surface area.
I want my cake, and eat it 😃
Steve.
Thanks Steve
Yes, I realise you've done this many times with different materials. Also as you describe sometimes with different aims ( noise reduction). (BTW. I haven't experienced any of the exciter noise issues you've found .... Could it be specific to the exciters you use?)
However I'm thinking of this as a specific 'recipe' approach which if beneficial will/may negate the need for such wide ranging experimentation.
Also there's no hole in the panel so it's a test of the effect of a sealed dome not only on hf but on vocal clarity
I know it works on 3mm ply and corflute, and 4mn cedar but I haven't tried other materials.
So my aim is ultimately to get feedback from others on this specific approach with different materials.
It may not give any benefit when used on very thin card panels for example. I simply don't know at this stage.
In any event though, I'm convinced and will continue to use it
Cheers
Eucy
Last edited:
Yep.. Interesting... Similar idea, with different approach.. Coke (beer) cans to the rescue. 👍
I haven't tried this but as offgrid says, it would have merit.
Eucy
More on the subject of domes (aargh!)🤭
This idea has been rattling around in my mind for a while
See the attached sketch
Call it a domeless dome... Or an infinity dome... The walls of a Coke can are very thin and reasonably high tensile so I pondered on using a disc of this with the
can curve giving some extra stress stiffness to the disc when flattened
Route a shallow hole approximately 1mm deep in the panel of similar diameter to the voice coil. Thoroughly seal the cutout against air leakage.
Scissor cut a disc from the side of the can about 6 to 8mm bigger in diameter than the cutout
Maybe a thin smear ring of damping material around the half diameter position
Glue the disc to the panel
The final size of the components could be adjusted based on test results
This will provide an air damped thin membrane in the exciter location and may give similar results to the dome I proposed
Being flat it may possibly be affected by hf 'flapping' but the size can be reduced to compensate ..
Another in the to-try basket
Eucy
This idea has been rattling around in my mind for a while
See the attached sketch
Call it a domeless dome... Or an infinity dome... The walls of a Coke can are very thin and reasonably high tensile so I pondered on using a disc of this with the
can curve giving some extra stress stiffness to the disc when flattened
Route a shallow hole approximately 1mm deep in the panel of similar diameter to the voice coil. Thoroughly seal the cutout against air leakage.
Scissor cut a disc from the side of the can about 6 to 8mm bigger in diameter than the cutout
Maybe a thin smear ring of damping material around the half diameter position
Glue the disc to the panel
The final size of the components could be adjusted based on test results
This will provide an air damped thin membrane in the exciter location and may give similar results to the dome I proposed
Being flat it may possibly be affected by hf 'flapping' but the size can be reduced to compensate ..
Another in the to-try basket
Eucy
Attachments
Eucy.
Do you measure your panel and exciter response ?
The resonant peaks will only be set off buy certain type of music being played.
Some music tracks will sound fine but others may set off these resonances, causing harshness or sibilance(10k or so).
If using a flat metal disk in the voice coil area ,you are going to have to be careful not to create the tin can effect.
Thinning the area within the voice coil into a concave dome similar to my eps panels might be a good idea,and drilling a 2mm hole and filling with bluetack or something similar might help , I'm not sure.
I never got around to try this as ply is not my favourite ( that is until I found the crate ply ).
Although I have drilled 2mm holes in some ply panels.
My ordinary 3mm ply panels always measured with plenty of HF, even too much , but always sounded lacking in this area ?
The podiums sounded the same to me.
I look forward to hearing about you results.
Steve.
Do you measure your panel and exciter response ?
The resonant peaks will only be set off buy certain type of music being played.
Some music tracks will sound fine but others may set off these resonances, causing harshness or sibilance(10k or so).
If using a flat metal disk in the voice coil area ,you are going to have to be careful not to create the tin can effect.
Thinning the area within the voice coil into a concave dome similar to my eps panels might be a good idea,and drilling a 2mm hole and filling with bluetack or something similar might help , I'm not sure.
I never got around to try this as ply is not my favourite ( that is until I found the crate ply ).
Although I have drilled 2mm holes in some ply panels.
My ordinary 3mm ply panels always measured with plenty of HF, even too much , but always sounded lacking in this area ?
The podiums sounded the same to me.
I look forward to hearing about you results.
Steve.
I doubt the "breath" is the solution. DAEX25HFE is the one has much vents (around spider and also the back center for cavity). But it is the noiser exciter I have.
If the exciter has no vent, just drill a hole on the panel under the voice coil with sponge in the hole can give the breath. However I didn't hear much improve, or maybe my 10W no-vent exciter is quite noise-free I don't know.
Another thought, the DAEX25HFE is the one has larger "voice coil moving distance", I call it springly. Maybe it's the nose reason.
Besides, it fabric spider is most soft (actually not the fabric soft, it's because the voice coil : spider size ratio too large, make there are many fabric area). Maybe stiffen the fabric spider will help. Such as apply super glue on some fabric area.
Besides, it fabric spider is most soft (actually not the fabric soft, it's because the voice coil : spider size ratio too large, make there are many fabric area). Maybe stiffen the fabric spider will help. Such as apply super glue on some fabric area.
One more case I remember. When the DAEX25HFE stay on a panel too long (I always changing exciter between panels). Listen it over a month for example. The exciter will becomes more slant(head down) by gravity. This make the top part of spider facric loose. The voice coil may get friction by this slant and cause noise. If this happen I hang the panel upsidedown, the noise will gone. But not last long until the fabric loose again.
So, when the panel not in use. I always place them upsidedown to balance the fabric loose.
A spine system can solve this gravity problem. However if the spine can't give a perfect horizontial placement for the exciter. It means the voice coil always slant, and no free space to move. It may even worst then the free hang.
So, when the panel not in use. I always place them upsidedown to balance the fabric loose.
A spine system can solve this gravity problem. However if the spine can't give a perfect horizontial placement for the exciter. It means the voice coil always slant, and no free space to move. It may even worst then the free hang.
Hi Steve
Yes, I measured the Dickens out of the few panels I've built so far. I haven't found any problems relating to the exciter area.
I did have a certain hollowness in the cedar panels when testing with spoken vocals ie.. News reports etc. This was solved by application of very thin self adhesive felt patches to the rear surface of the panel in two selected areas. I mentioned this previously.
Sibilance is not a problem for me, but I have found that these panels will definitely pick up poor recordings and I am surprised by the number of recordings which have been adversely manipulated to create a particular effect.. Notable in my opinion is the excessive use of echo in many recordings.
The flat disc may not work...I put it forward as just another idea to try.
Nothing ventured, nothing gained.
I'm to busy preparing to move house to try it now... But I will asap.
Wrt ply, have you tried any poplar material.?.. If not I recommend you do. I was a ply sceptic, but poplar is a good performer. (Christian will agree I'm sure)
Cheers
Eucy