Surround type does have an effect on the damping at cone breakup frequencies yes, that much is well known, but it's not a huge effect, and my attempts to add additional damping to the surround on various drivers have been unsuccessful.
The problem with adding foam rings to the surround is you're inevitably going to affect the operation of the surround at lower frequencies - it's going to make the surround stiffer, affecting the Fs, and Qms, and it could make the suspension significantly more non-linear as well, increasing distortion at low frequencies.
I don't think it's necessary though because it seems much the same effect can be had by adding the damping to the cone just inside the perimeter of the cone, as I've done with my strips.
Even though they're attached to the cone and nothing else and they're quite light, they have enough inertia at high frequencies that the vibration from the bending cone compresses and stretches the foam strip against it's own inertia - thus absorbing energy as heat.
The advantage of doing it on the cone is that below cone breakup frequencies the only change in response of the driver is a slight reduction in sensitivity and Fs, and a slight increase in Qms, exactly as you'd expect from a small increase in mass. There is no interference with the operation of the suspension.
Another factor is that depending on how far from the edge of the cone you place the damping, you primarily affect the response at different frequencies.
For example on the Flat 8 putting the strips very near the edge affects about 3Khz, another 10mm towards the centre affects about 4Khz, with 5Khz being around another 10mm towards the centre - about the position of the inner most decoupling ring.
Although it's all damping and does has a cumulative effect, you can target a particular frequency region for extra damping (such as ~4Khz, which is the trouble spot on the Flat 8) and by adjusting the location and number of strips tailor the frequency response until it's almost flat through that range - which I don't think would be possible simply by damping at the surround.
Too much damping all at the same radial distance from the centre can actually add a dip in the response, and in the Flat 8 Mk II I actually ended up alternating the distance from centre to strip between two positions for every alternate strip around the cone.
(Next time I have the drivers out of the cabinet I really should take some pictures of the strips)
The problem with adding foam rings to the surround is you're inevitably going to affect the operation of the surround at lower frequencies - it's going to make the surround stiffer, affecting the Fs, and Qms, and it could make the suspension significantly more non-linear as well, increasing distortion at low frequencies.
I don't think it's necessary though because it seems much the same effect can be had by adding the damping to the cone just inside the perimeter of the cone, as I've done with my strips.
Even though they're attached to the cone and nothing else and they're quite light, they have enough inertia at high frequencies that the vibration from the bending cone compresses and stretches the foam strip against it's own inertia - thus absorbing energy as heat.
The advantage of doing it on the cone is that below cone breakup frequencies the only change in response of the driver is a slight reduction in sensitivity and Fs, and a slight increase in Qms, exactly as you'd expect from a small increase in mass. There is no interference with the operation of the suspension.
Another factor is that depending on how far from the edge of the cone you place the damping, you primarily affect the response at different frequencies.
For example on the Flat 8 putting the strips very near the edge affects about 3Khz, another 10mm towards the centre affects about 4Khz, with 5Khz being around another 10mm towards the centre - about the position of the inner most decoupling ring.
Although it's all damping and does has a cumulative effect, you can target a particular frequency region for extra damping (such as ~4Khz, which is the trouble spot on the Flat 8) and by adjusting the location and number of strips tailor the frequency response until it's almost flat through that range - which I don't think would be possible simply by damping at the surround.
Too much damping all at the same radial distance from the centre can actually add a dip in the response, and in the Flat 8 Mk II I actually ended up alternating the distance from centre to strip between two positions for every alternate strip around the cone.
(Next time I have the drivers out of the cabinet I really should take some pictures of the strips)
While I'm not keen on dismantling my speakers right now just to take a picture of the back of the Flat 8 Mk II's I realised the Flat 8A's are just sitting on a shelf at the moment, so I've attached a quick snap of the rear taken with my phone, which gives some context to the location and size of the damping strips.
The size, number of strips, and their angular position around the cone (8 equally spaced around 360 degrees) are the same as on the other drivers, but their distance from the edge is a bit closer.
On the other drivers every second strip is straddling the outer most decoupling ring - eg about 5mm further away from the edge of the cone than seen here.
Next time I have the other drivers out of their cabinets I'll try to remember to take some rear pictures of them as well.
The size, number of strips, and their angular position around the cone (8 equally spaced around 360 degrees) are the same as on the other drivers, but their distance from the edge is a bit closer.
On the other drivers every second strip is straddling the outer most decoupling ring - eg about 5mm further away from the edge of the cone than seen here.
Next time I have the other drivers out of their cabinets I'll try to remember to take some rear pictures of them as well.
Attachments
Last edited:
fullranges
Not many to be honest. Just missed a tecnics eas 6 inch vintage japanese full range loudspeaker 6 inches Pair (eBay item 270762032068 end time 14-Jun-11 14:25:16 AEST) : Electronics
I have a pair of these. They are very good. I bought a new set of fe103 drivers for my tqwt I think you call them needles over here. I wanted to make the work for my mum. But there still way too thin. I don't know what I done wrong with these speakers. I followed the plans closely.
I have been testing some axiom 110 10 inch fullranges. A lot more treble then I expected. And I pulled my axiom 301 out and installed in my lounge. They have nice bass but the whisser just messes things up somehow. Very good though and very good treble too.
I have been planing and buying stuff for my multi way horn system. This thread even reinforced to me that full range is an almost impossible goal. Whizzer or no whizzer.
Not many to be honest. Just missed a tecnics eas 6 inch vintage japanese full range loudspeaker 6 inches Pair (eBay item 270762032068 end time 14-Jun-11 14:25:16 AEST) : Electronics
I have a pair of these. They are very good. I bought a new set of fe103 drivers for my tqwt I think you call them needles over here. I wanted to make the work for my mum. But there still way too thin. I don't know what I done wrong with these speakers. I followed the plans closely.
I have been testing some axiom 110 10 inch fullranges. A lot more treble then I expected. And I pulled my axiom 301 out and installed in my lounge. They have nice bass but the whisser just messes things up somehow. Very good though and very good treble too.
I have been planing and buying stuff for my multi way horn system. This thread even reinforced to me that full range is an almost impossible goal. Whizzer or no whizzer.
Hi D,
Looks like you're doing allot of interesting work. Excellent to see this effort going into drivers - nicely done.
But a note of caution, surrounds often play a significant roll in damping extraneous frequencies and maintaing power-train stability. Much depends on the relationship the designer made (or choose) between the cone's mass, its flex ratio and profile matched to the surrounds damping specifications and linear excursion envelope.
As I've learnt from countless hours of development work, surrounds are a significant component in the design and operation of drivers. I'd err some caution when making comments like "not huge effect". It may be the case on the particular drivers you're currently using but its certainly not the case for many drivers.
I wish driver design and making were more simple. I've so many grey hairs on by head from all hours of brain work and experimentation.
Cheers
Mark.
Last edited:
You're right of course, the surround does have more effect than I suggested on the high frequency response, and my own measurements do show that when comparing two different Coral drivers which are the same cone design but have surrounds of very different physical properties.
There was a good discussion going on in another thread where I presented some of the same graphs as I did here, (plus an additional driver) but I think explained myself a bit better, starting at post #42:
http://www.diyaudio.com/forums/multi-way/192215-phase-plug.html#post2640287
Post #45 shows a diagram of the exact damping strip layout I used on the Flat 8 II:
http://www.diyaudio.com/forums/multi-way/192215-phase-plug.html#post2640803
Post #53 shows the Coral Flat 8A (the black coned driver I showed the photo of earlier in this thread) although the strip configuration during those measurements is actually different to the one shown in the photo:
http://www.diyaudio.com/forums/multi-way/192215-phase-plug-2.html#post2659727
Then in post #54 I showed measurements of a "defective" Coral 8A-100 with really bad cone breakup issues which I couldn't solve:
http://www.diyaudio.com/forums/multi-way/192215-phase-plug-2.html#post2659751
All three drivers have the same cone design but measure very differently.
In regards to the effect of the surround it's interesting to compare the Flat 8 II and the Flat 8A. The Flat 8 II has a very compliant un-doped reverse roll fabric surround with a fairly large radius of curvature which allows for an Xmax of around +/- 3mm.
They have really excellent bass response for a FR driver but as you can see the unmodified driver had severe cone breakup from 2-5Khz. It seems to me that the surround was a very poor termination for standing waves. Normally that type of driver would have a doped surround but these have no doping in the fabric of the surround - they're a hand assembled not factory assembled pair, so I assume whoever assembled them neglected to add the prescribed doping compound to the fabric which would have better terminated the cone. Great for bass response, not good for midrange.
Despite that, with the right pattern of foam strips (8 near the edge and 4 offset further in on this driver) I was able to almost completely eliminate the uncontrolled cone breakup without making any modification to the surround whatsoever, therefore not sacrificing any bass response.
My old before/after measurements aren't great but in post #45 I also presented a modern gated measurement taken a few months ago with ARTA. There is still a small bump at 4Khz that I EQ out, but it's nothing like the saw blade response of the unmodified driver. Now that I have a better measurement system I may have another go at optimizing the strip layout even further, as I think I can eliminate that 4Khz peak completely.
Now compare the situation with the Flat 8A - it's basically the same driver, same cone, same everything, the only thing that is significantly different is the surround, which is extremely stiff. It's also a reverse roll "fabric" surround, but the roll is fairly flat in curvature, and it appears to have been heavily doped with something that has gone hard with age (nearly 40 years) - perhaps varnish, who knows. They're not originally this stiff.
Point is it's so stiff that it pushes the Fs up from 38Hz of the other driver to 105Hz, and can barely manage +/- 0.5mm Xmax. As far as bass goes, this driver is a complete waste of time.
However look at the before modification graph - yes it has a 3dB dip at 3.2Khz and then a 6dB peak at 3.8Khz, but that's a hell of a lot better controlled than the unmodified version of the other driver which had a saw blade response... So yes, you're right that the surround makes a big difference, assuming that's the only variable that's different between the two drivers.
But then look at how flat the response is with good surround damping and offset damping strips - I was able to get it flat to better than +/- 2dB right through the cone breakup region without much trouble at all. In this case the strip configuration was opposite to the other driver - 4 strips near the edge and 8 strips further in.
My thoughts on all this are:
Yes surround damping does matter, however I don't think any practical amount of surround damping on it's own (especially if you still want the driver to produce bass) will give a fully optimal result, because it's only applying damping right at the perimeter.
Because any imperfection in the termination impedance will allow standing waves to form and the distance from voice coil to surround termination is equal at all rotations of the cone, the standing wave pattern on the cone will be the same all the way around the cone causing response peaks and dips at discrete frequencies.
On the other hand, by applying damping directly on the cone before the wave reaches the surround, it seems to lessen the importance of the surround damping dramatically, allowing the surround to be optimized more for bass performance.
Although the best midrange result was achieved with offset damping strips and good surround damping together, the offset damping strips with poor surround damping still gave a better result than the good surround damping by itself, and didn't sacrifice any bass performance.
Also, the offset damping strips allow you to apply discrete damping at more than one radius, and this seems to be key - on every driver I tried I always got the best results when the strips alternated between 2 different radii rather than all of them being at the same radius from the centre.
I think this is because it breaks up the remaining standing wave pattern differently at different angular rotations of the cone - thus there are no frequencies where all of the cone is experiencing the same standing wave pattern at once, so standing waves on different sectors of the cone tend to cancel each other out at a given frequency, a bit like B&W's FST Kevlar midrange unit. (I went into this in more detail in the other thread) No matter how good the damping at the surround is, it's always going to be at one radius at the cone perimeter, with less than perfect termination resulting in peaks and dips in the response.
I think from a practical design perspective you would make the damping of the surround as good as reasonably possible without sacrificing the bass performance or mechanical properties of the surround, and then additionally add some offset damping in the right pattern to fully optimize the response. Together I think it's possible to get an extremely good result that exceeds what is possible with surround damping alone.
Last edited:
Wow am I waking up an old thread here. It is a very interesting read, and since I am trying to find a way to "polish" a Visaton BG20... I wonder if one could also use treatment like this on whizzer cones. Or is it better to intruduce "cracks" on the edges like this thread suggests: http://www.diyaudio.com/forums/full-range/110257-whizzer-distortion.html ?
It is a very interesting read, and since I am trying to find a way to "polish" a Visaton BG20... I wonder if one could also use treatment like this on whizzer cones. Or is it better to intruduce "cracks" on the edges like this thread suggests: http://www.diyaudio.com/forums/full-range/110257-whizzer-distortion.html ?I would try a Visaton B200 to hear what is possible without a whizzer. I never heard the BG20, but the B200 is a really nice sounding/balanced wide bandwidth whizzer-less driver. I would go so far as to say that a lot of manufacturers know the tradeoffs of implementing a whizzer cone vs no whizzer and a lot offer both options (Fostex). This is not to say that you cannot modify a whizzer to your liking. The drawback is that mods that involve cutting or coating are irreversible. Also, it is hard to know when to stop even if you like the results and you end up with an unusable driver. So, unless you have a dozen pairs, a lot of time and measuring equiptment I wouldn't mess with mods.
for what its worth, a de-whizzered BG20's graph "looks good" and is employed in several Klang + Ton "Cheap Trick" projects - so it would be a choice of helper tweeter to consider
http://www.referenzboxen.de/hifi/ct230_en.htm]Strassacker: Speaker Building, Components[/url]
here's a pdf with measurements, 3rd order highpass and lowpass topology filters, etc.
file:///C:/Users/User/Downloads/CT%20232.pdf
Modpodge might not hurt things - I don't know the effect on BG20's whizzer
http://www.referenzboxen.de/hifi/ct230_en.htm]Strassacker: Speaker Building, Components[/url]
here's a pdf with measurements, 3rd order highpass and lowpass topology filters, etc.
file:///C:/Users/User/Downloads/CT%20232.pdf
Modpodge might not hurt things - I don't know the effect on BG20's whizzer
Last edited:
Phase plug and removed whizzer or phase plug and whizzer? Iwas also thinking of a different shape dustcap, if that wuold even work. Doesn't a phase plug usually make stuff more directional?
I wanted to do it gradually and measure between the steps I take. BTW, can one suggest what are the absolute minimum requirements to perform basic measurements?
I have heard and read nothing but praise for b200, so there must be something to it, it is just too expensive to buy for testing.
What is modpodge? How come a dewizzered bg20 needs a helper tweeter, but a b200 usually doesn't. It looks like it has a shallower cone. I am really interested in differences between them. VAS is close together, so they should be somewhat equally "hard", but the Q's of the b200 are all higher which must be where the soundcrap ends, but continues to ring in the bg20.
I'm guessing I could do something about it by eliminating the "bell modes" which I believe DBMandrake was doing by glueing absorbant materials on the inside of the cone in this thread. Plus isolation of the basket, and the BG20 should already become a very decent driver.
Just went through this whole thread for the first time. Some very interesting things to look ionto. I've always blamed the whizzer what audible no-no's I could hear in fullrange drivers. I ended up cutting the whizzers rightr off my old Pioneer 8" cheapies. However, my Electro-voice LS12's are a whole different class of speaker. Considering they are from around 1958 and still sound good is amazing in itself. I run them in OB's have few complaints. However, there is a dip between 2kHz and 4kHz that I'd like to try and minimize without cutting the whizzer. I did remover the dust caps and put a phase plug in. Maybe I'll play with sticking some foam around the last couple of out sections of the cone. I can reverse that if I choose.
Below is a recent measurement from 1 meter with one speaker. I have Hawthorne Augies hooked up (obviously not tuned in yet). Also have cheap tweeter at 10kHz and up, but not having enough output to keep up with the efficient EV drivers. Don't let the graph scare you off from these nice drivers. Open baffle closeup measurements can look weired. Anyway, it's easy to see the problem area between 2kHz and 4kHz.
![URL]](/community/proxy.php?image=http%3A%2F%2F%5BURL%3Dhttp%3A%2F%2Fs16.photobucket.com%2Fuser%2Fhurdy_gurdyman%2Fmedia%2FEV%2520LS12%25201-meter%2520left%2520speakerw-Augie%2520paper%2520cone%2520tweeter%25201%252048th%2520smoothing%2520white%2520noise%25202%252011%25202017.jpg.html%5D%5BIMGDEAD%5Dhttps%3A%2F%2Fi16.photobucket.com%2Falbums%2Fb29%2Fhurdy_gurdyman%2FEV%2520LS12%25201-meter%2520left%2520speakerw-Augie%2520paper%2520cone%2520tweeter%25201%252048th%2520smoothing%2520white%2520noise%25202%252011%25202017.jpg%5B%2FIMGDEAD%5D%5B%2FURL%5D&hash=80f45bf9b53bfb0c240c2511ab4e1e4e)
Below is a recent measurement from 1 meter with one speaker. I have Hawthorne Augies hooked up (obviously not tuned in yet). Also have cheap tweeter at 10kHz and up, but not having enough output to keep up with the efficient EV drivers. Don't let the graph scare you off from these nice drivers. Open baffle closeup measurements can look weired. Anyway, it's easy to see the problem area between 2kHz and 4kHz.
I'm not sure at all if I can hear it seeing as there is no way to take it out and compare. As for the measuring part, I plan on redoing all my measurements using pink noise. I used white noise without realizing I'm getting more treble and bass than what's really there. Preliminary results shows there is a lot less treble present than what I was measuring. The notch is still present, though. Now, all I need is a good high efficiency tweeter...
Dave
Dave
Do Marc Wauters Lowther $0.98 tweak with different density materials to find what best overall smooths out the ~1 kHz-up BW: https://web.archive.org/web/20001015053453/http://www.goodnet.com/~darmah/lowther/mods.htm
GM
Last edited:
Thanks for the link, GM. I remember this tweak from back in the days before I went coaxial. Just slipped my memory. I tried this on my old Pioneer, but finally ended up removing the whizzer. I guess you can't expect perfection from a $25 speaker.
I am not going to cut the whizzer on the EV's. These will only have reversible mods. We'll see what happens.
Dave
![URL]](/community/proxy.php?image=http%3A%2F%2F%5BURL%3Dhttp%3A%2F%2Fs16.photobucket.com%2Fuser%2Fhurdy_gurdyman%2Fmedia%2FEV%2520LS12%252012ft%2520w-Utah%2520horn%2520w-15%2520Augies%2520XO-120Hz%2520PN%25202-26-2017.jpg.html%5D%5BIMGDEAD%5Dhttps%3A%2F%2Fi16.photobucket.com%2Falbums%2Fb29%2Fhurdy_gurdyman%2FEV%2520LS12%252012ft%2520w-Utah%2520horn%2520w-15%2520Augies%2520XO-120Hz%2520PN%25202-26-2017.jpg%5B%2FIMGDEAD%5D%5B%2FURL%5D&hash=5cdfc010211714da87a325f3c93b57a7)
I do have a single LS12 with a ceramic magnet in it. I think I'll dig it out and see if the bass damping and treble extension may be better. This driver is in excellent condition. One never knows...
Guess I'll retire from this topic until a future time.
Dave