This is the fifth Tang Band driver I have modified and posted details to this forum. I have yet to find anything that indicates that the company Tang Band knows what they are doing. They seem to be assembling parts and calling the result a loudspeaker driver because they can and not because they are trying to achieve some acoustical goal. Clearly they are trying to make money, but other than that I cannot detect any integrated design work.
This driver does use a polypropylene cone, aluminum phase plug and a "rubber" surround, but beyond that it is a TV or portable radio loudspeaker driver. The basket is stamped steel intended for rear baffle mounting and even has the mounting gasket already glued on the front. It may be a good portable radio loudspeaker driver, but it falls short of their claim of being hi-fi.
For those who have already bought the driver, I can offer improvements to increase the upper range clarity and smoothness. I can even, with extensive modifications bring it to hi-fi performance, but it is difficult to do, requires precision that few possess, and is entirely too sensitive to mounting and enclosure interactions. If you have not already bought this driver, my best advice is to not buy it.
As with a couple of my other driver modifications, the application of lines of Gem-Tac glue can reduce some upper frequency cone vibration modes. This is accomplished easily with two U shaped lines and two shorter arcs of glue as shown the photograph.
The more critical and intractable problem is the unevenness of response below three kHz. These problems are only partially due to the cone. Instead of a cone vibration mode, they are caused by a backside output reflection bleeding through the cone. The reflection is bouncing off the cup formed in the basket to mount the spider and magnet assembly.
While I have been able to minimize the reflection, the modification is extensive and difficult. First the basket must be nibbled away to allow room for mounting a complexly shape baffle/deflector assembly to be installed both between the cone and the spider and under the spider. This severely weakens the basket structure and makes the driver extremely sensitive to misalignment due to external shocks.
There is also little room on the backside of the cone and this part of the modification requires great precision in fabrication and placement to allow full excursion of the driver. This task, on an already assembled driver tests the limits of my skills. For others who need instructions on how to remove a dust cap on a poly cone, it is clearly beyond their skills.
Finally, to preserve the sub three kHz gains in performance, special care must be taken in mounting and enclosing the driver. You can neither front nor rear mount the driver to maintain hi-fi levels of performance. Instead, using a baffle the exact thickness of the mounting rim, you have to flush mount the driver both front and rear. The only way to do this is to glue the driver to a basket sized and shaped cutout in the baffle. Then you need to plug the basket mounting holes. Finally, the enclosure must be large in volume to avoid both reflections coming through the cone and also backside loading the cone.
If you are willing and capable of doing all of that, you can have a 2X3 driver of hi-fi quality. Unfortunately, you still have to give up low frequency performance with the small cone area and you have less high frequency performance than with certain three and four-inch drivers that I know of.
In sum, not recommended even with modification. Please keep in mind that I have been working with Tang Band drivers as part of a driver development project. I believe I have learned everything I can from Tang Band drivers. As a result, this is likely my last TB modification. If you can benefit from what I have learned about this small selection of TB drivers, then I am glad I have taken the time and trouble to share.
Although not in great supply, I wish you good designing,
Mark
This driver does use a polypropylene cone, aluminum phase plug and a "rubber" surround, but beyond that it is a TV or portable radio loudspeaker driver. The basket is stamped steel intended for rear baffle mounting and even has the mounting gasket already glued on the front. It may be a good portable radio loudspeaker driver, but it falls short of their claim of being hi-fi.
For those who have already bought the driver, I can offer improvements to increase the upper range clarity and smoothness. I can even, with extensive modifications bring it to hi-fi performance, but it is difficult to do, requires precision that few possess, and is entirely too sensitive to mounting and enclosure interactions. If you have not already bought this driver, my best advice is to not buy it.
As with a couple of my other driver modifications, the application of lines of Gem-Tac glue can reduce some upper frequency cone vibration modes. This is accomplished easily with two U shaped lines and two shorter arcs of glue as shown the photograph.
The more critical and intractable problem is the unevenness of response below three kHz. These problems are only partially due to the cone. Instead of a cone vibration mode, they are caused by a backside output reflection bleeding through the cone. The reflection is bouncing off the cup formed in the basket to mount the spider and magnet assembly.
While I have been able to minimize the reflection, the modification is extensive and difficult. First the basket must be nibbled away to allow room for mounting a complexly shape baffle/deflector assembly to be installed both between the cone and the spider and under the spider. This severely weakens the basket structure and makes the driver extremely sensitive to misalignment due to external shocks.
There is also little room on the backside of the cone and this part of the modification requires great precision in fabrication and placement to allow full excursion of the driver. This task, on an already assembled driver tests the limits of my skills. For others who need instructions on how to remove a dust cap on a poly cone, it is clearly beyond their skills.
Finally, to preserve the sub three kHz gains in performance, special care must be taken in mounting and enclosing the driver. You can neither front nor rear mount the driver to maintain hi-fi levels of performance. Instead, using a baffle the exact thickness of the mounting rim, you have to flush mount the driver both front and rear. The only way to do this is to glue the driver to a basket sized and shaped cutout in the baffle. Then you need to plug the basket mounting holes. Finally, the enclosure must be large in volume to avoid both reflections coming through the cone and also backside loading the cone.
If you are willing and capable of doing all of that, you can have a 2X3 driver of hi-fi quality. Unfortunately, you still have to give up low frequency performance with the small cone area and you have less high frequency performance than with certain three and four-inch drivers that I know of.
In sum, not recommended even with modification. Please keep in mind that I have been working with Tang Band drivers as part of a driver development project. I believe I have learned everything I can from Tang Band drivers. As a result, this is likely my last TB modification. If you can benefit from what I have learned about this small selection of TB drivers, then I am glad I have taken the time and trouble to share.
Although not in great supply, I wish you good designing,
Mark
An externally hosted image should be here but it was not working when we last tested it.
So are you backing away from your previous stance on Tangband or is it just this particular driver that frustrates you. I notice that you appear to have been working with drivers that PE carries. Have you worked with other models such as the W3-871S, W4-657S or W5-704S?
I do appreciate the time you've spent working with these drivers and how much you've been willing to share. Thank you.
You are most welcome.
My previous stand on Tang Band was that I liked the company and that they made interesting drivers. That hasn't changed. It is just that I have neither the time nor the energy to go through their entire catalog. Worse, I believe it likely that I could improve every driver they make. Which means that Tang Band could improve their drivers if they cared to. How could they miss a 12 db Everest of a peak in the W4-656? How could they care about the performance of their drivers when they have such a huge resonance well within the claimed frequency response and still release the driver? As I have shown, there is a lot of potential in Tang Band drivers, it is just that you have to be a budding loudspeaker designer to find it.
Now, I am a complex person and the reasons I shared what I have done with the Tang Band drivers are just as complex. One of those reasons was the amount of what I call myth that exists here and in other forums about what matters in drivers. I could have attacked each myth or I could put information out there by taking drivers that don't have those "qualities" and changing what does matter until they are just as good as anything else. To a large extent, I have suceeded. If the readers haven't been able to see that for themselves, then my attacking the myths they believe in would not have accomplished anything either.
You noticed that I focus on PE Tang Band and I agree that I have tended to focus on Parts Express stock. I live just 1.5 hours away from PE. Ground shipping is next day for me. Kind of hard to ignore that convenience. While I have bought samples directly from Tang Band, the cost of air shipping from China is prohibitive.
While I have no examples of the other two drivers you mention, I have worked with the 871 and I can modifiy the cone to eliminate the high frequency cone breakup. The modification, however, is too complex to be practical. Besides, we should be mere months away from the publication of my W3-881 article and the modified performance will top the 871 by a considerable margin (and be relatively easy to do to). Further, that modification will apply to any of the TB PPM three-inch that use the same short former voice coil as the 881. You should be able to take advantage of the modification even if you do not have access to Parts Express.
It is not so much that I am frustrated with Tang Band as it is that I have other interests. My next driver development project is a transient coherent five or six-inch nearly full-range. With a six inch cone I know we can get flat bass response to below 50 Hz, and if I can hold the top end to -3 db at 18 kHz, then I would be willing to call it full-range. I do not know if I can do it. There are several fundamental questions about what controls the top end of a non resonant cone structure that are beyond the mass and inductance loading you find in the text books. I will need to go beyond the Tang Bands to test my theories. I will have to be making these from scratch. Can't do that and continue testing Tang Band drivers at the same time. And since I believe I am at the limit of what I can learn using Tang Band drivers as a test bed, I will be leaving Tang Band behind.
Best,
Mark
Tang Band should hire you 
Anyway thanks for posting all your results, I found them interesting, and somewhat usefull (dont have the exact models you have modified, maybe I will buy them though).
"Further, that modification will apply to any of the TB PPM three-inch that use the same short former voice coil as the 881."
Does this apply to the W3-926S?
thanks
Anyway thanks for posting all your results, I found them interesting, and somewhat usefull (dont have the exact models you have modified, maybe I will buy them though).
"Further, that modification will apply to any of the TB PPM three-inch that use the same short former voice coil as the 881."
Does this apply to the W3-926S?
thanks
I feel you are expecting a little too much from these....almost sound cocky or something.
Have you looked at any Goldwood drivers?? I think your statement may apply more towards their drivers that Tang Band.
I think we should be more or less happy that we are able to clean up a "cheap" driver so effectivly instead of bashing the company. I do like your findings though...and your proof to back it up.
I will be interested in your take on 881s's...as I really enjoy mine so far. Although I will not hack up the frame I will be interested in seeing what can be extracted.
Have you looked at any Goldwood drivers?? I think your statement may apply more towards their drivers that Tang Band.
I think we should be more or less happy that we are able to clean up a "cheap" driver so effectivly instead of bashing the company. I do like your findings though...and your proof to back it up.
I will be interested in your take on 881s's...as I really enjoy mine so far. Although I will not hack up the frame I will be interested in seeing what can be extracted.
Thomas,
Does the W3-926 use a PPM or a paper cone? The spec sheets lists "black PP" in one place and states "paper cone" in another?
I cannot tell if this driver has the long or short former. TB driver dimensions do not state the height of the basket. I have asked TB about this and the people I can contact do not know the drivers that use the short or long former.
Hybrid,
This is, however, an excellent example of my characterizations of the company. The critical variable that determines the use of the short or long former is the height of the basket. The long former attached to the 53 mm PPM cone causes a very nasty peak that the short former does not. I have verified this by dissassembling and swapping parts. The problem follows the former.
Why does TB offer so many baskets for their 3 inch drivers? It does not seem to be pricing, it is not acoustical performance. They have both short and tall formers in die cast and stamped and every shape that is available. And that is what TB is doing. They are taking every part they can get their hands on and assembling it using every other part they can get their hands on and offering it for sale. This produces an extensive and interesting catalog that will surely appeal to everyone as long as acoustic performance is not a determinant selection variable.
Is saying this bashing the company? I do not see it that way. I am just describing how the company works based upon available information. I have no problem with the company operating this way because I am both cocky enough and skilled enough to out design them. The problems they sell are problems I can cure. When building only for myself and not trying to help others, I can even correct the problems with the 23-972.
If, however, it really bothers you that I have bashed Tang Band, then I apologize. I am certainly not promoting Goldwood over Tang Band and you, of course, were not bashing Goldwood because you would not do such a thing. Still, I believe I have demonstrated a serious and consistent problem with TB drivers that only someone of my cockiness and skills can deal with. I have never made any secret that I am pursing accuracy of reproduction along very explicit design goals. I have also stated repeatedly that if someone likes the sound they have and that is all they want, then they do not need anything that I could possibly offer.
As for the 881s, you can already see what I extracted on the system pics forum. Should be in the first third of the forum. I show before and after and a limited selection of driver variability. What I don't share is the details of the modification procedure. That is for the brick and mortar article.
Timn8ter,
Are you still checking in on this forum? Do you own the W4-657S or W5-704S? If you do, send me an e-mail. No promises, but we may be able come to an arrangement of mutual benefit.
Mark
Does the W3-926 use a PPM or a paper cone? The spec sheets lists "black PP" in one place and states "paper cone" in another?
I cannot tell if this driver has the long or short former. TB driver dimensions do not state the height of the basket. I have asked TB about this and the people I can contact do not know the drivers that use the short or long former.
Hybrid,
This is, however, an excellent example of my characterizations of the company. The critical variable that determines the use of the short or long former is the height of the basket. The long former attached to the 53 mm PPM cone causes a very nasty peak that the short former does not. I have verified this by dissassembling and swapping parts. The problem follows the former.
Why does TB offer so many baskets for their 3 inch drivers? It does not seem to be pricing, it is not acoustical performance. They have both short and tall formers in die cast and stamped and every shape that is available. And that is what TB is doing. They are taking every part they can get their hands on and assembling it using every other part they can get their hands on and offering it for sale. This produces an extensive and interesting catalog that will surely appeal to everyone as long as acoustic performance is not a determinant selection variable.
Is saying this bashing the company? I do not see it that way. I am just describing how the company works based upon available information. I have no problem with the company operating this way because I am both cocky enough and skilled enough to out design them. The problems they sell are problems I can cure. When building only for myself and not trying to help others, I can even correct the problems with the 23-972.
If, however, it really bothers you that I have bashed Tang Band, then I apologize. I am certainly not promoting Goldwood over Tang Band and you, of course, were not bashing Goldwood because you would not do such a thing. Still, I believe I have demonstrated a serious and consistent problem with TB drivers that only someone of my cockiness and skills can deal with. I have never made any secret that I am pursing accuracy of reproduction along very explicit design goals. I have also stated repeatedly that if someone likes the sound they have and that is all they want, then they do not need anything that I could possibly offer.
As for the 881s, you can already see what I extracted on the system pics forum. Should be in the first third of the forum. I show before and after and a limited selection of driver variability. What I don't share is the details of the modification procedure. That is for the brick and mortar article.
Timn8ter,
Are you still checking in on this forum? Do you own the W4-657S or W5-704S? If you do, send me an e-mail. No promises, but we may be able come to an arrangement of mutual benefit.
Mark
w3 926s former length
I took apart one of my 926s and measured the former length to be 21.5 mm, would this be considered a short fomer? I assume it is the larger former since the w3-881s former is around 17.5mm. I phoned tangband the other day to find out what happened to the shorter basket, they told me the tooling had broken and is not repairable. However, they did mention the possibility of using a shorter basket using stamped steel. I assume that if all other components are kept the same that maybe this version would still have the possibilities for modification as the original w3-881s. What do you think Mark?
I took apart one of my 926s and measured the former length to be 21.5 mm, would this be considered a short fomer? I assume it is the larger former since the w3-881s former is around 17.5mm. I phoned tangband the other day to find out what happened to the shorter basket, they told me the tooling had broken and is not repairable. However, they did mention the possibility of using a shorter basket using stamped steel. I assume that if all other components are kept the same that maybe this version would still have the possibilities for modification as the original w3-881s. What do you think Mark?
Bringing this thread back from the dead a bit, aren't we?
Sounds interesting...
May I ask, why would you want to push a 5" or 6" driver down below 50hz?. Wouldn't a bigger driver be better equipped to handle the bass range, leaving the 6" to cover from there on up.
At any rate, it'd be nice to get an update on current developments
Sounds interesting...
May I ask, why would you want to push a 5" or 6" driver down below 50hz?. Wouldn't a bigger driver be better equipped to handle the bass range, leaving the 6" to cover from there on up.
At any rate, it'd be nice to get an update on current developments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.