Sansui SP-3500 woofer too stiff

A few days ago a customer brought in a pair of SP-3500 speakers with rubbing voice coils in their woofers. Fortunately for him the rubbing was caused by misalignment, rather than his speakers being smoked. Unfortunately, they still needed to be surgically repaired, as there is no reliable way to fix the rub outside of removing and reinstalling the entire cone assembly.

In the process of determining the correct approach to restoring his model W-114, 14" woofers, it became clear that their suspensions were much too stiff for woofers. I checked their free air resonance and found it to be 75Hz!

The W-114 woofer uses a doped cloth, inverted half roll, suspension. While the "goo" that is used to dope the cloth is very similar to what Altec used on their woofers, the Altec dope does not stiffen up. The dope on the W-114s had dried out, thus their being overly stiff. Years of experience, and the knowledge that those suspensions had to go anyway in the process of rebuilding the W-114s to be rub free, prompted me to try to resoften the dope. Using a new (clean) acid brush, I applied a liberal amount of MEK (methyl ethyl ketone) to the cloth roll suspension and exercised the cone in and out while the MEK evaporated. I then retested the free air resonance of the speaker, and found it to have dropped to 42Hz!

I suspect that most any Sansui speakers of that vintage which used a woofer with a doped cloth suspension have lost some of their low frequency performance due to this stiffening. As MEK is almost certainly the solvent that was originally used to thin the dope, I suggest that anyone with old Sansui speakers whose woofers have that suspension investigate the possibility that their woofers have stiffened up. If so, those woofers can be treated as I've described to restore the resiliency of their suspensions.

TG

Sansui SP-3500 woofer too stiff: Addenda

If you decide to treat excess stiffness of your Sansui woofers with MEK I as I previously described, do remove them from the cabinet, prior to treatment, and set them upright on a work surface that you won't feel bad about damaging with spilled or dribbled MEK

TG

And be careful, MEK {methyl ethyl ketone} is a great solvent, however highly flammable, and definitely not good to inhale any fumes.
Burns in any breaks in your skin too.:thumbsdn:

Thanks for the advice!

Woofers stiffened up again

Well, it's been a few days since I treated the cloth roll suspensions on the W-114 woofers with MEK in order to restore their resiliency. I decided to test them again and, unfortunately, they are back to having their free air resonance at about 75Hz. Once the MEK entirely dissipated, they stiffened back up.

As I mentioned previously, these woofers had coil rubs, so they were destined for surgery anyway. I cut the suspension out, leaving just the spider, and the free air resonance dropped to about 26Hz. I'm going to rebuild them with a foam half roll, which I expect to raise the resonance to somewhere around 35 Hz.

I'll have their owner bring in one of his SP-3500 enclosures so I can measure the internal volume and the port dimensions. With that information, I can calculate the Theil-Small parameters needed for the woofers to get the best response. Hopefully, everything will line up without needing any further work, but if not, I can add a bit of latex to the foams to stiffen them and raise the resonance, or a bit of mass to the cone to lower it further, which ever would be needed, to get the tuning in order.

It's too bad the MEK trick didn't last. It was easy to do.

TG

MEK is extremely potent (and carcinogenic ie cancer causing) stuff. It usually melts or dissolves most things besides metals and wood. I'm surprised it didn't damage the actual suspension itself, especially as these would be a lot less sturdy than the day they were made. MEK may have been used as a thinning agent in the manufacture of the coating but to add it straight full strength to the suspension, doesn't sound good. It's similar to putting paint thinners (which is about 60% MEK co-incidentally) straight onto a surface which has been painted with a paint which had been originally reduced with thinners before it was applied. After a short while the paint will break up and dissolve.

Quite familliar with MEK

I've been using MEK in my speaker lab for over 30 years. My work space is cross ventilated and has a powerful exhaust fan.

Having said that, while the dangers of MEK are real, they are somewhat overblown. Carbon tetrachloride was banned as a cancer causing substance. If you knew to what extreme the government went to come to such a conclusion, you would probably feel safe in drinking the stuff. I know people that are now into their late 80s, who are as healthy as they have a right to be at that age, and who worked at companies like 3M where they had open 55 gallon drums of carbon-tet sitting along side production lines for dip cleaning.

As to MEKs solvent abilities:
MEK is the solvent of choice in almost all traditional speaker adhesives and dopes, and still is found in the bulk of such materials currently in use. Modern binders that do not incorporate MEK: Cyano-acrylate, epoxy and urethane.

The foam that is used for most woofer suspensions is not attacked by MEK. I've repaired many hundreds of speakers and never run into any suspension material that was attacked by MEK, other than some of the adhesives that bound them to the cone. Among the materials that are not affected by MEK: Every speaker cone paper I've ever seen in my many years performing factory service for JBL, Altec, Gauss, E-V, Celestion, Cerwin-Vega, etc. as well as polypropylene (Infinity for example) , fiberglass, etc.

The suspension material Sansui used on the W114 is cloth, probably cotton. It is not effected in the least by MEK. The dope they treated the cloth with to seal it and control its flex properties is solvable by MEK, but there was either some other much less volatile solvent contained in the dope that took years to evaporate or the material itself was originally naturally soft and pliable.

It's possible that they treated the cloth with something like polyisobutylene, which does not require a solvent to stay soft, although it is effected over time by ozone and other environmental hazards, but I'm not an organic chemist, so my speculations in that regard are untutored.

In any case, the suspensions of those old Sansui speakers, and probably those of Pioneer and other brand speakers as well, have not retained the required suppleness, having lasted decades beyond the expectations of their designers. If the dope can't be restored, or removed and replaced, the entire suspension should be replaced by one made of foam or some other physically similar (hopefully with more longevity than foam) material.

TG

Sansui SP-3500 woofer too stiff: Fixed

The W-114 woofer, being an odd size, does not match up with any of the many aftermarket foam rings I evaluated for a fit. After experimenting with a few different surround options, I settled on using the medium compliance foam that is intended for a JBL 128H 12" woofer. That surround has the proper inside and outside flange widths as well as the correct size half roll to fit the space between the cone and the frame. Unfortunately, its circumference is a bit more than 4" too small. To overcome this, I cut a piece from a second 128H surround and used it to fill the gap. It's not perfect as the smaller radius of those foams means that a bit of stretching and squeezing are in order to get the foam to work as required.

Never the less, with a careful approach to the process of gluing the foam pieces to the cone, and using a fairly fast set Bostic solvent based adhesive, the conversion turned out fairly well. The free air resonance of the first reworked W-114 is 36 Hz.

The SP-3500 enclosure is really too small for a 14" woofer, even one with a fairly high Vas of about 7.5 Cu ft, which is the value of the reworked woofer. The enclosures net volume is only about 1.3 cu ft. and it is tuned fairly high (about 45Hz). Consequently, the alignment is far from what you would expect from any modern design system, having about a 2dB rise in the 70 to 80 Hz range and starting to drop off at around 60Hz. However, it actually sounds pretty good with a fair selection of known reference discs. The misalignment was probably done on purpose in order to push the efficiency of the low frequency speaker high enough to balance with the very efficient mid range and high frequency components.

After the glue on second woofer sets, and I measure its parameters, I'll come up with a composite parameter set and post it, along with a photo of one of the woofers. Since there are two glued seams in the surround, they look obviously reworked, but I think they look reasonable, considering what had to be done.

TG

Has anyone furthered the work stated here previously over the last however many months? Treatments, modifications?

xo, Biggles