Tymphany custom OEM surplus drivers group buy

[Hari Iyer]

I just managed to assemble a pair of TABAQ-like MLTL cabinets with the Tymphany drivers for testing and auditioning. Here's a pic of one of them:

I did not use any stuffing and omitted the BSC network, just
for convenience to get a quick idea of the sonics.

My preliminary impressions: It works, and sub-100 Hz bass is
present in abundance. However, it is too bright in the upper-mids
and highs, which will have to be tamed with a combination
of light foam lining of the closed end of the pipe, as well as
a Baffle-Step Compensation (BSC) network.

@linuxguru
It will be interesting to see the impedance curve of the stuffed and unstuffed line. If you have the required gear and time, please post them too.

 

[Hari Iyer]

@Linuxguru
You can measure speaker system impedance with speaker workshop, ARATA, REW etc which are freeware, you may still require an impedance jig for doing so.

 

[linuxguru]

@Linuxguru
You can measure speaker system impedance with speaker workshop, ARATA, REW etc which are freeware, you may still require an impedance jig for doing so.

Thanks - I'll try to rig up an impedance and T/S parameter measurement jig as soon as I can, but it may take a few weeks until I rig up an old Pentium box with a good full-duplex Soundblaster or similar card for impedance measurements. Speaker Workshop seems perfect, but updates stopped in 2002, so I'll probably have to dig out a Win 95 or 98 install CD somewhere.

 

[linuxguru]

I managed to kludge together a BSC network for the folded MLTL based on the following posts at diyAudio:

TABAQ TL for Tang Band - Page 76 - diyAudio

I settled on 0.75 mH || 5 ohms || 0.47uF for the first prototype. In preliminary listening, the brightness in the upper-mids and highs is almost completely cured. The 0.47 uF can probably be omitted, but it seems not to hurt and gives a certain sparkle and air to the highs. Some small changes in values (larger L, larger R) may help - I'll run this in for a few weeks before deciding on further tweeks.

Here's a pic of the BSC network, kludged on a piece of perfboard:

The inductor is hand-wound on a locally-available mains-transformer bobbin. The resistors are Dale CW-2B wirewounds, and the capacitor (optional) is a Wima MKP10.

Edit: The line is still fully unstuffed, i.e. empty.

 

[baswamin]

Hi Linuxguru,

I was planning on building tabaq with this driver. It is good to see you are trying to build it yourself. This will help others who want to build the same using this driver. After your first trial, I was thinking about BSC and stuffing. BSC seemed to have cured the issue with the high's. I dont know if stuffing will only affect bass performance or overall balance.

Can you tell me if these drivers are available?

Thanks

 

[tnvijay]

siva sir,
pls try stuffing too and let all of us know the result. According to many previous TABAQ builders,BSC was dependent on personal taste and some did omit it.But almost all benefitted from stuffing (IMHO)

 

[linuxguru]

@baswamin: The drivers are available.

@tnvijay: I'll probably have to build another pair for experimenting with stuffing and for A-B comparison. The improvement in overall timbral balance with BSC alone was fairly remarkable, and these BSC values seem to be close to optimal - maybe L=1 mH may improve it slightly.

I think it's good enough now that I can temporarily shelve the idea of using dual 12-ohm drivers for increased effective area as well as power - the single 6-ohm driver can itself handle 30W, and it seems to have sufficient Xmax (2.5 mm). It is surprisingly loud for indoor listening even at modest power.

 

[tnvijay]

@linuxguru
How do you compare in terms of sonics compared to your technics 2 way?

 

[baswamin]

Cant you stuff by removing the driver from the box through the opening for speaker? Atleast you can stuff half the waveguide and see how it impacts. Also, was the inside edge smoothened (rounded inner edge of speaker opening)?

Thanks

@tnvijay: I'll probably have to build another pair for experimenting with stuffing and for A-B comparison. The improvement in overall timbral balance with BSC alone was fairly remarkable, and these BSC values seem to be close to optimal - maybe L=1 mH may improve it slightly.

 

[linuxguru]

Cant you stuff by removing the driver from the box through the opening for speaker? Atleast you can stuff half the waveguide and see how it impacts. Also, was the inside edge smoothened (rounded inner edge of speaker opening)?

Yes, the only way to stuff it with polyfill is through the speaker cutout, and only the 1/3 of the line at the closed end needs to be stuffed, if at all.

The speaker cutout has not been chamfered on the inside - I didn't want to risk weakening the MDF near the mounting flange of the speaker, though perhaps it can be done carefully with a file or dremel.

 

[linuxguru]

How do you compare in terms of sonics compared to your technics 2 way?

I haven't done an A-B comparison of these two, but will check within a few days.

 

[keith_correa]

I think a Bessel Array with these would make a killer CC speaker. Of course - you will need to be far away so definitely not for folks in small apartments [like me]...

 

[goldyrathore]

Would'nt the high CTC distance cause problems for any horizontal array?
BTW, not to digress, but can someone point me to directivity or polar plots for a Bessel array in the far field?

 

[linuxguru]

I think a Bessel Array with these would make a killer CC speaker. Of course - you will need to be far away so definitely not for folks in small apartments [like me]...

The same source from where I got the Tymphany drivers in this thread also has an oval tiny full-range speaker for notebook applications - it's about 0.5" x 1". 5 or 6 of those can be crammed into a small sound-bar type setup for a near-field Bessel array, not more than ~3-6 inches in width. These are rated at 8 ohms, 1W.

I haven't investigated it further because it is moderately expensive (just slightly lower than the Tymphany) and there are some specific issues related to the design of the baffle cutout - it has to be precisely machined or moulded to the dimensions of the driver, as well as driver spacing in the array.

If somebody wants to take up the design, I'd be happy to post the pics, dimensions, etc. of the driver. T/S parameters will have to be measured from samples, of course.

 

[goldyrathore]

Siva,

Can you pls post the impedance curve or Fs of this other driver, the driver is so small that T/S params are not much useful? Pls post a picture too, if possible along with a ruler.
I ve tested some very small fullrangers and observed that they dont have flat response and their highs roll off very early at somewhere near 12k.

Thanks.

 

[linuxguru]

Can you pls post the impedance curve or Fs of this other driver, the driver is so small that T/S params are not much useful? Pls post a picture too, if possible along with a ruler.

I don't yet have an impedance measuring setup, but will try to rig one up within a few weeks. I don't have the driver with me, but managed to take a pic earlier with my cellphone cam (which isn't very good at low-light or close-up pics):

You can get a rough idea of dimensions from the size of the 4-pin connector next to it. One detail that may not be visible is that the rear vent behind the cone is present only on the short edges - i.e. when placed side-by-side with the long edges abutting each other, the rear vents are not obstructed.

 

[viki_v2]

Driver seems good enough for the price.
I will be looking to purchase 4 of them.

If members are building a common design, I can help with 3d modelling to get enclosures CNC Machined.

Also I can get the enclosure CNC cut if members come together.
This way we can get neatly finished enclosures for cheap.

 

[linuxguru]

You're welcome to design and offer an enclosure based on the published specs if you wish. However, the earlier GB for a common enclosure designed by FM Hari Iyer hasn't really taken off - there was an expression of interest for about 2 pairs of cabinets, but it needs at least 5 pairs to be viable. Most buyers of the drivers seem to want to build their own enclosures or experiment with other stuff like Line Arrays or refurbishing existing speaker cabinets.

I have already arranged for the fabrication of an inexpensive small cubic sealed enclosure and another larger folded MLTL similar to the TABAQ box. You may want to look into the design of either a Cornu spiral or an advanced compact folded-horn or resonator or TL - anything that can be made very compact in the BS form-factor, like the Pico Lino 2 or Karlsonator. The Tymphany driver itself is very versatile, like its siblings in the Vifa line-up, the paper-cone TC9FD and the glass-fibre cone TG9FD.

 

[Vinay2013]

Hi Siva - It is great to see that you are trying different enclosures/designs for these drivers. A great help for people like me.

I think it would be great if you could start shipping the drivers so that other FMs can also start their projects/experimentation and share their experience.

Look forward to see shipping details soon.

Regards,
Vinay

 

[tnvijay]

Hi to all Open Baffle Gurus,
I have a pair of tymphany drivers bought from linuxguru.Since its a fullrange, is there a possibility of a OB? If yes, pls post dimensions .I can add a sub if necessary. TIA.