Modifying the Tannoy GRF horn cabinet- will this work?

[DeonC]

Hi guys

I have made a modified drawing of the Tannoy GRF horns. The basic difference is that I have enlargened the front chamber by adding another 100mm (4") to it. I have also extended the flare of the horn with the same amount (attached to this post will be the original design; attached the one that follows will be my modified design). A further difference is that I intend using it with a pair of Tannoy DU386 drivers. These are a domestic (read lower quality) version of the HPD386. The TSPs of these and the Monitor Gold (for which the GRF was designed) are different. I have a few questions for the horn fundis though:

1. Although the horn remains basically unchanged (the flare rate is the same, just the mouth is bigger and it is slightly longer), the compression chamber is much larger. Now this is not an issue as I can make the volume smaller (by adding some bracing, etc), but how big must it be to be optimal?

2. The mouth is a bit larger (about 20% larger)- how will this affect the low-end response? And how low will she go?

3. Will the DU386 work in these horns?

The TSPs of the two drivers are as follows:

Monitor Gold 15" (for which they were designed):
Vas: 350 litres
Fs: 26 Hz
Qts: 0.2
Qa: 2
Qe: 0.22
Mmd: 90 g
Cms: 0.42 mm/N
Sd: 0.077 m2
BL: 19.2 N/A
Re: 5.5 Ohm
P (thermal): 50 W
Sensitivity (@2.83V, 1m): 94.8 dB
Rms: 7.45

DU386 (my current drivers):
Vas: 0.483 m3
Fs: 22 Hz
Qts: 0.18
Qa: 2.4
Qe: 0.2
Mmd: 68 g
Cms: 0.58 mm/N
Mma: 22 g
Sd: 0.077 m2
BL: 18.7 N/A
Re: 5.7 Ohm
Z, efficiency into half-space (2-pi): 2.6 %
VD, Peak linear displacement volume: 0.00031 m3
a, effective piston radius: 0.156 m
P (thermal): 85 W

Thanks,
Deon

[DeonC]

Here is my modifed design. Note that the driver does not pass through the wall of the horn. The depth is based on measurements taken of my own DU386 driver.

[GM]

Greets!

While it's technically a horn, it's really just a TL, so model it in MJK's offset driver TL WS to see if it preforms acceptably to you.

GM

[DeonC]

Hi GM

...so model it in MJK's offset driver TL WS...

Could you point me in the right direction to get this program. I have been on the Transmission Line Speakers webite as well as Martin J. King's Quarter Wavelength Loudspeaker Design website, and the only program I saw was software to model T-lines in Mathcad. I don't have ANY experience with Mathcad so if this is it, then I will try to learn, but I was wondering if there are any other sites I can try?

Thanks,
Deon

[maxlorenz]

Great project!
I'm a Tannoy fan and I dream of making an "Autograph" for wich I already made a carton model based on the diagrams from:

http://www.users.bigpond.com/tunnelg...autograph.html

I'm only guessing but I presume that you will have even more LF from the horn and even midbass contamination due to bigger chamber diameter unless you manage to filter it (wich may render the speaker "slower" if wool or similar). I've seen freq response chart of DIY GRF wich show an ample bump below +/- 80 hz if I recall well.
Here's a link from a guy that used a "tear drop" to block midbass.

http://www.hilberink.nl/codehans/tannoy31.htm

I would try to make the internal panel binding non-permanent to fine tune them by ear

Please remember that I have never build anything like this before
I'm only trying to help.

Good luck!
Mauricio
A Tannoy fan

[GM]

Originally posted by DeonC
Could you point me in the right direction to get this program. I don't have ANY experience with Mathcad so if this is it, then I will try to learn, but I was wondering if there are any other sites I can try?

Greets!

?? MJK has a link to the MC8 demo, so download/install it and download his MC8 worksheets: http://www.quarter-wave.com/Models/MathCad_Explorer.zip

Then it's just a matter of double clicking on the various driver, speaker, box parameters and typing in yours. Since you can't save the design, copy/paste it into a MS WORD doc if need be.

Anyway, there's a bunch of folks familiar with them and if you do a search, a number of posts have answers to many Qs that have popped up, and of course you can ask specific Qs to the various lists.

GM

[DeonC]

Hi Mauricio

I'm only guessing but I presume that you will have even more LF from the horn and even midbass contamination due to bigger chamber diameter unless you manage to filter it

I was thinking of adding extra bracing to the chamber to reduce the size. This will have the added benefit of stiffening the cabinet too which is a good thing.

Here's a link from a guy that used a "tear drop" to block midbass.

I wonder if the same cannot be accomplished by simply lengtening the baffle behind the driver downwards a bit. That will make the gap feeding into the horn a bit.

I dream of making an "Autograph"

I have a friend that has a pair of Autograph Signature cabinets that was made by a professional carpenter. You can have a look at his cabinets at the following link:
The cabinets of Chris. I have also added a picture of his cabinets below as a teaser. The other cabinets ae identical, except that they have the flat-sided exponensial front-horn as per the original plans. He is selling them for US$3000 the pair, and the buyer can choose which veneer to have them finished in (exotic hardwoods will add to the price though). The only problem is that they are in South-Africa, so the buyer will have to pay for shipping from here. They are great however if you are a pipe organ fanatic like Chris. He has a full pipe organ built into his listening room which he plays for fun. It also provides a good reference point to compare with the output of the Tannoys (very little to no difference, even on the 32ft pipes ).

Thanks for all your help and I hope you get yor dream Autographs.

Enjoy,
Deon

[DeonC]

Hi GM

Thanks for the link. I tried to download everything on the site, but I did not have much time to surf it, so I might have missed that ZIP. Thanks for the link- I got it and will commence playing.

Just one quick Q though- I did download an Excel spreadsheet from another website where you can simulate exponential horns in. The thing that bothers me is that according to this spreadsheet they won’t go very low (or I am reading it wrong). The mouth cut-off frequency (Fc) for the mouth on my modified GRFs (which is 42.7cm X 55cm, or 16.8” X 21.7”) is only about 99Hz – 100Hz (for wall loading). Corner loading lets it go down to about 60Hz. Can this be correct? If so then how will it produce low bass? I was thinking that it should be able to go down to about 30Hz at least. Where am I going wrong?

Thanks,
Deon

[DeonC]

Hi Mauricio

I forgot ther picture to my previous post. Oops, sorry. Here it is though.

Enjoy,
Deon

[maxlorenz]

Thanx for the pics. I knew these wonders already
Mine will be fed with Hypex' UCD180' in active biamp way
I just have to get a pair of DC speakers

I wonder if the same cannot be accomplished by simply lengtening the baffle behind the driver downwards a bit. That will make the gap feeding into the horn a bit.

I doubt it. I think it has to be with throat diameter. But you can check it with your simulation. This way, even if it sounds bad, at least you can show the graphics to your friends

I have a friend that has a pair of Autograph Signature cabinets that was made by a professional carpenter

As this is DIYaudio I want to make them on my own (with carpenters help, of course)
I envy your friend

Just one quick Q though- I did download an Excel spreadsheet from another website where you can simulate exponential horns in. The thing that bothers me is that according to this spreadsheet they won’t go very low (or I am reading it wrong). The mouth cut-off frequency (Fc) for the mouth on my modified GRFs (which is 42.7cm X 55cm, or 16.8” X 21.7”) is only about 99Hz – 100Hz (for wall loading). Corner loading lets it go down to about 60Hz. Can this be correct? If so then how will it produce low bass? I was thinking that it should be able to go down to about 30Hz at least. Where am I going wrong?

Take a look at the link I gave you: surelly the mesured LF response is BIG.
http://www.hilberink.nl/codehans/tannoy31.htm

I don't know if it is controled/articulated, though. Be aware that all my predictions are based only in non-linear logics, i.e. hunch


Best wishes from,
Mauricio