Bass Driver for a transmission line speaker

[Paul N]

Greetings from the Netherlands

I would like some help in picking a good driver for my self-built Rogers Monitor, a 34 x 44 x 91 cm (external b x w x h) 4 way speaker. Behind the bass driver is a folded horn (labyrinth). Total net volume of the horn is about 70 liters. Length of the horn is about 210 cm. I have tried to simulate the horn in hornresp, but apparently with wrong or missing input. Quite different drivers give more or less the same result.

The present driver is a 10” Monacor SPH265, with an fs of 23Hz, a qts of 0,27 and a VAS of 199 liter. Xmax is 5.5mm. The Monacor seems to have been a wrong decision by me. On its own, the speaker produces relatively little sound below 100Hz.

At the moment I am considering a 10” or 12” Dayton reference driver, or alternatively a 10” or 12” Scan-Speak 4558T00. These drivers have good ratings, are easily available and price wise acceptable to me. A 12” inch requires a larger opening in the baffle and will decrease the stifness of the speaker.

I do not want to build a new pair of speakers, but want my old companions to sound as good as they can. The speakers are driven by a Pioneer SC-LX79 using bi-amping, with a separate channel for the bass driver.

Looking forward to your suggestions, kind regards, Paul

 

[Kvn_Walker]

There is complex math that’s required to match a transmission line enclosure and woofers.

TLSGuy has built custom TL speakers, so he‘s probably best qualified to help you out.

 

[Pogre]

There is complex math that’s required to match a transmission line enclosure and woofers.

TLSGuy has built custom TL speakers, so he‘s probably best qualified to help you out.

@TLS Guy.

This seems a little backward? I thought it more ideal to choose the drivers first, then figure out the cabs...

 

[TLS Guy]

@TLS Guy.

This seems a little backward? I thought it more ideal to choose the drivers first, then figure out the cabs...

Yes, unfortunately he is backward. Not only that, I feel he does not know what a transmission line is. A transmission line is NOT a horn, it is a modified Gedeckt organ pipe. The physics of horns and pipes is totally different.

The driver he chose would be OK for horn loading, but useless for a TL. As with ported enclosures you have to model the pipe to the driver and NOT the other way round.
It is virtually impossible to match a driver to an existing TL, just like you are generally out of luck trying to replace a driver to an existing ported enclosure.

The OP can down load George Auspurger's TL modelling program from my website. He was the first to publish the first accurate mathematical TL model just over 20 years ago. I can assure, and attest to the fact, that his is the only accurate model. There are others, but they are in error.

George is now well into his nineties, but has kindly consented to allow me to make his model available. I feel highly privileged that he has allowed me to have access to, and make the model available.

The OP can download the model here from my website.

Unfortunately there are very, very few of us alive who really know how to properly design TLs, which is a pity, as they produce the most perfect, accurate and realistic bass of any form of loudspeaker loading.

I do not have enough information to begin to help the OP, as the cross sections, taper, length of the pipe, port area, volume of the pipe, type and weight of the damping material are crucial.
There is a lot that goes into the design of a TL and they are highly intolerant of errors.

 

[ryanosaur]

Greetings from the Netherlands

I would like some help in picking a good driver for my self-built Rogers Monitor, a 34 x 44 x 91 cm (external b x w x h) 4 way speaker. Behind the bass driver is a folded horn (labyrinth). Total net volume of the horn is about 70 liters. Length of the horn is about 210 cm. I have tried to simulate the horn in hornresp, but apparently with wrong or missing input. Quite different drivers give more or less the same result.

The present driver is a 10” Monacor SPH265, with an fs of 23Hz, a qts of 0,27 and a VAS of 199 liter. Xmax is 5.5mm. The Monacor seems to have been a wrong decision by me. On its own, the speaker produces relatively little sound below 100Hz.

At the moment I am considering a 10” or 12” Dayton reference driver, or alternatively a 10” or 12” Scan-Speak 4558T00. These drivers have good ratings, are easily available and price wise acceptable to me. A 12” inch requires a larger opening in the baffle and will decrease the stifness of the speaker.

I do not want to build a new pair of speakers, but want my old companions to sound as good as they can. The speakers are driven by a Pioneer SC-LX79 using bi-amping, with a separate channel for the bass driver.

Looking forward to your suggestions, kind regards, Paul

If you built this, you should be able to post the plan so we can see what you are talking about. What you describe using three different terms: Horn, TL, Labrytnth... is confusing.
Something isn't adding up and it is making it difficult for even the most likely to assist you... to assist you!
Please clarify!

 

[Paul N]

Thank you all for your quick reply. TLS Guy, I am aware that my backwards way of working is far from ideal. I have read your post on the TL principle and was aware of the works George Auspurger. Thanks for the link to his model. I have just downloaded it. Is it correct that the speaker database is empty?

Herewith more detailed information on the Rogers Monitor that I built around 1980. First the drawing:

 

[Paul N]

Information on the design (translated from Dutch)
"The system contains a relatively large bass chamber, which is coupled to a duct with a length of 1/4 of the resonance frequency fs of the bass driver. The duct diameter gradually decreases amd ends in an opening. To keep the box diameter acceptable for living room purposes, the duct is folded a couple of times, creating a labyrinth.
One of the special properties of a well designed labyrinth is the reflection of the air that reaches the backside of the cone exactly in opposing phase at fs, which results in damping of the driver resonance at fs. Above and below fs, the bas driver is not hindered by the stifness of the enclosed air as in a closed box, resulting in a increased fs.
The output below fs is amplified, so that this system is able to even produce an audible 20Hz. This is contrary to a closed box, which reduces the output below fs with 12db per octave
Because the labyrinth has less parallel walls, standing waves and unwelcome harmonics hardly happen.
The duct has to be filled with an exact quantity of absorbing material. According to research by dr Bailey the only right material for that purpose is clean long haired sheepwool. Every other material results in lesser performance.
The appliace of Dr Bailey's longhair reduces the speed of sound to less than 300 m/s, which reduces the duct length and so the cabinetsize to a minimum. Also the unwelcome harmonics are effectively absorbed.
A right dose of Dr Bailey's long hair results in the targeted Q-factor. In case more wool is applied, the bass tones will be less strong and the Q-factor will be lower. Less wool results in stronger bass tones and a strong reduction in the amount of wool applied will result in an unwanted peak in the rendition of the bass notes. The amount of wool in the bass chamber has the most effect. In the buiding instructions, the location and amount of wool to be applied is indicated."

From the time that I built it, I remember that I was told that the wool in the duct was applied at the location of the antinodes at fs.

The original speaker which was used was a Keff B139. Below its datasheet.

 

[Paul N]

The datasheet for the current speaker, the Monacor SPH265

 

[Paul N]

Meanwhile the B139 has been replaced. Version 5.0 is said to have a special version of the Davis 25SCA10W, the 25 SCA 10 WW. The WW version is mentioned on the Davis website, so it has been produced. The below presented specs are probably not completely right, but might be of interest.
25 SCA 10 WW
outside diameter: 262 mm
mounting cutout: 245mm
Power rating: 80W
SPL: 90 (dB/W ??)
Fs: 25Hz
Qts: 0.51
Vas: 181 liter
Mms: 45.4 gram
Cms: 0.40 mm/N
Rms: 3.9 kg/s
Sd: 330 cm^2
Bl: 10.3 N/A
Re: 6.0 ohm
Le: 0.60 mH
Weight: 3.0 kg

The 10W specs are quite different:
25 cm bass driver with carbon cone.

• resonance frequency fs = 28 Hz
• impedance R = 6 Ohm
• sound pressure level SPL = 92 dB (2,83V; 1m)
• DC resistance Re = 5,96 Ohm
• force factor BL = 10,08 N/A
• voice coil inductance L = 0,53 mH
• effective mechanical mass incl. air load mms = 30,9 g
• equivalent volume of compliance Vas = 109 l
• total Q factor Qts = 0,33 (Qms=2,23, Qes=0,39)
• maximum peak linear excursion vibration xlin = +/- 8 mm
• overall diameter d = 280 mm

If there is any other info that you might need to have to help me further, please let me know.

Thanks again, Paul.

 

[TLS Guy]

Meanwhile the B139 has been replaced. Version 5.0 is said to have a special version of the Davis 25SCA10W, the 25 SCA 10 WW. The WW version is mentioned on the Davis website, so it has been produced. The below presented specs are probably not completely right, but might be of interest.
25 SCA 10 WW
outside diameter: 262 mm
mounting cutout: 245mm
Power rating: 80W
SPL: 90 (dB/W ??)
Fs: 25Hz
Qts: 0.51
Vas: 181 liter
Mms: 45.4 gram
Cms: 0.40 mm/N
Rms: 3.9 kg/s
Sd: 330 cm^2
Bl: 10.3 N/A
Re: 6.0 ohm
Le: 0.60 mH
Weight: 3.0 kg

The 10W specs are quite different:
25 cm bass driver with carbon cone.

• resonance frequency fs = 28 Hz
• impedance R = 6 Ohm
• sound pressure level SPL = 92 dB (2,83V; 1m)
• DC resistance Re = 5,96 Ohm
• force factor BL = 10,08 N/A
• voice coil inductance L = 0,53 mH
• effective mechanical mass incl. air load mms = 30,9 g
• equivalent volume of compliance Vas = 109 l
• total Q factor Qts = 0,33 (Qms=2,23, Qes=0,39)
• maximum peak linear excursion vibration xlin = +/- 8 mm
• overall diameter d = 280 mm

If there is any other info that you might need to have to help me further, please let me know.

Thanks again, Paul.

There are plenty of KEF B139s on eBay. Many were made, and are still in very good working condition. I have six in regular use at the moment, and some spares.

The KEF B 139 has the most perfect T/L parameters for TL loading and was the basis for many. Jim Rogers was part of the Radford group, that did the really ground breaking work on TLs. Raymond Cooke had TLs in mind when he designed the B139. There really is no adequate replacement for it. In any event using a different driver, apart from miss alignment, would require a complete crossover redesign.

There are two for sale in eBay right now.

The KEF B 139 is a really special driver and certainly in the top five of all time. So you need to pony up and snag those drivers. Anything else you could do, would be way inferior than replacing with the genuine article.

Two B139s in their old location.

These speakers in our new home. They are the rear backs, in my old and new AV room. They used to be my reference monitors in my former studio I did a lot of mastering in and edited radio broadcasts.

This is those dual TLs in their current location.

Here is a B139 in a three way B4 design.

This is the front of our new AV room, with full range triamped dual TL mains, and biamped TL center.

In wall TL for my wife's great room system.

The take home message is that you need to use the B 139 drivers, those excellent lines were designed for, and not try and substitute. The only alternative is a totally new design and build.

 

[Paul N]

There are plenty of KEF B139s on eBay. Many were made, and are still in very good working condition. I have six in regular use at the moment, and some spares.

The KEF B 139 has the most perfect T/L parameters for TL loading and was the basis for many. Jim Rogers was part of the Radford group, that did the really ground breaking work on TLs. Raymond Cooke had TLs in mind when he designed the B139. There really is no adequate replacement for it. In any event using a different driver, apart from miss alignment, would require a complete crossover redesign.

There are two for sale in eBay right now.

The KEF B 139 is a really special driver and certainly in the top five of all time. So you need to pony up and snag those drivers. Anything else you could do, would be way inferior than replacing with the genuine article.

Two B139s in their old location.

These speakers in our new home. They are the rear backs, in my old and new AV room. They used to be my reference monitors in my former studio I did a lot of mastering in and edited radio broadcasts.

This is those dual TLs in their current location.

Here is a B139 in a three way B4 design.

This is the front of our new AV room, with full range triamped dual TL mains, and biamped TL center.

In wall TL for my wife's great room system.

The take home message is that you need to use the B 139 drivers, those excellent lines were designed for, and not try and substitute. The only alternative is a totally new design and build.

I am very impressed by your knowledge and collection and will follow your advice by trying to buy a set of Rogers Monitor built with the B139. I have just sent a message to someone in the Netherlands offering a set. Only problem I have with them is that they are in black oak instead of birch, which I have used.

 

[Paul N]

I am very impressed by your knowledge and collection and will follow your advice by trying to buy a set of Rogers Monitor built with the B139. I have just sent a message to someone in the Netherlands offering a set. Only problem I have with them is that they are in black oak instead of birch, which I have used.

And if that does not work out, I have at the moment two other options to obtain them

 

[Paul N]

I have received a pair of B139´s in good order. After our holidays, I will make a new baffle for them. As it turns out, none of the drivers that I used were in the original Rogers Monitor design. So I will have to replace the crossover filter, also because the B139 will be amped separately.

 

[Paul N]

As far as I know, this is version (no 1) of the housing of the Rogers Monitor that I built. I managed to find a drawing of the inside with dimensions in mm., which I edited. The thickness of the panels is 18 mm.

The capital letters are added by me to calculate the volumes of each part of the Transmission Line.

The TL is stuffed with Dr. Bailey's longhair wool, which is kept in place by nets. The location of the nets is indicated by dashed lines. Please note that the Capital letters refer to different area's then in the drawing above.

 

[TLS Guy]

As far as I know, this is version (no 1) of the housing of the Rogers Monitor that I built. I managed to find a drawing of the inside with dimensions in mm., which I edited. The thickness of the panels is 18 mm.

The capital letters are added by me to calculate the volumes of each part of the Transmission Line.

View attachment 49205
The TL is stuffed with Dr. Bailey's longhair wool, which is kept in place by nets. The location of the nets is indicated by dashed lines. Please note that the Capital letters refer to different area's then in the drawing above.

View attachment 49206

This is the model for the KEF B 138 TL.

This is the pipe data.

You did not give me a width of your cabinet, that I can see. But I have shown you the optimal length. Throat is the area at the driver. I modelled it with wool damping as per your instructions.

If you are using different drivers for mid and tweeter, then the crossover will need a total redesign.

The KEF B 139 starts to resonate around 750 Hz, and so it need a fourth order low pass filter no higher than 400 Hz. You can see F3 is around 30 Hz. That is the biggest problem working with the KEF B139. My lines are crossed at 180 Hz third order, and the ported enclosure, 400 Hz fourth order. That keeps them out of trouble.

 

[Paul N]

This is the model for the KEF B 138 TL.

This is the pipe data.

You did not give me a width of your cabinet, that I can see. But I have shown you the optimal length. Throat is the area at the driver. I modelled it with wool damping as per your instructions.

If you are using different drivers for mid and tweeter, then the crossover will need a total redesign.

The KEF B 139 starts to resonate around 750 Hz, and so it need a fourth order low pass filter no higher than 400 Hz. You can see F3 is around 30 Hz. That is the biggest problem working with the KEF B139. My lines are crossed at 180 Hz third order, and the ported enclosure, 400 Hz fourth order. That keeps them out of trouble.

Thank you for your help with my Rogers Monitor. Much appreciated.
The width of the inside of the housing is 300 mm.
With my lines you mean your TL speakers?
I had expected that the throath would be the beginning of the duct. Anyway, I will try inputting the data in George’s his model in the same way as you have.
You are right that the filter for the other 2 drivers might very well need to be changed. I plan to measure these drivers in a couple of weeks And compare the results with the measures for the original speakers.

 

[TLS Guy]

Thank you for your help with my Rogers Monitor. Much appreciated.
The width of the inside of the housing is 300 mm.
With my lines you mean your TL speakers?
I had expected that the throath would be the beginning of the duct. Anyway, I will try inputting the data in George’s his model in the same way as you have.
You are right that the filter for the other 2 drivers might very well need to be changed. I plan to measure these drivers in a couple of weeks And compare the results with the measures for the original speakers.

My rear back TLs are crossed at 180 Hz from the B 139s active. My ported reflex enclosures in the family room are crossed 400 Hz passive.

I see your speakers have the driver right at the beginning of the line. This is not correct design. The driver should actually be placed near the antinode of the third harmonic to prevent excitation of the odd harmonics, of which the third will be the most prominent. So that is between one quarter and third from the closed top of the line.
So you will have increased odd harmonic distortion from that part of the design. I usually place my drivers at the quarter mark from the closed end, and so three quarters from the open end.

 

[Paul N]

It seems like my first try at simulating my RM with the B139 (description above graph immediately below is wrong) SPH-265 comes very close to what you simulated for the B139, although the input clearly differs

I will have another go at it with the info I got from you, the speaker data for the B139B and my more precise figures for the box of my RM.

 

[TLS Guy]

It seems like my first try at simulating my RM with the B139 comes very close to what you simulated, although the input clearly differs
View attachment 49256

View attachment 49257

I will have another go at it with the info I got from you and my more precise figures for the box of my RM.

Your taper should be entered as linear, as the restriction is constant per foot. A 3/1 taper is most common, but I think yours is about 4/1

 

[Paul N]

Below my second simulation, now with the B139B.
Piston area is offered by George's TL simulator and as I do not have data for that, I left it unchanged.
I used 7.2 Ohm for calculating BL2/R and mH/R. For one reason or another Rmin/R does not accept a value below 1.

The enclosure data that I used in my second simulation:

For total pipe length, I used the length of the acoustical path. If the coupling chamber should be excluded, pipe length is 1,72 m. For stub length, I used the minim distance between the port and the surround of the B139B.

Simulation 2 results:

I did a couple of simulations in which I only changed the damping. No damping gave an ugly result but going down to 4mm (the minimum for wool) did not make much of a difference. Also just changing the pipe length did not make a lot of difference.