New Transmission Line BS speakers using Dayton Drivers

Hari Iyer

Well-Known Member
Joined
Feb 8, 2010
Messages
3,822
Points
113
Location
Mumbai
Started a new TL BS project last week using Dayton Reference woofers and tweeters.

Only one raw TL box is ready as of now for measurements and design of crossovers.

Some images -

14v1clv.jpg


33yp7rd.jpg


Measured free air impedance and phase of the woofer

zlxkpv.jpg


There is a variation of around 14% in resonance from the published values. Also other TS parameters are different by around 8% to 10%. So higher cost of drivers does not mean close tolerances - unless i am doing drastically something wrong in measuring the TS parameters using REW which i do not think so.


Measured unstuffed impedance and phase of the woofer

331ghhv.jpg


The tuning of the line is not ideal - this is due to the reduced cross-section of the line near the terminus causing the impednace peak to be less than equal. The tuning of the line is around 66Hz


Measured stuffed impedance and phase of the woofer

2diisko.jpg


The impedance @ resonance is around 42% less than the free air impedance which indicates TL loading and removal of all odd harmonics standing waves. The line is tuned at exactly 66 Hz which is the gometric mean of the 2 peaks of the unstuffed line.


Combined impedance and phase of the woofer

2wqwm79.jpg


Simulated cross-over for the TL speaker system
The cross-over frequency is at 1930Hz which is same as the Wave length of the centre to centre distance between the woofer and tweeter - 7 inches. Design goal was to have tight and punchy bass.

m8nsxd.png


Tweeter reverse connected to check filter phase

2mpkhp4.png


The deep notch at the cross-over point with perfect phasing with tweeter reverse connected.


Will take another 2 weeks to build the crossover as per design and finish the box. Will keep updating as per progress. Thanks.
 
Last edited:
wow... i see better drivers now as compared to the past.

The front looks PMC inspired.. and the speakers look tall as well.

:)

good luck !
 
wow... i see better drivers now as compared to the past.

The front looks PMC inspired.. and the speakers look tall as well.

:)

good luck !
@mpw Thanks.

Yeah, the box is 19" tall to accommodate the line length.

PMC probably has classic TL loading and terminus at the rear of the box. This box has a shorter but thicker line and tuned much higher than resonance and is front terminus.
 
Completed the first iteration of the crossover network. The component tolerance is <1%.

Now waiting for the tough part - Carpenters to finish the box and deliver for me to complete the prototype.

After subjective listening and objective measurements will apply tweaks if at all necessary. Am still looking for good quality finishes for the box. diyaudiocart laminates looks tempting and good economical option too. Will checkout this options too.
 
coincidentally I am also building a TL based on Peerless 6" drivers :) which looks similar to what you are building. But in my case my carpenter is damn slow, he is part of HFV and coincidentally has the same handle as mine :).
WP_20151205_010.jpg
 
@Venki.
Good to know about your project too. You dhould have used interlocking joints for a more rigid cabinet. My spkrs crosssection area is much higher than this tuned to around 66hz. May be your Perrless drivers allow this.

BTW what crossover network hv you planned? My crossover design details are in the first post. Eager to know about yours too.
 
@Venki.
Good to know about your project too. You dhould have used interlocking joints for a more rigid cabinet. My spkrs crosssection area is much higher than this tuned to around 66hz. May be your Perrless drivers allow this.

BTW what crossover network hv you planned? My crossover design details are in the first post. Eager to know about yours too.

Yes sire I should have used interlocking joints, I get very little time to plan so when I do start cutting without thinking too much :(. If I remember correctly, this is tuned to around 50Hz (need to check the design which is in my PC). I am trying to do something different this time by using AMT tweeters. AMT tweeters doesnt have any T/S parameter (this cant be measured atleast). So I am going to do a LPF at around 5khz after correcting impedance. Then would adjust the probably xover and attenuation after measuring the system.
 
My prototype is now ready and is singing. The designed / simulate values seems to agree with the subjective listening test and am quite happy with the outcome. IMO will not require any tweaks atleast for now.

Will carry out objective measurements in the evening and post result to analyze and correlate with my subjective findings. Post that will complete the asthetics and finishing work. Any FM who are keen for a listen of this unfinished SS are welcomed. Am available post 6PM.
 
103xtgm.jpg

Measurement microphone (AKG) and Pre-amplifier (Shure)
jskiuq.jpg

Frequency Response indicates flat response @ 1meter/1 watt measuring distance. Roll off happens around 66Hz @ 6dB/octave.
243lxqp.jpg

Measured impedance and Phase agrees with the simulated value as below
6h14c3.png

Simulated impedance and phase as per original design
10hunv6.jpg

Impulse response displays floor reflection around 3 ms which corresponds to around 290Hz.
rjo9p1.jpg

Step Response indicates clean integration of woofer and tweeter with smooth decay after the first woofer peak.
24yo01v.jpg

RT60 indicates no boom bass and storage of energy for my room. Decay at all frequencies are well controlled which indicates the speaker can also be used in smaller rooms too.
2jafa1g.jpg

Waterfall plot indicates even response in entire frequency spectrum with no major peaks or dips.
2ufu35u.jpg

Spectrogram indicates smooth even energy across spectrum with no major room boom.

Spike at 74Hz is due to room boundary effect and needs to be corrected by adding bass traps at appropriate locations. The one near 240Hz indicates additional stuffing requirement for the 3rd harmonics for the line tuning frequency causing response anomolies. Will add stuffing to correct this at the 3rd harmonic wavelength.

Spike near 20KHz is dummy and is due to microphone measurement curve which I get even without connecting any speaker. This is also indicated in the waterfall plot. Need to invest in a good measurement microphone asap.
 
Last edited:
Today I got chance to listen these speakers.
The first thing I noticed was low noise and calmness of these drivers.They are very easy to follow and musical.
Secondly the space between the instruments can be easily realized. Sound-stage is wide and speakers simply disappears.Not to mentioned that vocals were in dead center with 3D sound.You get feel of live music.They are flat sounding still not clinical.
Drums had enough punch and attack.There is no boom at all.Bass is tight and fast.I can say they sound like good studio monitors.To my liking,I feel vocals may have little more body.When discussed the same with Hari, he said it can be achieved easily.Even speakers need some time to open to its best.
Nice work again Hari.
 
I am yet to listen to this one which I want to but could not get time to visit his place.
Hari, will meet soon.

Till then, keep up the good work.
 
Will plan and modify a little as per Spiros suggestion over the weekend. Will let you know once done.
 
Added some more stuffing and modified the crossover to make a tiny amount of midrange clutter a bit smooth and airy. Yet to listen them.
 
Back
Top