A new Linear TL Floor Standers for my next upgrade to 845 tubes

[kapvin]

And here is a quick & dirty MLTL using the same driver and enclosure



and the impedance



-6dB around 35hz.

this is exactly the same cabinet, open end closed off, with a 3inch ID port of 7" length in the closed off end.

stuffing is important. -- top 70% of the cabinet to be loosely filled with 200g of polyfill; bottom of cabinet kept free.

bottom of port (about 1") loosely stuffed with 25-40g of polyfill

remember- this is barely optimised.. can model and optimise it further over the weekend if you are interested.

(BTW- took all your dimensions as inner dimensions.)

best wishes

Kapil

 

[keith_correa]

Looks like F3 of ~35hz to me. And these sims look way better than the previous. Nice work!!

Also, in the worksheets, where do you specify that stuffing is below the port? Do you add another section to do this? What effect does that have? Can you post a sim of a "before" and "after" to show the effect of stuffing below the port? Would be interesting to see.

 

[Hari Iyer]

Thanks for the offer. We can plan this Saturday if time permits for you. BTW, are you back at Thane? I can ask the carpenter to make a quick jig where the port can be placed on a plywood plank that can be inserted from below the terminus with the port on it. Also can make the stuffing as per your plan. Will let you know if i can complete this before the weekend.

BTW, i have modified the crossover completely after you last heard the speakers and now the speakers sound quite different from earlier. Check it out.

cheers,

 

[Hari Iyer]

Hi Hari,

this is the modelled TL response based on your Geometry.



and this is the modelled impedance



the dotted lines are infinite baffle and solid line is your enclosure.

impedance looks familiar?

Thanks for the images. I once again checked my data today and following are the data for your re-simulation - All are internal dimensions

Length of the line - 43"
Cross-section area of the line - 35.35 sq.in
Line tuning frequency - 71 Hz
Stuffing density - 1.7 pounds/cubic feet
stuffing material - Dacron polyfill
Characteristic Impedance - 0.275
Total stuffing used - 720 grams - I reduced this to around 630 grams

Calculated pipe resonance freq with damping - 69Hz
Calculated pipe resonance freq without damping - 76Hz

 

[kapvin]

Thanks for the offer. We can plan this Saturday if time permits for you. BTW, are you back at Thane? I can ask the carpenter to make a quick jig where the port can be placed on a plywood plank that can be inserted from below the terminus with the port on it. Also can make the stuffing as per your plan. Will let you know if i can complete this before the weekend.

BTW, i have modified the crossover completely after you last heard the speakers and now the speakers sound quite different from earlier. Check it out.

cheers,

No, I am not in thane in the near future.

the jig seems appropriate.

the dimensions change will impact tuning. will do so when I have a bit of free time.

glad that you have modified the crossover. the previous one was too low, for a first order. i am sure it is sounding excellent.

best wishes.

 

[kapvin]

Also, in the worksheets, where do you specify that stuffing is below the port? Do you add another section to do this? What effect does that have? Can you post a sim of a "before" and "after" to show the effect of stuffing below the port? Would be interesting to see.

Sure.. when I next get time.

maybe i can email you the graphs.. imageshack has become very restrictive. are there any other photo hosting sites?

Kapil

 

[captrajesh]

imageshack has become very restrictive. are there any other photo hosting sites?

Image Shack has become useless. I've moved on to Photobucket. As of now, it is quite good.

 

[Hari Iyer]

I am using tinypic and have not faced any issues till date.

 

[Hari Iyer]

No, I am not in thane in the near future.

the jig seems appropriate.

the dimensions change will impact tuning. will do so when I have a bit of free time.

glad that you have modified the crossover. the previous one was too low, for a first order. i am sure it is sounding excellent.

best wishes.

They are sounding good, but will like the bass to be less compressed. Moving the speakers to the rare wall increases bass by almost 3-4 dB, but i do not like boom bass and hence the speakers are moved away from the rare wall. Will check out your suggestion too.

 

[Hari Iyer]

Made some new observations in my TL speakers that i would like to share with FM w.r.t. measurements,

The single peak is the free air resonance of the woofer which is around 57.6 Hz.
The line impedance measurements indicate a 1st peak at 42Hz and the 2nd peak at 89 Hz. Notice the unbalance in the peak amplitude in the 1st and 2nd peak which indicates issues in pipe tuning of an unstuffed line. This may be an indication of why the bass is compressed.

On further observation the geometric mean of these two peak frequencies is at 62Hz which is very far away from the free air resonance of the woofer indicating a misalignment in the tuning frequency.

The 3rd peak is at 206Hz.

Considering the line tuning frequency of 68Hz the 3rd peak should be at 204Hz which is correct.

I thought of how to balance the amplitude of the 1st and 2nd peak and bring balance in the tuning. After considering the wisdom of FM Kapvin thought of reducing the cross-section area and do an aperiodic loading to the line. The result is as below,

Now both the 1st and 2nd peak is balanced are at 38Hz and 80Hz with the geometric mean at 56Hz which is very close to the driver free air resonance.
Converting the TL loading to an aperiodic loading brought back the pipe tuning frequency to be balanced.

I will be implementing this tomorrow and check the final impedance with the stuffing added. The key is to have the impedance peak near the 2nd peak ie. at 80Hz with 40% to 50% less peak impedance at pipe resonance.

 

[Hari Iyer]

After simulations and measurement did not go with the Aperiodic loading and settled with TL loading. The stuffing of the line allowed the single impedance peak from rising and was in the region of 67Hz which was only 1Hz difference from the calculated value. Also the impedance at tuning freq was around 19Ohms which was around 50% lower than driver impedance at resonance. Also now moved the speakers closer to the wall to increase the low end support with excellent results.

 

[Hari Iyer]

Re-simulated the cross-over for filter phase corrections and measured the actual impedance of the speaker in the box and re-tweaked the cross-over once again for proper filter phase. The phase corrections made signification improvement in the low frequency imaging and allowed accurate imaging for the lows with correct balance of the overall sound. I can now say that unless the phase issues with the filters are addressed an overall coherent sound can never be obtained. Also now moved the speaker back to the original position for enhanced imaging with success.

 

[kapvin]

Hi Hari,

When did you get back to town. I thought you were in Jakarta.

Will be in for a brief period in mid dec. hope to meet then if you are there.

 

[Hari Iyer]

Hi Hari,

When did you get back to town. I thought you were in Jakarta.

Will be in for a brief period in mid dec. hope to meet then if you are there.

Hi Kapvin,

Came last week after completion of phase 1 of project. Will not go for phase 2. Am awaiting allocation for other projects. Should be in Mumbai for more time. December is possible to meet. Also give my refurbished TL a listen. Its not the same when you heard last.

Cheers,

 

[Nikhil]

Hi Hari,

Nice to see the updates on the speaker. If it's not too much trouble, can you explain what is filter phase correction?
Is this in context of phase alignment of the drivers?

Regards.

 

[Hari Iyer]

Hi Hari,

Nice to see the updates on the speaker. If it's not too much trouble, can you explain what is filter phase correction?
Is this in context of phase alignment of the drivers?

Regards.

@ Nikhil,

Earlier i calculated the cross-over filter values with the woofer free-air measured impedance. The filter phase is the phase at the cross-over point. But the cross-over does not see the free-air impedance of the woofer. Instead it sees the impedance of the woofer loaded in the TL box with the stuffing which is very different from the free-air impedance. I measured this impedance and re-calibrated the cross-over for bringing the filter phase at the cross-over point. This change caused the low frequency and mid-range to be precise, with perfect imaging and tight bass.

This is more of the cross-over filter phase correction (electrical). The acoustic phase correction is already done with the woofer and tweeter off-set.

 

[Hari Iyer]

Complete all upgrades to the speaker and restored the original Peerless Kevlar woofer which was designed for the TL box. It took me close to 15 months to conceptualize and complete this project. Now that the speaker system is complete in all technical aspect will now take the opportunity to also mention all other material sources for this build as per below,

Woofer Peerless Fabrikkerne India Ltd, Mumbai
Tweeter - Vifa (sourced from forum member Kapvin)
Capacitors - DEC, Desai Electronics Pvt. Ltd., Mumbai
Resistors - Raatronics, Mumbai
Inductors - Wires used for winding are from RR cables, Mumbai
Damping Material - Recron polyester fiber cotton, Reliance Industries Limited, Mumbai
PCB - Designed and fabricated locally at Thane
Solder - RCS, Alpha Alent, Chennai
Gold plated Terminals - MX - MDR TECHNOLOGIES LIMITED, Thane
Wood - MDF, Sourced locally from Mumbai
Veneer - Century Plywood, Mumbai
spikes - Soundfoundations, Mumbai
hexnut screws - theaudiophile.net, Gurgaon
Labour - Box is built at Thane with high quality MDF
Speaker System - Conceptualized, Designed and Assembled at Thane

some images below of the final build. Only re-polishing is still remaining which i may delay for some more time.

 

[jenson]

looks excellent!

 

[Hari Iyer]

Some technical details,

Image of the measured impedance curve of the speaker is as below,

Image of the simulated impedance, Frequency Response, Filter Phase and Impedance phase is as below,

Measured impedance and simulated impedance is within 5% of the value. I have not done any FR measurement as it will require a more accurate measurement microphone, stands and a good quality sound card to do it. But impedance measurement only require an impedance jig which is easy to build.

As frequency response is a function of impedance i can safely infer that the measured FR will be almost same as the simulated FR. If you look at the impedance phase of the simulated response you can see that the speaker system provides close to resistive load to the amplifier from 700Hz to around 10KHz. The filter cross-over is around 1830Hz and the FR shows a drop of around 1.5dB from around 1.5KHz till 10KHz after which there is a small roll-off till 20KHz to keep the sound mellow and fatigue free.

 

[shafic]

Nice work!!:clapping: