Transmission Line Speaker Build using Alpair 7.3 speakers, first build.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Wow, very nice. I'd really be curious about how they sound. I have a pair of 7.3 in the FH cabinet, but my build wasn't so great so I'm thinking of something else for them.

So, did you create a router template for the cutout and do the layers individually?

Hi Zacster, these were made on a CNC router by a gentleman in France called Christian Pop, yes each part was cut and then layered. All i had to do was glue and clamp them.

Ill post the results of the frequency response and testing i got from REW with my mini dsp mic.

But presently they sound wonderful and surprisingly the bass is clean and present!
 
Was this designed for the A7.3 driver or was it borrowed from the commercial one for a Fostex I have seen somewhere on a Japanese website?


Hi xrk971

These were designed around using the Alpair 7.3, Parameters were Qts 0.4-0.6 and Fs between 70 - 80hz.

I've always wanted to try this type of design as i have built a few normal bass reflex boxes. I also wanted them to be efficient as i am using a valve amplifier with only 3.5w of power.

Not only that with the use of CNC routers makes this type of build possible and easier to construct. I was very impressed by the build, obviously it needed some work but that's the fun of it.
 

Attachments

  • speakertl.jpg
    speakertl.jpg
    640.1 KB · Views: 171
  • speakertl2.jpg
    speakertl2.jpg
    826.3 KB · Views: 167
Last edited:
REW speaker testing

In order for me to really know how these speakers sound i wanted to do some loudspeaker measurements.

I didn't want to spend a great deal of money so with some research i invested in the mini dsp UMIK-1 and the brilliant free software REW (Room EQ Wizard)

https://www.minidsp.com/applications/acoustic-measurements/loudspeaker-measurements

An externally hosted image should be here but it was not working when we last tested it.


I used a DAC Modi from Schiit Audio connected to a class D amplifier.

What i wanted to see was how different types of wadding and also placement of wadding had an effect on the characteristics of the loudspeaker.

I did this until i got the sound i liked and the response i though was better (based on the SPL and phase).



The first measurement is just the loudspeaker in the box with no wadding.





This next screenshot is with wadding compared to without





This screenshot is a mixture of all the different waddings.






Any advice or tips would be most welcome!
 
It's fun to take measurements right in front of the speaker and see how they work.

But, ultimately, you'll want to place them in your room, put the mic at the listening position and level, and measure from there.

Playing with speaker position, you'll probably see that huge 10k shelf disappear.
 
Perceval - while I think you're right about the measurement set up as shown being problematic, what I'd be far more concerned about is the crazy almost 30dB spike centering at 200Hz - which is almost exactly the wavelength of the 1.7m noted as length of the line.

Something very similar to vishalk's measurement is seen in MA's published graphs - although the to not quit the same extreme, and FWIW it should be noted that the latter extend to 40Kz

I've used this driver in literally dozens of enclosures from 6liter sealed ( great for desk-top, but they deserve better IMO)Woden Maeshowe, Pensil, Frugelhorn, several MarKens, custom FASTs, etc. They are definitely capable of much better low frequency response than that shown in the measurements here.
 
Last edited:
I'd recommend simulating this in Hornresp, just as an exercise. The peaks and dips in the measurement are all there, but slightly different shapes and sizes to those simulated.
In theory, bottom end tuning will come out in the region of 50Hz, but it doesn't look like the cabinet is helping much at all down at the bottom end.

Try measuring the driver close-in (don't let the cone hit the mic!), and then the port exit seperately. You might find it interesting.

It looks like that cabinet has been a lot of work, but I'm afraid the inherent design isn't fantastic. Smooth curves in a horn path are great for passing the midrange through, but in this case, we want the opposite: just the bass. Sharp internal angles help to form an acoustic low-pass filter that stops the midrange getting through the port and interfering above 100Hz or so, which is certainly what's happening here.
If it was mine, I'd probably convert it to a ported box, keeping the L-shaped port behind the driver, and then putting holes in that until you're happy with the tuning frequency. Longer port = lower tuning.

Chris
 
Perceval - while I think you're right about the measurement set up as shown being problematic, what I'd be far more concerned about is the crazy almost 30dB spike centering at 200Hz - which is almost exactly the wavelength of the 1.7m noted as length of the line.

Something very similar to vishalk's measurement is seen in MA's published graphs - although the to not quit the same extreme, and FWIW it should be noted that the latter extend to 40Kz

I've used this driver in literally dozens of enclosures from 6liter sealed ( great for desk-top, but they deserve better IMO)Woden Maeshowe, Pensil, Frugelhorn, several MarKens, custom FASTs, etc. They are definitely capable of much better low frequency response than that shown in the measurements here.

Hi Chris

What do you suggest in terms of testing? I placed the mic 4 inches away from the cone directly in the middle of the speaker, i also placed it 1 meter away and took measurements.

How would you change the way i test, let me know i don't mind trying other ways.

Yes i agree with you about the enclosures i was thinking of trying the Frugels but wanted to try and test something smaller hence this project.

I am wondering if there is anything internally i can do? I was also thinking of trying a different driver like a tang band or fostex.
 
I'd recommend simulating this in Hornresp, just as an exercise. The peaks and dips in the measurement are all there, but slightly different shapes and sizes to those simulated.
In theory, bottom end tuning will come out in the region of 50Hz, but it doesn't look like the cabinet is helping much at all down at the bottom end.

Try measuring the driver close-in (don't let the cone hit the mic!), and then the port exit seperately. You might find it interesting.

It looks like that cabinet has been a lot of work, but I'm afraid the inherent design isn't fantastic. Smooth curves in a horn path are great for passing the midrange through, but in this case, we want the opposite: just the bass. Sharp internal angles help to form an acoustic low-pass filter that stops the midrange getting through the port and interfering above 100Hz or so, which is certainly what's happening here.
If it was mine, I'd probably convert it to a ported box, keeping the L-shaped port behind the driver, and then putting holes in that until you're happy with the tuning frequency. Longer port = lower tuning.

Chris


Thank you Chris for your suggestions, you know what it was a starting project just to get me back into audio, had an itch that needed scratching and this was a good start :)

Dont really regret the time or work involved, it was kind of therapeutic that was until i saw the measurements and had feedback from you guys!

I will try what you said, if you were me and had those enclosures how and what you modify to help?

I like tinkering so going to try a different driver, i admit i like the Alpair 7s and they deserve to sing at there best. Really wanted this enclosure to work :confused:

Anyway just keep testing and going at least i'm learning something. Thanks for the tip on Hornresp, is it a difficult software to learn to use?
 
Last edited:
It's fun to take measurements right in front of the speaker and see how they work.

But, ultimately, you'll want to place them in your room, put the mic at the listening position and level, and measure from there.

Playing with speaker position, you'll probably see that huge 10k shelf disappear.


Its great fun ;) i have done the above measurements but my thoughts were the mic needs to be a certain distance from the speaker so you get a true measurement (or a good representation of what the speaker is doing).

Surely you have to do measurements from the speaker, the port, 1m away and then a room measurement?
 
Surely you have to do measurements from the speaker, the port, 1m away and then a room measurement?

Not if you want them tailored to your room.

Having a great looking response at 10cm in front of the driver is great to publish but it does not represent actual real world factors.

Sitting and listening at 1 or 3 meters away, the room will come into play, and all those nice measurements will be for nothing, and you will have to start all over again.

You liked the sound initially. So go ahead. Place the speakers into their respective space, put the mic at head level at your listening position, and measure there.

You might find out why you liked the sound then. It's ok, we all hear things differently, and enjoy our music in different ways. Just like ice cream !
 
Thank you Chris for your suggestions, you know what it was a starting project just to get me back into audio, had an itch that needed scratching and this was a good start :)

Dont really regret the time or work involved, it was kind of therapeutic that was until i saw the measurements and had feedback from you guys!

I will try what you said, if you were me and had those enclosures how and what you modify to help?

I like tinkering so going to try a different driver, i admit i like the Alpair 7s and they deserve to sing at there best. Really wanted this enclosure to work :confused:

Anyway just keep testing and going at least i'm learning something. Thanks for the tip on Hornresp, is it a difficult software to learn to use?

Hornresp has a bit of a learning curve, but there's plenty of help around here to get you started. It's a very powerful bit of software, though, so well worth learning.

If it was mine, I'd probably take a hole saw to the panel directly behind the driver, and make it pass as much air as possible. That'll effectively get you a ported box, which should give better performance.

Chris
 
Its great fun ;) i have done the above measurements but my thoughts were the mic needs to be a certain distance from the speaker so you get a true measurement (or a good representation of what the speaker is doing).

Surely you have to do measurements from the speaker, the port, 1m away and then a room measurement?
You need to minimize the variables when you are measuring, otherwise the source cause is masked.

Near field measurements (4cm) from both the driver and the port will tell you what the speaker is doing without involving the room at this point. The far field measurement is a combination of the two outputs (driver + port) and can be done after the cabinet is working.

It appears you might have a few internal reflection points (175Hz, 210Hz) identified by the curve starting a local peak before abrupting diving asymptotically and returning. My ABC has one at 175Hz, and HornResp predicted it. By comparison, a symmetric dip and return would be a function of sharing between port and driver, like a Bass Reflex.

I should add, depending on where and how wide the resonance or reflection freq band is will determine the audible effect. So it may sound OK, but graph ugly. I would still try a few fixes. Does you cabinet have a removable panel?
 
Last edited:
You need to minimize the variables when you are measuring, otherwise the source cause is masked.

Near field measurements (4cm) from both the driver and the port will tell you what the speaker is doing without involving the room at this point. The far field measurement is a combination of the two outputs (driver + port) and can be done after the cabinet is working.

It appears you might have a few internal reflection points (175Hz, 210Hz) identified by the curve starting a local peak before abrupting diving asymptotically and returning. My ABC has one at 175Hz, and HornResp predicted it. By comparison, a symmetric dip and return would be a function of sharing between port and driver, like a Bass Reflex.

I should add, depending on where and how wide the resonance or reflection freq band is will determine the audible effect. So it may sound OK, but graph ugly. I would still try a few fixes. Does you cabinet have a removable panel?

This is why i came on here to get the advice i need, thank you all.

Yes DonVK it does have a removable panel, i do want to fix this. As you can see i have tried various types of fill in the enclosure and it did help a little.

I was considering what Chris661 suggested which was to cut a hole behind the driver with a hole saw and see the effect it has.

For the moment ill take measurement's from the speaker and port and see what is produces.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.