Hi,
I just finished my first TL box and started making some impedance and frequency response measurements. I used some synthetic pillow stuffing, so I'm not sure if this is suitable? I stuffed about 80% of the line, only light stuffing and leaving the last 20% of the line unstuffed.
First some pictures of the cabinet:
Impedance response unstuffed cabinet:
Next measurement is of cabinet with light stuffing:
If I'm correct I need to use more stuffing as I see still 2 peaks?
I did some frequency response measurements as well, with the stuffed cabinet.
30 cm from side wall of cabinet:
20 cm in port:
15 cm above port:
20 cm in front of cabinet:
Simulation of final design:
I expected a bit more LF range so I'm wondering if I did something wrong? Should I change stugging? (add more stuffing)
I just finished my first TL box and started making some impedance and frequency response measurements. I used some synthetic pillow stuffing, so I'm not sure if this is suitable? I stuffed about 80% of the line, only light stuffing and leaving the last 20% of the line unstuffed.
First some pictures of the cabinet:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Impedance response unstuffed cabinet:
An externally hosted image should be here but it was not working when we last tested it.
Next measurement is of cabinet with light stuffing:
An externally hosted image should be here but it was not working when we last tested it.
If I'm correct I need to use more stuffing as I see still 2 peaks?
I did some frequency response measurements as well, with the stuffed cabinet.
30 cm from side wall of cabinet:
An externally hosted image should be here but it was not working when we last tested it.
20 cm in port:
An externally hosted image should be here but it was not working when we last tested it.
15 cm above port:
An externally hosted image should be here but it was not working when we last tested it.
20 cm in front of cabinet:
An externally hosted image should be here but it was not working when we last tested it.
Simulation of final design:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
I expected a bit more LF range so I'm wondering if I did something wrong? Should I change stugging? (add more stuffing)
Last edited:
Hmm, I haven't tried to do a complex fold in LA, but your measurements Vs sim implies that the latter is assuming the terminus is near/at the floor whereas it's well above; plus, with a Qts < ~0.4 the TL's tuning should ideally be above Fs unless some form of EQ is used to flatten the response over a wider bandwidth [BW]. That said, your port measurement is showing plenty of output, but it's not blending well with the driver's as you move away from the speaker, so remeasure it with it near/at a wall to see if it's more to your liking.
GM
Edit: just noticed the simmed impedance showing the opposite of your measurements and since tuning is < Fs, it's not right.
GM
Edit: just noticed the simmed impedance showing the opposite of your measurements and since tuning is < Fs, it's not right.
Last edited:
Hmm - there should be two impedance peaks, as shown in the simulation. The simulated frequency peaks seem to match pretty well to what you're measuring on the unstuffed line.
However, your tuning is too low for the driver, I think.
Too much stuffing has a tendency to kill bottom end response, but helps with the wiggles further up. It also "kills" the quality, so an overstuffed line sounds dull. By using tapering and driver offset from the end, the goal is to minimize the amount of stuffing you need. Stuff more in the closed end and less towards the opening.
You'll need to first run in the drivers a bit before starting to measure.
Try measuring the response at 1 meter distance. I think you're loosing some of the port contribution when you're measuring too close to the driver.
Also, in general, it seems to be a consensus that the lines work a little better with more stuffing in the closed end, and lighter towards the open end.
BTW: you don't need any stuffing in the small, triangular dead ends. In fact, the deflectors are unnecessary - they don't really have any practical mission. Try simulating with and without them - you'll see very little difference.
Johan-Kr
However, your tuning is too low for the driver, I think.
Too much stuffing has a tendency to kill bottom end response, but helps with the wiggles further up. It also "kills" the quality, so an overstuffed line sounds dull. By using tapering and driver offset from the end, the goal is to minimize the amount of stuffing you need. Stuff more in the closed end and less towards the opening.
You'll need to first run in the drivers a bit before starting to measure.
Try measuring the response at 1 meter distance. I think you're loosing some of the port contribution when you're measuring too close to the driver.
Also, in general, it seems to be a consensus that the lines work a little better with more stuffing in the closed end, and lighter towards the open end.
BTW: you don't need any stuffing in the small, triangular dead ends. In fact, the deflectors are unnecessary - they don't really have any practical mission. Try simulating with and without them - you'll see very little difference.
Johan-Kr
Thanks for your suggestions - this is really appreciated. I will try to use less stuffing and compare.
Yesterday I implemented a software erqualizer as a quick and dirty test, I used the subwoofer together with my full range DIY electrostatic loudspeakers and I crossed over at 80 Hz / 12 dB / oct, and it sounded ok - not (much) worse than my esl I think, but it didn't sound as much more low range extension - maybe it feels just a little bit stronger compared to the low range of my esl.
By the way - I'm considering to try a mass loaded (tapered) TL as well - as according to the simulation software I can use my current cabinet and just close the port and add a small tube. If I'm correct I will need less stuffing compared to my current TQWT.
Hi,
today I did new outdoor measurements. I measured my transmission line woofer together with a 45 liter sealed enclosure that I built for reference, just to check if the TL would sound better or not.
I measured in my garden to minimize reflections, although there is still a fence at 2 meter distance. Most measurements were made with mic 2 meter away from woofer cabinet. I finished my second TL woofer, but left the back panel loose so I'm able to shift it in and out in order to change (decrease) line length and increase tuning frequency. I did this in 10 cm increments, so calculated line length -10, -20 and -30 cm.
Measurement #1: mic 2m distance, line 1 = original line length, line 4 = -10 cm, line 5 = -20cm, line 6 = -30cm.
Comparing 0 cm and -10 cm line length
Mic at 0.5, 1 and 2 meter distance from cabinet
Influence of stuffing: +/- first 70% of line stuffed (blue line) compared to only stuffing in first bend. (red line)
Comparing 45 liter sealed unstuffed test enclosure with stuffed TL (+/- first 40% of line stuffed)
My first impression is that the TL port adds extra sound pressure at lower frequencies (last picture) - so this seems to be working, although I'm still nog 100% sure whether the cabinet is tuned correctly. I did also a listening session where I compared the TL with the sealed cabinet and I got the idea that the TL produces deeper bass. Maybe the sealed woofer sounds a bit tighter, but I'm not sure.
By changing stuffing I can get a few dB's more or less output. Changing port length influences freq. response, but not as dramatic as I expected. I made some near field measurements (indoor) as well, I will publish these measurements too - these graphs suggested that decreasing port length gives less LF output, but I just measured port output and driver output seperately.
today I did new outdoor measurements. I measured my transmission line woofer together with a 45 liter sealed enclosure that I built for reference, just to check if the TL would sound better or not.
I measured in my garden to minimize reflections, although there is still a fence at 2 meter distance. Most measurements were made with mic 2 meter away from woofer cabinet. I finished my second TL woofer, but left the back panel loose so I'm able to shift it in and out in order to change (decrease) line length and increase tuning frequency. I did this in 10 cm increments, so calculated line length -10, -20 and -30 cm.
Measurement #1: mic 2m distance, line 1 = original line length, line 4 = -10 cm, line 5 = -20cm, line 6 = -30cm.
An externally hosted image should be here but it was not working when we last tested it.
Comparing 0 cm and -10 cm line length
An externally hosted image should be here but it was not working when we last tested it.
Mic at 0.5, 1 and 2 meter distance from cabinet
An externally hosted image should be here but it was not working when we last tested it.
Influence of stuffing: +/- first 70% of line stuffed (blue line) compared to only stuffing in first bend. (red line)
An externally hosted image should be here but it was not working when we last tested it.
Comparing 45 liter sealed unstuffed test enclosure with stuffed TL (+/- first 40% of line stuffed)
An externally hosted image should be here but it was not working when we last tested it.
My first impression is that the TL port adds extra sound pressure at lower frequencies (last picture) - so this seems to be working, although I'm still nog 100% sure whether the cabinet is tuned correctly. I did also a listening session where I compared the TL with the sealed cabinet and I got the idea that the TL produces deeper bass. Maybe the sealed woofer sounds a bit tighter, but I'm not sure.
By changing stuffing I can get a few dB's more or less output. Changing port length influences freq. response, but not as dramatic as I expected. I made some near field measurements (indoor) as well, I will publish these measurements too - these graphs suggested that decreasing port length gives less LF output, but I just measured port output and driver output seperately.
Last edited:
The interesting part is where stuffing only the first 70% of the line gives more output in the bottom. Too much stuffing sucks life ot of the line. You need some stuffing to lessen wiggles above 100Hz, but the less you need, the better.
The measured outdoor response seems to be well suited for use in a room. If you can get 12dB/octave rollof below 40Hz or so, it'll match well to an average size room.
I'm wondering a bit about the dip at 72Hz on both sealed and tline. Outside, you shouldn't have any room-related problems.
Johan-Kr
The measured outdoor response seems to be well suited for use in a room. If you can get 12dB/octave rollof below 40Hz or so, it'll match well to an average size room.
I'm wondering a bit about the dip at 72Hz on both sealed and tline. Outside, you shouldn't have any room-related problems.
Johan-Kr
Last edited:
Hi Fenalaar,
thanks for your response and suggestions - I got curious too about the dip at 72 Hz. I did some new measurement today and measured at bigger distances from the cabinet. It seems that the dip is increasing when cabinet - mic distance is increased: (top / purple line is mic at 0.25m from cabinet, bottom yellow, brown line is mic at 3.5 meter from cabinet)
So I'm wondering what could be the cause. I have a narrow garden (about 5 meters wide) and at one side there is a wooden fence (about 2 meters high). The cabinet - fence distance is about 1.5 - 2 meters. Could it be the fence causing the dip? Or could it be an interaction between the port at the back of the cabinet with the driver output at the front of the cabinet?
When measuring driver output at 1 cm in front of the dust cap, (indoor) there is no dip at 72 Hz:
Regards, J.
thanks for your response and suggestions - I got curious too about the dip at 72 Hz. I did some new measurement today and measured at bigger distances from the cabinet. It seems that the dip is increasing when cabinet - mic distance is increased: (top / purple line is mic at 0.25m from cabinet, bottom yellow, brown line is mic at 3.5 meter from cabinet)
An externally hosted image should be here but it was not working when we last tested it.
So I'm wondering what could be the cause. I have a narrow garden (about 5 meters wide) and at one side there is a wooden fence (about 2 meters high). The cabinet - fence distance is about 1.5 - 2 meters. Could it be the fence causing the dip? Or could it be an interaction between the port at the back of the cabinet with the driver output at the front of the cabinet?
When measuring driver output at 1 cm in front of the dust cap, (indoor) there is no dip at 72 Hz:
An externally hosted image should be here but it was not working when we last tested it.
Regards, J.
Last edited:
I forgot that the sealed cabinet showed the same dip, so I guess that's not the cause of the dip at 72 Hz.
Hi, it has been a while since my last visit here but I made a lot of progress: I concluded that the woofer is not adding much to the bigger electrostatic loudspeaker. Is seems to fit better to the smaller electrostatic loudspeaker and I painted the enclosures and the combination of esl + woofer looks and sounds very nice in my opinion. I'm very satisfied with the result so far. I think I overstuffed the woofer as it sounds a bit life-less, as Fenelaar stated, so I will remove some of the wool / stuffing.
By the way - I got rid of all the bumps and the whole system runs from about 20 Hz up to 13 kHz. The only problem is an irritating vibration between the woofer frames and the woofer enclosure. The Tangband drivers have only four holes for attachement which is too few. Do you have any suggestions how to fix this problem? Should I use a gasket or some sort of silicone sealant? I prefer to keep everyting removable.
By the way - I got rid of all the bumps and the whole system runs from about 20 Hz up to 13 kHz. The only problem is an irritating vibration between the woofer frames and the woofer enclosure. The Tangband drivers have only four holes for attachement which is too few. Do you have any suggestions how to fix this problem? Should I use a gasket or some sort of silicone sealant? I prefer to keep everyting removable.
I'm going to guess that the "vibration" you're hearing may actually be air escaping because the seal between the driver and the box is not airtight. In any case, 1/8" weatherstripping is all you need to fix the problem. Just apply it around the hole in the box before screwing the driver in.
Impressive performance
Hi there S: Impressive performance from a reasonable sized enclosure and great driver choice. What are the box dimensions? How does it sound for music reproduction? ...regards, Michael
Hi there S: Impressive performance from a reasonable sized enclosure and great driver choice. What are the box dimensions? How does it sound for music reproduction? ...regards, Michael
Hi Brian,
thanks for your suggestion, I will add the weatherstripping, sounds like an easy job
Last edited:
Hi Michael, thanks for your compliment, the outer dimenstions of the enclosure are: 68x43x35 cm (27x17x14 inch).
How it sounds: in my unprofessional opinion it sounds tight, LF is not overly present / loud, just fine. I don't really have any serious reference, but I listened to a few loudspeaker systems of friends of mine, in a price range of 1000 - 2000 euros, and LF of my woofers is more extended, and I prefer the more tight sound of this TL woofer compared to the BR woofers of my friends systems, although I only compared by using my memory.
Although it seems to me that many people don't prefer to mix an electrostatic driver with a dynamic driver, I think they blend quite well together.
A few pictures of the result so far:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.