Ive been fooling around with Martin Kings ML-TQWT program and have come to an impass. The below output is as optimal as I have gotten thus far. I would like to flatten the response out a bit more but cant seem to get the parameters dialed in to do so. Does this look about as good as can expected or can it still be manipulated further?
As if so, what combination will do this?
Thanks, amt
As if so, what combination will do this?
Thanks, amt
Attachments
Great! I conjured up the experts.
Since I cant save the work, Ive had to write down the parameters and so here they are. I havent added any series resistance and to be honest, hadnt thought of doing so.
Fs 45
Re 5.66
Le .67
Bl 13.8
Sd .0449 m2
Vas 73.51 liters
Qe .24
Qm 7.07
Qt .23
L = 52
driver position = .4
So = 1
SL = 3
Density = .25
rPort = 2
lPort = 5.5
Im more concerned with efficiency than lowering the response.
amt
Since I cant save the work, Ive had to write down the parameters and so here they are. I havent added any series resistance and to be honest, hadnt thought of doing so.
Fs 45
Re 5.66
Le .67
Bl 13.8
Sd .0449 m2
Vas 73.51 liters
Qe .24
Qm 7.07
Qt .23
L = 52
driver position = .4
So = 1
SL = 3
Density = .25
rPort = 2
lPort = 5.5
Im more concerned with efficiency than lowering the response.
amt
I, too, am interested in this driver. I was able to model the attached response. Perhaps a bit of an improvement?
My parameters where:
L = 50
driver position = .45
So = 1
SL = 3
Density = .6
rPort = 4 (don't want any port noise or compression)
lPort = 6
I notice from the published frequency response that there is a dip between around 300 and 700 hz. Perhaps that would even out the response a bit.
I wonder what the response would look like in this enclosure above 1k.
BTW, I've always been uncertain about the driver position. Is it the % from the closed end? As in (50 x 0.45) = 22.5 inches from the closed end?
Thanks.
Paul Ebert
My parameters where:
L = 50
driver position = .45
So = 1
SL = 3
Density = .6
rPort = 4 (don't want any port noise or compression)
lPort = 6
I notice from the published frequency response that there is a dip between around 300 and 700 hz. Perhaps that would even out the response a bit.
I wonder what the response would look like in this enclosure above 1k.
BTW, I've always been uncertain about the driver position. Is it the % from the closed end? As in (50 x 0.45) = 22.5 inches from the closed end?
Thanks.
Paul Ebert
Attachments
), I'm thinking about giving it a go.Probably very close to the driver's measured response o a spec sheet.
You have it correct.
I still think that the comments in my second post hold. Paul's design is better but is still down about 5 dB from the response at 1 kHz. I think that without a correction circuit the bass will be very weak.
That looks much better. But you still need to deal with the baffle step in some way. For my Lowther systems, I end up using an 8 to 10 ohm parallel resistor in the correction circuit to overcome baffle step and the low Qts of the different Lowthers drivers. I use SS amps, a tube amp would require a lower value of resistance. You can use a circuit, try room placement, or some combination of the two to get a flat response and decent bass.
Try this on for size.
The MathCAD worksheet looks like this
Bob
An externally hosted image should be here but it was not working when we last tested it.
The MathCAD worksheet looks like this
Bob
Bob, that looks pretty flat. I notice that the software version I have doesnt include the extra resistance field, and yours doesnt include the driver placement field. Where is the driver placed, or have I overlooked something?
Martin, so the BSC is really manditory for this drive, or else the frequency response will really be down by 3-4 db below 100hz or so?
Paul, Ive been eyeing this driver for a couple years, but once the Druid made its debut, I felt compelled to try and use it. A friend thats bass savy has said that they are a nice sounding speaker and has listened to the them numerous times. I will probably use a Fostex FT17H.
amt
Martin, so the BSC is really manditory for this drive, or else the frequency response will really be down by 3-4 db below 100hz or so?
Paul, Ive been eyeing this driver for a couple years, but once the Druid made its debut, I felt compelled to try and use it. A friend thats bass savy has said that they are a nice sounding speaker and has listened to the them numerous times. I will probably use a Fostex FT17H.
amt
amt said:
I did indeed edit MJK's worksheet to add the Rs fields. Also, you can do calculations of the right side of the screen to confirm area, etc.
The driver placement is xo in the Closed End of Transmission Line section.
Paul Ebert said:
The whole combination of driver Fs/Vas, length, geometry and post size/length determines the shape of the FR plot. In general, a 40" straight MLTL or low taper TQWT will have F3 in the 40's.
The larger the port, the stronger the high order harmonics. You want the smallest port that you can live with before port noise becomes a problem. Given the low power levels this design will encounter before significant distorting sets in, 4" should be adequate.
BTW, one hint that your design needs some work is that you needed 0.6 lb/ft^3 stuffing to flatten out the humps and valleys. If you need more than 0.25 lb/ft^3, rethink what you are doing.
Understand that I spend maybe 15 minutes modeling this design. You should take this as a starting point and go from there.
GM said:
Oops!
Fortunately, the resulting plots are nearly identical. I've replaced the pdf.
Revised MathCAD worksheet
Bob
Greets!
FYI, since neither sections are 'locked', you can cut n' paste the 'X0 = L*x.xxx' into the 'ML TQWT Geometry......' section if you prefer.
Hmm, the way I calc them, I get the same FR plot except with a slightly smoother (though I imagine inaudible) HF BW with a much shorter straight ML-TL and slightly longer vent:
L = 26.1"
SO/SL = 171.434"^2
X0 = L*0.466
rp = 2"
Lp = 3.5"
same stuffing density/location as yours
Anyway, FWIW from my own stone age (minimal/coarse measurements) research I concluded that the pipe length basically just reduces the vent length, net Vb determining the shape ('fullness') of the FR/roff-off slope with driver location (and taper if applicable) determining the overall smoothness. This assumes the vent is near/at the bottom.
Hmm, I deleted both 'Qtd' calcs and the WS seems to work fine, with no obvious changes that I could see, so just brought it to your attention in case you/others used it for doing 'effective' Qts calcs.
GM
FYI, since neither sections are 'locked', you can cut n' paste the 'X0 = L*x.xxx' into the 'ML TQWT Geometry......' section if you prefer.
Hmm, the way I calc them, I get the same FR plot except with a slightly smoother (though I imagine inaudible) HF BW with a much shorter straight ML-TL and slightly longer vent:
L = 26.1"
SO/SL = 171.434"^2
X0 = L*0.466
rp = 2"
Lp = 3.5"
same stuffing density/location as yours
Anyway, FWIW from my own stone age (minimal/coarse measurements) research I concluded that the pipe length basically just reduces the vent length, net Vb determining the shape ('fullness') of the FR/roff-off slope with driver location (and taper if applicable) determining the overall smoothness. This assumes the vent is near/at the bottom.
Hmm, I deleted both 'Qtd' calcs and the WS seems to work fine, with no obvious changes that I could see, so just brought it to your attention in case you/others used it for doing 'effective' Qts calcs.
GM
Bob and Martin,
Thanks so much for your help. I'm learning some good info, here. I've got a lot of more general questions related to full range design (definitely, a novice). Since I don't want to hijack this thread, I think I'll take those to a new one.
There is one thing I'm curious about regarding the B102. An obvious difference between it and the driver used in the Druid is the aluminum phase plug. Ignoring the issue of making such a plug, is adding one something that can be done DIY? Also, what does such a plug buy you?
Paul
Thanks so much for your help. I'm learning some good info, here. I've got a lot of more general questions related to full range design (definitely, a novice). Since I don't want to hijack this thread, I think I'll take those to a new one.
There is one thing I'm curious about regarding the B102. An obvious difference between it and the driver used in the Druid is the aluminum phase plug. Ignoring the issue of making such a plug, is adding one something that can be done DIY? Also, what does such a plug buy you?
Paul
I have no idea what the B102 driver is but I am sure a phase plug could be added in some form. Either a shaped plug or a socket head seems to be used most often. If you look in my gallery I added a picture of a homemade phase plug for a Lowther DX4 ML TQWT design. I have also seen people suspend a ping-pong ball in front of the driver, so suspending an object might be a method that does not require altering the driver if that is an issue.
I have experimented a little with phase plugs on my Lowther drivers. I have listened with no phase plug, the original bullet phase plug, the "shower head" phase plug from the DX4s, and a lightbulb phase plug from one of the horn applications. The light bulb phase plug seemed to degrade the high end sound. There was not much difference with the other three and if anything the original bullet sounded marginallly better then the other options. These are my opinions for my drivers in my room, your driver and situation might yield different results.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.