Hello
I tried to measure a driver parameters using LIMP, but the results sounds not correct:
Thiele-Small parameters:
Fs = 108.07 Hz
Re = 3.90 ohms[dc]
Le = 9.61 uH
L2 = 4785.81 uH
R2 = 25.74 ohms
Qt = 0.02
Qes = 0.02
Qms = 0.14
Mms = 1.80 grams
Rms = 8.688394 kg/s
Cms = 0.001208 m/N
Vas = 1.65 liters
Sd= 31.17 cm^2
Bl = 15.489085 Tm
ETA = 10.09 %
Lp(2.83V/1m) = 105.26 dB
Here is the impedance curve measured with the added mass method (added mass: 2.3grams):
It's a 3" full range driver, 4 ohms, I tested it in free air with this schematic (I used only left channel line out since line out is stereo...)
Is it really possible to have a Qt of 0.02 ?
I tried to measure a driver parameters using LIMP, but the results sounds not correct:
Thiele-Small parameters:
Fs = 108.07 Hz
Re = 3.90 ohms[dc]
Le = 9.61 uH
L2 = 4785.81 uH
R2 = 25.74 ohms
Qt = 0.02
Qes = 0.02
Qms = 0.14
Mms = 1.80 grams
Rms = 8.688394 kg/s
Cms = 0.001208 m/N
Vas = 1.65 liters
Sd= 31.17 cm^2
Bl = 15.489085 Tm
ETA = 10.09 %
Lp(2.83V/1m) = 105.26 dB
Here is the impedance curve measured with the added mass method (added mass: 2.3grams):
An externally hosted image should be here but it was not working when we last tested it.
It's a 3" full range driver, 4 ohms, I tested it in free air with this schematic (I used only left channel line out since line out is stereo...)
Is it really possible to have a Qt of 0.02 ?
Last edited:
Hello
Is it really possible to have a Qt of 0.02 ?
Hi, In short, no its not possible, rgds, sreten.
Hey,
I had so many issues with getting Arta to run right, I finally gave up. I was about to break down and buy a WT2 or WT3 when a used WT3 came up on the forums and I snagged it. Works like a charm. Although I don't use it to measure T/S.
Looks like the same issues I had too. Sometimes it looked accurate. Sometimes it would read 30ohms. Even 200ohms.
Good luck
I had so many issues with getting Arta to run right, I finally gave up. I was about to break down and buy a WT2 or WT3 when a used WT3 came up on the forums and I snagged it. Works like a charm. Although I don't use it to measure T/S.
Looks like the same issues I had too. Sometimes it looked accurate. Sometimes it would read 30ohms. Even 200ohms.
Good luck
If Re is 3.9R (which is easy to check with a multi-meter), but your impedance curve starts at over 26R at low frequencies, then it is obvious straight away that something is grossly wrong!
As I replied in another thread about using LIMP, first of all attach a known resistor to the jig (10R is ok) and see if the Z plot is a straight line at 10R. If not then something has screwed up.
I find LIMP easy to use and very accurate, provide that your sound card can drive an undistorted signal into the load, which is the case for all sound-card based measurement software.
As I replied in another thread about using LIMP, first of all attach a known resistor to the jig (10R is ok) and see if the Z plot is a straight line at 10R. If not then something has screwed up.
I find LIMP easy to use and very accurate, provide that your sound card can drive an undistorted signal into the load, which is the case for all sound-card based measurement software.
I wonder if it's a soundcard impedance problem, because the only thing wrong is the vertical offset, FS looks ok. I don't have a lot of noise/distorsion if I set the recording volume not too high.
I will try you idea tonight. My resistor is about 100ohm, I only get realistic results if I tell LIMP my resistor is something like 4 ohm (the impedance curve is a lot lowered). It's why I think of an headphone-out impedance problem. Maybe I can find a "cheating formula" by comparing several resistors with several results given by LIMP, then applying a correction factor
I will try you idea tonight. My resistor is about 100ohm, I only get realistic results if I tell LIMP my resistor is something like 4 ohm (the impedance curve is a lot lowered). It's why I think of an headphone-out impedance problem. Maybe I can find a "cheating formula" by comparing several resistors with several results given by LIMP, then applying a correction factor
with me Arta is working great, but i have not measuring box, instead i have JIG from speaker workshop tutorial but with different resistor value
????? ??? - ?????? - SW tutorial ENG.pdf
????? ??? - ?????? - SW tutorial ENG.pdf
The Z(out) of the headphone socket should be compensated for by the fact that the soundcard measures the voltage each side of the series resistor in the jig.
It's ok with the resistor alone:
When I connect the speaker : (without changing anything else)
The reference channel is set properly, if I invert it I get a reversed curve (lowest impedance at fs)
I just tried with a 23 ohm resistor and I get better results :
The whole impedance offset seems sensitive to the resistor I use (even if I always tell LIMP its value)... So which value should I use ?
An externally hosted image should be here but it was not working when we last tested it.
When I connect the speaker : (without changing anything else)
An externally hosted image should be here but it was not working when we last tested it.
The reference channel is set properly, if I invert it I get a reversed curve (lowest impedance at fs)
I just tried with a 23 ohm resistor and I get better results :
An externally hosted image should be here but it was not working when we last tested it.
The whole impedance offset seems sensitive to the resistor I use (even if I always tell LIMP its value)... So which value should I use ?
As the reference resistor is in series with the device being measured, then 33 Ohms will be ok, as all headphone outputs will drive a nominal 32 Ohm headphone load.
I don't understand why LIMP isn't accepting the value of your reference resistor.
If you do a run without any load connected, the plot will be of the input impedance of the sound card, usually around 10k. If you ask LIMP do perform a calibration without any load connected, then it will calibrate out the soundcard Z(in). The soundard Z(in) doesn't matter much if you're measuring 8 Ohm speakers, as the maximum Z at resonance might only be 30 Ohms or so, but I measure much higher Z speakers (for 100V line operation) so calibrating out the effect of Z(in) is required.
Put a 10 Ohm resistor in place of the speaker and see what you get. With a 10 Ohm load LIMP should give an accurate result with a reference resistor in the range 22R to 100R.
I don't understand why LIMP isn't accepting the value of your reference resistor.
If you do a run without any load connected, the plot will be of the input impedance of the sound card, usually around 10k. If you ask LIMP do perform a calibration without any load connected, then it will calibrate out the soundcard Z(in). The soundard Z(in) doesn't matter much if you're measuring 8 Ohm speakers, as the maximum Z at resonance might only be 30 Ohms or so, but I measure much higher Z speakers (for 100V line operation) so calibrating out the effect of Z(in) is required.
Put a 10 Ohm resistor in place of the speaker and see what you get. With a 10 Ohm load LIMP should give an accurate result with a reference resistor in the range 22R to 100R.
Oh I just find what was my mistake : the output and input grounds were connected together... I disconnected them and all seems correct now :
Sorry stupid mistake ^_^'
Can't wait to measure my other unknown cheap drivers. Hope the values are reliable enough to do some simulations in WinISD
An externally hosted image should be here but it was not working when we last tested it.
Sorry stupid mistake ^_^'
Can't wait to measure my other unknown cheap drivers. Hope the values are reliable enough to do some simulations in WinISD
hmm maybe i've been doing the same thing.
could you draw what the circuit would look like if you were drawing it using a crocodile clips, and jack to phono leads etc. kind of like an individualised component over-view so i can check if i've cocked it up to?
An externally hosted image should be here but it was not working when we last tested it.
Sorry i'm not very good at paint !
I have the same problem, data not oke with the visaton high end wsp26s who i did use as a reference.
FS is not 19 but 29 QMS is a lot higher.
I did use the pc headphone output, my jig with amp is not ready yet.
Fs = 29.06 Hz
Re = 5.96 ohms[dc]
Le = 95.13 uH
L2 = 967.48 uH
R2 = 5.04 ohms
Qt = 0.35
Qes = 0.37
Qms = 6.43
Mms = 43.88 grams
Rms = 1.194908 kg/s
Cms = 0.000688 m/N
Vas = 115.94 liters
Sd= 346.36 cm^2
Bl = 11.068253 Tm
ETA = 0.70 %
Lp(2.83V/1m) = 91.82 dB
Added Mass Method:
Added mass = 18.00 grams
Diameter= 21.00 cm
WSP 26 S 8 OHM
Did read input output grounds, but not much differents. Vas of this speaker is 230 liter, itr give
6.22
Loudspeaker parameters:
Fs = 29.40 Hz
Re = 5.95 ohms[dc]
Le = 98.13 uH
L2 = 967.90 uH
R2 = 5.09 ohms
Qt = 0.36
Qes = 0.38
Qms = 6.55
Mms = 40.97 grams
Rms = 1.126327 kg/s
Cms = 0.000718 m/N
Vas = 6.22 liters
Sd= 78.54 cm^2
Bl = 10.788861 Tm
ETA = 0.04 %
Lp(2.83V/1m) = 79.31 dB
Added Mass Method:
Added mass = 18.00 grams
Diameter= 10.00 cm
thanks for now.
FS is not 19 but 29 QMS is a lot higher.
I did use the pc headphone output, my jig with amp is not ready yet.
Fs = 29.06 Hz
Re = 5.96 ohms[dc]
Le = 95.13 uH
L2 = 967.48 uH
R2 = 5.04 ohms
Qt = 0.35
Qes = 0.37
Qms = 6.43
Mms = 43.88 grams
Rms = 1.194908 kg/s
Cms = 0.000688 m/N
Vas = 115.94 liters
Sd= 346.36 cm^2
Bl = 11.068253 Tm
ETA = 0.70 %
Lp(2.83V/1m) = 91.82 dB
Added Mass Method:
Added mass = 18.00 grams
Diameter= 21.00 cm
WSP 26 S 8 OHM
Did read input output grounds, but not much differents. Vas of this speaker is 230 liter, itr give
6.22
Loudspeaker parameters:
Fs = 29.40 Hz
Re = 5.95 ohms[dc]
Le = 98.13 uH
L2 = 967.90 uH
R2 = 5.09 ohms
Qt = 0.36
Qes = 0.38
Qms = 6.55
Mms = 40.97 grams
Rms = 1.126327 kg/s
Cms = 0.000718 m/N
Vas = 6.22 liters
Sd= 78.54 cm^2
Bl = 10.788861 Tm
ETA = 0.04 %
Lp(2.83V/1m) = 79.31 dB
Added Mass Method:
Added mass = 18.00 grams
Diameter= 10.00 cm
thanks for now.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.