Although I'm a novice at cabinet design, I'm helping someone with his. The resulting SPL plot showed an odd dip around 425hz.
At first I thought there was crossover error but I had him run an SPL sweep of the woofer (in cabinet) with no crossover.
The dip was still there. I then had him run and impedance sweep. Low and behold there is a big dip in impedance at that frequency.
The impedance at that point is 2 ohms.
My question is this. How can the impedance ever drop below the DCR (5.08) of a driver (Celestion TF1225) with no crossover?
Thanks in advance.
Mike
At first I thought there was crossover error but I had him run an SPL sweep of the woofer (in cabinet) with no crossover.
The dip was still there. I then had him run and impedance sweep. Low and behold there is a big dip in impedance at that frequency.
The impedance at that point is 2 ohms.
My question is this. How can the impedance ever drop below the DCR (5.08) of a driver (Celestion TF1225) with no crossover?
Thanks in advance.
Mike
I'll venture a wild guess:
The impedance at resonance is also much lower than the specs for the 8 Ohm version.
In the datasheet Celestion shows something like 67 Ohm - but your measurement shows less than half of that.
This driver also comes in a 4 Ohm version - could this be the reason?
Cheers
Martin
The impedance at resonance is also much lower than the specs for the 8 Ohm version.
In the datasheet Celestion shows something like 67 Ohm - but your measurement shows less than half of that.
This driver also comes in a 4 Ohm version - could this be the reason?
Cheers
Martin
The free air impedance sweep matched the specs (8 ohm version). It's seems to be the ported box that is messing with things in a way
I don't understand.
I don't understand.
I'll have him do that, but the dip in the SPL (with crossover) to me confirms that the impedance dip is there. The voltage drop across the woofer drops by 30(?) db in that area.
There is also something odd going on with the resonance: In the Celestion datasheet the resonance frequency is quoted as 55.6 Hz - but your measurement shows 86.1 Hz.
I will not try to explain this - but it makes me wonder if the speaker is a properly working TF1225 and/or if there could be some kind of fault in the measurement setup.
Cheers
Martin
I will not try to explain this - but it makes me wonder if the speaker is a properly working TF1225 and/or if there could be some kind of fault in the measurement setup.
Cheers
Martin
https://audioxpress.com/article/testing-loudspeakers-which-measurements-matter-part-2
"A vented enclosure will have a double peaked impedance plot at low frequencies. In this case the saddle point between the peaks approximates the vented box tuning frequency. (The woofer voice coil inductance and the crossover circuit may cause a slight shift in the saddle point relative to the actual box frequency.)"
Thanks for the replies. His DATS hardware seems to be working fine. Unfortunately nothing seems to explain to me how a ported box can create a 427hz notch filter on a raw driver 😕
Since it can't, seems likely there is something in parallel with the driver under test.
The phase of the impedance measurement is also odd. Are you sure the plot above is the woofer only? And have you confirmed the DATS is producing good data with a known speaker where you can compare his measurement to manufacturer's or someone else's data for magnitude and phase? Yes, I saw that you ran a free air sweep, but I assume that was just looking at magnitude.
Has he swapped woofers to see if they are both behaving the same way (assuming he has two)?
Has he measured nearfield frequency response (less than 1/2 inch microphone distance) for the woofer and the port with no crossover? Anything weird there? And yes, I saw that you said this originally started with a frequency response measurement. Nearfield measurements remove a lot of room effects/measurement problems. There are many ways to generate inaccurate acoustical measurements, so getting back to basics can help.
Has he swapped woofers to see if they are both behaving the same way (assuming he has two)?
Has he measured nearfield frequency response (less than 1/2 inch microphone distance) for the woofer and the port with no crossover? Anything weird there? And yes, I saw that you said this originally started with a frequency response measurement. Nearfield measurements remove a lot of room effects/measurement problems. There are many ways to generate inaccurate acoustical measurements, so getting back to basics can help.
Last edited:
Problem solved. So much for long distance trouble shooting. We're trying to add a front port to a Klipsch Hersey cabinet and make it a 2-way. I reused the squawker hole for a front port and plugged the tweeter hole. Using an external crossover he was to bypass the internal crossover and connect the speaker terminal directly to the woofer. He thought he could use the internal crossover to mount the wires. The woofer circuit has a 4mh series inductor and a 33uf shunt capacitor. Those ended up in series to ground, and turned into a 430hz notch. Once I told him to completely bypass the internal crossover the problem went away.
Sorry for the trouble. If I was there to see what he did it would have been a quick fix.
Mike
Mike,
No trouble, like Sherlock Holmes liked to say: "when you have eliminated the impossible, whatever remains, however improbable, must be the truth".
That works just as well for long distance trouble shooting as near, though I didn't think it improbable there was something in parallel with the driver under test.
Art
As you say, impedance is the sum of an inductor's resistance and reactance, but sums (the total amount resulting from the addition of two or more numbers, amounts, or items) add.
8 ohms of DCR and 8 ohms of reactance sum to a greater impedance.
https://unacademy.com/content/neet-.../differences-between-reactance-and-impedance/
If the opposite were true, we would see impedance curves with impedance curves lower than DC resistance.
The fact is we never see a valid impedance curve of a driver's impedance falling to less than it's DCR, though at very low frequencies the impedance may be equal to the driver's DCR.
That fact made it easy to deduce there was a circuit in parallel with the driver in the original post reducing the impedance below it's DCR.
Art