SY
It is clear you have no idea what you are talking about. It is just your usual snide, unproductive comments. If you actually had something to say you would be specific.
I will be interested in comments from the competent folks like Vaccy or Brad.
BTY the last time you were this grumpy was just before you got a new job.
I take it the scale is not milli-anythings, is it ohms? Is this just the real part of the impedance? What about the imaginary part or phase angle? What are the blue lines?
I look forward to seeing the before and after plots of the speaker output level. Does the Zobel network dissipate much power?
The blue lines are fixed resistors of 4, 8, 12, 16.....
This is the real part of the plot. The resistors in the network are rated at 150 watts. The capacitors for 30 amps.
I can look at phase angle with a bit of effort. But the design goal is to flatten the frequency response for the long 500'+ cable lengths. Or resistances above an ohm. As the impedance swing now is limited to less than 2 to 1 this will give a frequency response variation that should be much smaller and smooth enough to be equalized for even better results.
The secondary effect of the network chosen is to provide strong damping matching the resonances. This seems to have worked.
SY if you have a question just ask.
Last edited:
I did. You didn't care to answer. I've never seen an impedance plot that looked that bad so naturally wondered what went wrong.
Carry on.
Ohms? Are they in your picture? Are they part of this Zobel circuit?
So in the "before" plot what are we looking at? Does that impedance include the long cable?
Ed,
Is that impedance graph a measurement at the end of the 500' of cable and terminated with the zobel and nothing else or is there a fixed load or speaker hanging on the end of the 500' line? The yellow line looks pretty bad for just an impedance measurement, why so ragged? The red and green lines look like a typical reflex impedance curve with the rise I suspect is caused by the long 500' of cable.
Is that impedance graph a measurement at the end of the 500' of cable and terminated with the zobel and nothing else or is there a fixed load or speaker hanging on the end of the 500' line? The yellow line looks pretty bad for just an impedance measurement, why so ragged? The red and green lines look like a typical reflex impedance curve with the rise I suspect is caused by the long 500' of cable.
Last edited:
Real and imaginary go together while magnitude and phase do also. Real alone is not totally useful except when looking at resistors. That yellow plot looks like dozens of high Q resonances, just curious. BTW some of the plots on your 1K - 2 Ohm divider experiment had sudden jumps in the baseline which also seemed strange.
Yes ohms. No they are from the test runs to verify the set up.
The impedance is derived by using a current source and looking at the voltage.
As above using a current source to test impedance. Acoustic tests will be from a voltage source using 1000' feet of cable.
As to the HF bouncing so much, I suspect what is being shown are the reflections off the front grill. This is an extreme weather loudspeaker. The outer most grille layer is perforated metal with about 65% open area. Behind that is foam which is probably 85% open or better and then a layer of stainless steel wire woven screen with maybe 70% open area. So I see a very clear resonance bounce. The change from virtually no damping to a very high one seems to have cleared that up quite a bit. But it does still show up.
Last edited:
Yes Scott it is magnitude. Yes as mentioned the yellow HF plot is lots of resonances that are not damped either acoustically or electrically.
As to the the baseline experiment jumps are normally from mismatched auto ranging. Of course as the first test was horrible it could be from all sorts of other issues including some we clearly don't agree about.
Last edited:
All looks normal/familiar resonances except the yellow data line...... interactions?
High distortion at those resonance freqs.
THx-RNM
As long as oatmeal and milk are in the diet I can't see what would be missing. Vegans would have it tough, but us veggies do fine. Oatmeal with milk is also amino complete which is handy.
About a hundred bucks would get you a "Woofer Tester" or "DATS" device. It makes speaker impedance measurements, mag and phase, trivially easy (andf the plots readable).
Thanks I have one and am aware of the limitations. Had a consultant talk to me about the results a woofer tester yielded in a baseball park. They are fine for the intended purpose. Not enough drive current for long cable runs.
I have no trouble reading the curves.
Last edited:
too much noise pickup by long cable runs? Other than maybe that, I'd think impedance is impedance. No offense on the readability comment, but just about everyone had trouble reading the one you posted, is all. Having a scale in the low millithingys is either a ribbon driver, or needs more explaining.
No I only take offense from SY. 🙂 Just having tried the woofer tester myself and seeing how it was powered it just didn't have enough current to produce good signal to noise.
The blue lines were 4 ohms apart. They are the calibration run with resistors. If you look back a bit you can see the results of using a solderless breadboard to hold the chain. It wasn't exactly successful. So I used direct connections. That gave nice straight lines.
So post Zobel the midrange is between 4 and 8 ohms and the problem HF runs between 8 and 16 ohms. The driving cable has four sections, two run about 8 ohms and two about 16 ohms.
- Status
- Not open for further replies.