I bought an LCR meter to measure the values of unmarked crossover components. I now realise that the meter measures impedances and calculates values for LCR. The issue is that the values displayed depend on the frequency setting. However when I buy an inductor/resistor/capacitor, they aren't specified as having a particular value at a particular frequency. So how do I know which measurement to use to obtain the actual value of the component in crossover? Thanks.
Check this out. Which meter do you have?
https://www.keysight.com/us/en/assets/7018-06840/application-notes/5950-3000.pdf
https://www.keysight.com/us/en/assets/7018-06840/application-notes/5950-3000.pdf
Last edited:
Thanks. I'll have a read through. I have a reasonable low end meter, a Tenma 72-10465
https://www.newark.com/tenma/72-104...52e2f2f&CMP=KNC-BUSA-GEN-SKU-Test-Measurement
https://www.newark.com/tenma/72-104...52e2f2f&CMP=KNC-BUSA-GEN-SKU-Test-Measurement
The chart on pp. 12-14 of the manual should help you to select the mode needed.
TENMA 72-10465 OPERATING MANUAL Pdf Download | ManualsLib
TENMA 72-10465 OPERATING MANUAL Pdf Download | ManualsLib
Yes, that´s he cheap way to do it.
The better way is to use a real LCR bridge.
They "shouldn´t", at least within the Audi8o range, but now you know displayed value is an approximation, not the real deal.
That said, it´s usually "close enough".
I still use my trusty Leader 740 LCR bridge bought in the 80´s 😀
but as a Plan B, I would use the one you have but if possible using frequencies relatively close to the crossover frequency.
Say, testing a 2kHz crossover?
If possible try to use that, or, say, within an octave.
So called "real life testing".
See above suggestion.
Drivers are artfully crafted to produce flat output for constant voltage (as far as possible).
A crossover is a circuit that provides 50% of the that voltage at the point you want as your crossover (not that you can't finagle slopes and frequencies of the upper and lower drivers any way that delivers good measurements and good sound to your chair).
In the case of a treble driver, there's also the L-pad.
So working in the abstract is just plain abstract.
B.
A crossover is a circuit that provides 50% of the that voltage at the point you want as your crossover (not that you can't finagle slopes and frequencies of the upper and lower drivers any way that delivers good measurements and good sound to your chair).
In the case of a treble driver, there's also the L-pad.
So working in the abstract is just plain abstract.
B.
Last edited:
Yes, thank you JM, I might have been wiser to save up for a bridge. The nearest frequency to crossover on my meter is 1kHz.
Ben, if I'm interpolating correctly, you don't have much faith in these LCR meters?
Ben, if I'm interpolating correctly, you don't have much faith in these LCR meters?
There is nothing wrong with your LCR meter.
Its accuracy is better than any loudpspeaker crossover components.
It also offers a wide selection of measurement frequency, which you do not need.
For a 2-way crossover, the crossover frequency is likely to be above 1kHz.
So all the components can be measured at 1kHz.
Afterall, you are after the nomial value, not absolute accuracy.
For a 3-way crossover, the components of the bass driver are probably not required to work at 1kHz.
So you can measure those at 100Hz or 120Hz (either will do, does not matter).
And the rest at 1kHz.
At least that it what I would do.
If you want to find out how good those components are, you can measure them at 100Hz, 1kHz and 10kHz.
Compare those values will tell you how wide-band your crossover components are.
Cheers,
Patrick
Its accuracy is better than any loudpspeaker crossover components.
It also offers a wide selection of measurement frequency, which you do not need.
For a 2-way crossover, the crossover frequency is likely to be above 1kHz.
So all the components can be measured at 1kHz.
Afterall, you are after the nomial value, not absolute accuracy.
For a 3-way crossover, the components of the bass driver are probably not required to work at 1kHz.
So you can measure those at 100Hz or 120Hz (either will do, does not matter).
And the rest at 1kHz.
At least that it what I would do.
If you want to find out how good those components are, you can measure them at 100Hz, 1kHz and 10kHz.
Compare those values will tell you how wide-band your crossover components are.
Cheers,
Patrick
Last edited:
Those type of meters are ok, I own a Keysight LCR meter that measures up to 100KHz and I love it.
The thing with real inductors and capacitors is that they do not exhibit the same inductance/capacitance at all frequencies, however as JMFahey said, at audio frequencies they should stay pretty much constant. Try measuring them close to the frequency that you are going to be using them, 1KHz should be ok for a crossover.
I've seen cheap film capacitors that at low frequencies give their rated capacitance but at higher frequencies drop down dramatically, so its important to test a component at different frequencies to check its stability. Since you are probably using air core inductors your meter should work just fine!
Last edited:
Thanks for that Patrick. You would say from experience that the tolerance of measurements made this way will not substantially effect the performance of the crossover?
> You would say from experience that the tolerance of measurements made this way will not substantially effect the performance of the crossover?
If any "high-end" crossover component money can buy is not better than 5%, what is the point of measuring better than 0.5% ?
Components | Hifi Collective
To think that any loudspeaker manufacturer would use better components is an illusion.
Irrespective of how expensive.
Patrick
If any "high-end" crossover component money can buy is not better than 5%, what is the point of measuring better than 0.5% ?
Components | Hifi Collective
To think that any loudspeaker manufacturer would use better components is an illusion.
Irrespective of how expensive.
Patrick
No, having such a meter is wonderful. The components of a crossover have very big tolerances and should always be measured to find matching pairs. There are much cheaper ways to match, but the meter is a sophisticated method. I would say I have no faith in any other aspect of the process of making a crossover and therefore like an empirical approach.
When you have found the components which produce the audio result you want, then you can use your meter (1) to determine what their values are and (2) to measure the components that get delivered to your house in the future for use in that crossover.
(Of course, as I have stated many times in this forum, it makes little sense today for a DIY person to use this kind of crude passive crossover when DSP can be used.)
B.
Last edited:
That's correct if you are in control of the whole process. If I'm coming along later and measuring one of your components with my meter, I won't necessarily get the same reading as you and if I order a replacement based on my measurement I may end up with a different value to your intended one. However, it seems that the manufacturing tolerances involved may make such issues less significant.
"Later" is just when the meter is valuable. You measure the component from the proband crossover, order 3 nominally of the same value, and then use whichever one comes closest to the proband.
B.
Your method is too out of fashion. Serious members are not used to improvise on impromptu instrumentation.Lojzek said:
This is one of the major ailments of these fora. Technical creativity is frowned upon.
About caring about the component value at crossover frequency:
This is the intuitive way to think about it, but not good enough. A case in point is the inductors that go in series with a woofer. Usually, in a passive crossover, system sensitivity is governed by the sensitivity of the woofer. A good practice would therefore be to employ inductors with the lowest series resistance. In this scenario, a mere 0.5 ohm can significantly affect system efficiency or cause imbalance between driver levels.
The very first thing one can do to ensure that the measured value of an inductor is correct is to measure its series resistance a.k.a. DC resistance. A good inductance meter will report an R component of the L+R measurement correctly, or at least close to its DC resistance.
FWIW, REW recently got an upgrade to its passive component measuring ability. It does not do single-point measurement, but also does a curve-fit of its determined L+R parameters over the measured impedance plot. The calibration procedure for the measurement setup includes compensation for the series resistance of the measurement probes, so results are very good.
This is the intuitive way to think about it, but not good enough. A case in point is the inductors that go in series with a woofer. Usually, in a passive crossover, system sensitivity is governed by the sensitivity of the woofer. A good practice would therefore be to employ inductors with the lowest series resistance. In this scenario, a mere 0.5 ohm can significantly affect system efficiency or cause imbalance between driver levels.
The very first thing one can do to ensure that the measured value of an inductor is correct is to measure its series resistance a.k.a. DC resistance. A good inductance meter will report an R component of the L+R measurement correctly, or at least close to its DC resistance.
FWIW, REW recently got an upgrade to its passive component measuring ability. It does not do single-point measurement, but also does a curve-fit of its determined L+R parameters over the measured impedance plot. The calibration procedure for the measurement setup includes compensation for the series resistance of the measurement probes, so results are very good.
Last edited: