LCR meter at 10KHz

[StarryNight]

in scoping (haha) out some LCR meters I came across this one which seems to measure at 100HZ 1KHz and 10KHz. I think I remember a thread where some of you were saying there are ACR characteristics at the 10KHz level that would be good to measure. Seeing as it's the same price as the Extech (which does 120HZ and 1KHz) I wonder if it would be a better choice? Any thoughts?

http://cgi.ebay.com/NEW-PROFESSIONAL...QQcmdZViewItem

[Possum]

....

I keep waiting for someone to buy one of these and report back :-) Supposedly, no, they won't read inductance correctly at 10khz, but they WILL read AC resistance correctly. That alone would make it useful to me as I use those readings in conjuction with inductance....

[Mike Sulzer]

I keep waiting for someone to buy one of these and report back :-) Supposedly, no, they won't read inductance correctly at 10khz, but they WILL read AC resistance correctly. That alone would make it useful to me as I use those readings in conjuction with inductance....

A meter operating at a frequency f can measure the amplitude and phase of the complex impedance. That is just two numbers. At 10 KHz, a typical impedance is capacitive, but not purely so because the inductance still matters. There is the resistance of the coil that potentially matters at all frequencies, and there are losses from eddy currents that vary in frequency in a complicated way. It is difficult to know what the ac resistance measurement says about this. You have too many variables and not enough information.

On the other hand, why is 10 KHz important? If you load the pickup with the cable capacitance, it has little response there. If you turn the volume down part way, you have instead an RC filter that cuts high frequencies. The response of a guitar speaker is way down at 10KHz. So is the hearing of the typical rock guitarist.

I believe that the response at 10 KHz is important because it is necessary to model the pickup in that range in order to understand its complete equivalent circuit. But that is too academic for most people.

[Possum]

...

I think it is important because the distance between 1khz and 10khz is a heck of alot of missing information, having a plot point at 10khz gives you maybe an idea of what is happening along the way. Maybe it might not even be accurate but if you get used to what kind of readings you get at that frequency with certain types of pickups it can give you an idea of what it means, and also will give you readings when you change one thing in the pickup for comparison. Those meters aren't cheap though so, probably a cheaper way to do something decent. Audacity will generate a frequency response chart, just never find the time to rig something up to drive the pickup for testing....

[Mike Sulzer]

I think it is important because the distance between 1khz and 10khz is a heck of alot of missing information...

Yes! What happens at the loaded resonance, 2-5 KHz, is really important.

A follow-up discussion to the one Joe started a while ago would be useful. I now have a system that measures amplitude and phase at an arbitrary number of frequencies, say 512 to 2048. (The software is not cheap, though). The problem is now to establish a useful circuit model of the pickup. There is one surprising aspect to this which explains why this is difficult for LCR meters. I will start a discussion tonight or on the weekend.

[Possum]

...

What do you mean a "circuit model of a pickup?" If you're talking simulations thats going to be totally useless. In my opinion its best to just measure the pickup and not try to simlulate pots, caps and a guitar cord and amp added on. the pickup itself isn't going to change, but all that other stuff will never be the same for every different player. Guitar cords are all over the place for instance....

[Mike Sulzer]

What do you mean a "circuit model of a pickup?" If you're talking simulations thats going to be totally useless. In my opinion its best to just measure the pickup and not try to simlulate pots, caps and a guitar cord and amp added on. the pickup itself isn't going to change, but all that other stuff will never be the same for every different player. Guitar cords are all over the place for instance....

No, I mean the circuit components that adequately describe the electronic behavior of a pickup. Not what you connect to it, although that matters for the overall performance of the device in a guitar. For example, the pickup is often represented by an inductor with a resistor in series and some capacitance. But this is not adequate.

[salvarsan]

P90 at 10kHz

This is an impedance amplitude plot (i.e., has no phase info)
of one of Steve Kersting's P90 bridge pickups. Pickup Impedance
is not constant with frequency, nor is it's phase shift. It is not linear
through 10kHz.

This means that inductance has to compete with other electrical properties
and the best you can do is specify the test conditions along with your
measurement.

From 100 to 2000 Hz, you get a gently curved slope that implies
a reasonably pure inductance. After that, the distributed capacitance
manifests more strongly and you get a lumpy non-linear curve that
makes inductance measurements problematic.

-drh

[Mike Sulzer]

...you get a lumpy non-linear curve that
makes inductance measurements problematic.

-drh

This is an example of why measurements of a pickup over a wide frequency range are most useful if you model all the "components" that make up the equivalent circuit of the pickup.