Anyone else measured inductors with WT3?

[XtremeRevolution]

This is thoroughly confusing. I'm getting the feeling that my WT3 might be defective. Every single inductor I test is too high of a value. I calibrated the WT3 using the provided resistor and the shorted leads calibration, and here's what I'm coming up with.

Jazten .30mH - measures .34mH
Jazten .25mH - measures .28mH
Erse 3.5mH solid core - measures 3.8mH

The other day, I was building my Florians crossovers which I had to unwind a 2.0mH inductor, and it measured 2.2mH. I unwound it to where it would measure 1.85mH thinking it was just a fluke. I then measured the .15mH inductor and it came up .17mH. I ended up unwinding it since I needed a .12 value.

Anyone having the same experiences? Am I missing something, or are these inductors all actually higher in value than they should be?

[johnnyrichards]

I have the same results. It is what it is. I have no problems with caps and resistors, but I do with coils.

[XtremeRevolution]

I have the same results. It is what it is. I have no problems with caps and resistors, but I do with coils.

Ok, so should I be unwinding them, or should I just leave them alone? Does that mean that ones I already unwound are now a lower value than they should be? If so, does it mean that I need to go pick up an LCR somewhere just so I can be able to unwind my own coils accurately?

[woodtradesman_audio]

Time for an LCR maybe?

[Chris Roemer]

This is thoroughly confusing. I'm getting the feeling that my WT3 might be defective. Every single inductor I test is too high of a value. I calibrated the WT3 using the provided resistor and the shorted leads calibration, and here's what I'm coming up with.

Jazten .30mH - measures .34mH
Jazten .25mH - measures .28mH
Erse 3.5mH solid core - measures 3.8mH

The other day, I was building my Florians crossovers which I had to unwind a 2.0mH inductor, and it measured 2.2mH. I unwound it to where it would measure 1.85mH thinking it was just a fluke. I then measured the .15mH inductor and it came up .17mH. I ended up unwinding it since I needed a .12 value.

Anyone having the same experiences? Am I missing something, or are these inductors all actually higher in value than they should be?

I'm SURE you're measuring Le at 1k (and not at 10k - which is in the "preferences"), 'cause otherwise I think your values wouldn't even be close.

In general, coils measure high for me, but where yours are running some +10% high, mine only measure a percent or 2 high.


Chris

[PWR RYD]

I use mine to measure inductors all the time. I feel mine measures inductors VERY accurately and VERY repeatably. I have compared it to a high quality LRC meter (B&K) as well as a test rig I made up using a sine wave generator and calibrated resistor. All values agree within 0.5%. Like Chris mentioned, I set mine to measure at 10 kHz for small values (under 1 mH) and at 1 kHz for larger values.

FWIW - every Jantzen inductor I've purchased whether its a 20 gauge air core or a 15 gauge P-core has measured a few percent over spec. With the current cost of copper this really surprises me

[XtremeRevolution]

I'm SURE you're measuring Le at 1k (and not at 10k - which is in the "preferences"), 'cause otherwise I think your values wouldn't even be close.

In general, coils measure high for me, but where yours are running some +10% high, mine only measure a percent or 2 high.


Chris

You really think that will skew my numbers by 10%? See below:

I use mine to measure inductors all the time. I feel mine measures inductors VERY accurately and VERY repeatably. I have compared it to a high quality LRC meter (B&K) as well as a test rig I made up using a sine wave generator and calibrated resistor. All values agree within 0.5%. Like Chris mentioned, I set mine to measure at 10 kHz for small values (under 1 mH) and at 1 kHz for larger values.

FWIW - every Jantzen inductor I've purchased whether its a 20 gauge air core or a 15 gauge P-core has measured a few percent over spec. With the current cost of copper this really surprises me

This is really strange. Like I said, mine just doesn't seem to be very measuring correctly right now. I might even call PartsExpress to have them replace it.

The thing is, if the default is 1khz for large values, my 3.5mH inductor still shouldn't be measuring 3.8mH.

[XtremeRevolution]

Paging John Murphy?

The tech at PartsExpress I just chatted with told me I should talk to him about this.

[PWR RYD]

I don't believe it automatically defaults to 1 kHz. If you change it to 10 kHz under the "Edit>Preferences" tab then it stays 10 kHz until you change it to 1 kHz and vise versa. You need to go to that tab and see what it is set for.

[Chris Roemer]

I don't believe it automatically defaults to 1 kHz. If you change it to 10 kHz under the "Edit>Preferences" tab then it stays 10 kHz until you change it to 1 kHz and vise versa. You need to go to that tab and see what it is set for.

Yup.

Not really sure what 10kHz is for (tweeter T/S?).

I remember old CLIO data that would show Le at BOTH 1k and 10k. The values were NOT the same (I don't even think they were close? - but CRR).

With my (older) WT3, it doesn't measure uF directly, you have to measure resistance at a certain freq and do the math. NOW I always take my reading at 1000 Hz. uF and Le values near 1k are more relevant than at 10k anyway, 'cause that's a LOT closer to where your trying to roll something off (even most 3-way tweets cross closer to 1k than 10k).

Chris

[XtremeRevolution]

Well, its too late to save the inductors on my Florians speakers as they've already been unwound. I've entered the predicted values (accounting for the 10% error), and the crossover actually looks slightly better than what I had originally planned on. The top end is about 1db quieter, which I don't mind much.

I will re-measure the inductors I measured earlier, specifying a 1k Le and report back tonight.

[billfitzmaurice]

Mine never reads accurately, but it's no problem. I 'calibrate' it with the initial measurement, and assume that the coil is accurately spec'd from the manufacturer. Also, be careful not to coil the test leads or have them resting on any ferrous material, you'll read the inductance created in the leads by so doing.

[XtremeRevolution]

Mine never reads accurately, but it's no problem. I 'calibrate' it with the initial measurement, and assume that the coil is accurately spec'd from the manufacturer.

I guess that's what I'll have to do. Thanks.

[daver]

I use mine to measure inductors all the time. I feel mine measures inductors VERY accurately and VERY repeatably.

My WT3 also seems to measure accurately and repeatably.

From time to time, I sometimes take the impedance measurement file and run a curvefit through the data as a check. Typically, the air core coils that I use (18 or 20 ga for instance) are fairly consistent with the WT3 value, and fairly constant regardless of frequency. Often, they are slightly below the "nominal" value.

Maybe the 10% difference that you see consistently has something to do with your WT3 or computer settings?

Even so, all coils will have a tolerance. A lot of them that I use are stated to be +/- 3%. For all practical purposes, if you have a simulation program, see what happens if you vary the "nominal" value of the specified inductor by +/-3%, +/-5%, or even +/-10%. Is there an appreciable difference? Maybe or maybe not.

[XtremeRevolution]

My WT3 also seems to measure accurately and repeatably.

From time to time, I sometimes take the impedance measurement file and run a curvefit through the data as a check. Typically, the air core coils that I use (18 or 20 ga for instance) are fairly consistent with the WT3 value, and fairly constant regardless of frequency. Often, they are slightly below the "nominal" value.

Maybe the 10% difference that you see consistently has something to do with your WT3 or computer settings?

Even so, all coils will have a tolerance. A lot of them that I use are stated to be +/- 3%. For all practical purposes, if you have a simulation program, see what happens if you vary the "nominal" value of the specified inductor by +/-3%, +/-5%, or even +/-10%. Is there an appreciable difference? Maybe or maybe not.

Earlier today I plugged in the vales of what my coils probably are after I unwound them too far in PCD, and the difference was minimal. I had a bit of a higher bump between 1000hz and 2000hz, and my tweeter's output dropped about 1-2db overall. Nothing too serious. My reverse null looks prettier and has a -43db dip. Still looks pretty good.

[daver]

Earlier today I plugged in the vales of what my coils probably are after I unwound them too far in PCD, and the difference was minimal. I had a bit of a higher bump between 1000hz and 2000hz, and my tweeter's output dropped about 1-2db overall. Nothing too serious. My reverse null looks prettier and has a -43db dip. Still looks pretty good.

Depending on where the coils are in the circuit, you may notice a slight shift in crossover frequency. Phase tracking through the crossover still looks good.

How about the impedance? Any significant changes with that?

Looking at the frequency response that you have, the slight 1 or 2 dB dip from 2k to 10k might make these easier listening as well.

Keep us posted.

[XtremeRevolution]

Depending on where the coils are in the circuit, you may notice a slight shift in crossover frequency. Phase tracking through the crossover still looks good.

How about the impedance? Any significant changes with that?

Looking at the frequency response that you have, the slight 1 or 2 dB dip from 2k to 10k might make these easier listening as well.

Keep us posted.

The crossover frequency didn't change by a consequential amount. Impedance was virtually unchanged. I may have made these a bit better by accident. In any case, I hope to get them built this week, and may have them ready for DIY Chicago.

[Whitneyville1]

XR, are you loosely twisting the test leads? That does make a difference, even at 1K. Not owning a WT3, but seeing the basic design/operation principles, I wonder about plug contact and clip stray resistance easily causing errors. I'm basing this on the Speaker Workshop DIY version. I wonder if working on an anti-static/grounding mat could help? We always used those on our benches at "Fish Finders" when hooking up to the test gear ("Octapus"). A good 1% 1K resistor should "calibrate" fine at 1Khz, unless the internal connections inside the "black-box" are bad. Maybe call to P-E is in order?

[carlspeak]

This is thoroughly confusing. I'm getting the feeling that my WT3 might be defective. Every single inductor I test is too high of a value. I calibrated the WT3 using the provided resistor and the shorted leads calibration, and here's what I'm coming up with.

Jazten .30mH - measures .34mH
Jazten .25mH - measures .28mH
Erse 3.5mH solid core - measures 3.8mH

The other day, I was building my Florians crossovers which I had to unwind a 2.0mH inductor, and it measured 2.2mH. I unwound it to where it would measure 1.85mH thinking it was just a fluke. I then measured the .15mH inductor and it came up .17mH. I ended up unwinding it since I needed a .12 value.

Anyone having the same experiences? Am I missing something, or are these inductors all actually higher in value than they should be?

I think you've got a calibration issue. Just look thru the PE coil offerings. The largest tolerence I see is 5%. Your measurements are consistently well above 5%.
Maybe you'll get lucky and John L. will chime in here and make a suggestion about your question.

[John L. Murphy]

This is thoroughly confusing. I'm getting the feeling that my WT3 might be defective. Every single inductor I test is too high of a value. I calibrated the WT3 using the provided resistor and the shorted leads calibration, and here's what I'm coming up with.

Jazten .30mH - measures .34mH
Jazten .25mH - measures .28mH
Erse 3.5mH solid core - measures 3.8mH

The other day, I was building my Florians crossovers which I had to unwind a 2.0mH inductor, and it measured 2.2mH. I unwound it to where it would measure 1.85mH thinking it was just a fluke. I then measured the .15mH inductor and it came up .17mH. I ended up unwinding it since I needed a .12 value.

Anyone having the same experiences? Am I missing something, or are these inductors all actually higher in value than they should be?

Hello Xtreme,

If the unit completes the 1 k Ohm calibration and the test leads calibration with good smooth plots for both magnitude and phase then it should measure inductors accurately. Are the plots of magnitude and phase smooth or somewhat noisy?

I assume these are the +/-3% Jantzen coils sold by Parts Express that you are measuring. Your measurements are more than 10% high which is definitely outside the range of what seems reasonable to me. I would troubleshoot the setup a bit. In particular be keenly aware of any iron or steel near the coil as you are making measurements as this will skew inductance measurements to the high side. For example a screwdriver near the coil will definitely affect its inductance value.

Are you using the inductance measurement routine in the new v.2.0 software or the Le reading from an earlier version of software?

Measuring my own sample 0.30 mH coil from PE I consistently get 0.303 for ten measurements in a row. That's 1% high but well within the coil's tolerance. If after careful calibration you are still not getting measurements within about 5% of the nominal values for these coils I would talk with PE about a replacement hardware unit.

Regards

John