Help measuring leakage inductance please.

Help Support GroupDIY:

Joined
Apr 17, 2021
Messages
5
Hi Guys, first post here, but I have been reading for a while. I have recently started winding my own transformers and it has been a steep learning curve.

Right now I have 4 Fender style deluxe output transformers on my desk, I have wound 2 and 2 are purchased.

They all have the same stack and are m6 lams

two have 2 sections
one has 3 sections
one has 4 sections

Primary inductances and Primary resistances of all three are fairly close

When measuring leakage inductance I have found the leakage inductance increases with more layers, from everything I have read the leakage inductance should reduce with more layers.

Have I misunderstood something? Secondaries are shorted, measuring at 1kH with my Philips Fluke PM 6303

TIA
 

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
13,532
Location
Norfolk - UK
Leakage inductance is basically a measure of how much flux from one winding fails to couple to the other winding. Coupling will be better if the two windings are very very close together. If you wind them separately , one on top of the other, quite a lot of each winding is not close to the other. If you interleave the windings i.e wind half the primary, then half the secondary then the other half of the primary followed by the other half of the secondary then they will be closer to each other. The more you interleave like this the closer they will be to each other and the lower will be the leakage inductance.

So my first question is are you actually interleaving the windings?

Cheers

Ian
 
Joined
Apr 17, 2021
Messages
5
Hi Ian, thanks for the reply.

the units with the lowest measured Leakage are the ones without interleaving, they are wound primary/secondary

The 3 layer one is wound primary/secondary/primary

The 4 layer one is wound secondary/primary/secondary/primary and has the highest Leakage.

I must be doing something wrong on the measurement side as the results are the direct opposite of what one would expect. I will go read the manual again.
 
Joined
Apr 17, 2021
Messages
5
Ok, so I just used the autoranging function on the meter, If I set it to Ls then I get numbers that make more sense relative to the interleaving.

That make sense to me since it is in series with the winding resistance, correct?
 

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
13,532
Location
Norfolk - UK
Ok, so I just used the autoranging function on the meter, If I set it to Ls then I get numbers that make more sense relative to the interleaving.

That make sense to me since it is in series with the winding resistance, correct?
What frequency are you measuring at?

Cheers

Ian
 

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
13,532
Location
Norfolk - UK
I have often found 1KHz inductance readings unreliable except for ferrite cores. So I routinely measure at 100Hz these days.

Cheers

Ian
 
Joined
Apr 17, 2021
Messages
5
Thanks, Think I need another meter, but I am currently only using this to compare different trafo's winding methods on similar units
 

MisterCMRR

Well-known member
Joined
Jul 1, 2019
Messages
93
Location
Ventura, CA
Since leakage inductances are small (at least hopefully, they should be only 1% or so of the open circuit inductance), I'd suggest 10 kHz (instruments in general have better accuracy when the test circuit impedance is neither extremely low or high). Also, leakage inductance will generally affect transformer performance only at high frequencies, so this is where it matters (and permeability of the core may be quite different at 10 kHz compared to 1 kHz or 100 Hz). Just my 2 cents worth!
 

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
13,532
Location
Norfolk - UK
Do you know of a readily available instrument that will meaus
Since leakage inductances are small (at least hopefully, they should be only 1% or so of the open circuit inductance), I'd suggest 10 kHz (instruments in general have better accuracy when the test circuit impedance is neither extremely low or high). Also, leakage inductance will generally affect transformer performance only at high frequencies, so this is where it matters (and permeability of the core may be quite different at 10 kHz compared to 1 kHz or 100 Hz). Just my 2 cents worth!
Do you know of a readily available not too expensive instrument that will measure at 10KHz? I have an Agilent U1731C that measure at 100Hz, 120H and 1KHz only and that was not cheap.

Cheers

ian
 

MisterCMRR

Well-known member
Joined
Jul 1, 2019
Messages
93
Location
Ventura, CA
I use a Reed Instruments model R5001. See REED R5001 LCR Meter. It costs around $250 and is available from Digi-Key and others. It's a pretty amazing instrument and measures most parameters at 100/120 Hz, 1 kHz, 10 kHz and 100 kHz with an accuracy better than 2%. I like it a lot ...
 

MisterCMRR

Well-known member
Joined
Jul 1, 2019
Messages
93
Location
Ventura, CA
For those working with transformers, it's wise to remember that the open-circuit inductance of a winding or Lpri dominates low-frequency performance and leakage inductance (secondaries shorted) Llkg dominates high-frequency performance. The former should be measured at 100 Hz or lower and the latter at 10 kHz or higher. Also remember that the permeability of most core materials (especially steel) varies considerably with level.
 

Brian Roth

Well-known member
Joined
Aug 20, 2005
Messages
1,743
Location
Salina Kansas
I use a Reed Instruments model R5001. See REED R5001 LCR Meter. It costs around $250 and is available from Digi-Key and others. It's a pretty amazing instrument and measures most parameters at 100/120 Hz, 1 kHz, 10 kHz and 100 kHz with an accuracy better than 2%. I like it a lot ...
Hello, Bill. Reed Instruments is a new one for me. I see they make a wide variety of meters and seem to be based in the USA. In addition to a LCR meter, I've also been shopping for a new "clamp-on" meter for AC/DC current measurements. Interesting to see a Fluke alternative!

Bri
 

MisterCMRR

Well-known member
Joined
Jul 1, 2019
Messages
93
Location
Ventura, CA
If you do much audio system noise troubleshooting, I'd recommend a sensitive clamp-on ammeter. Ground loop currents only need to be a mA or so to cause noise in unbalanced interconnections. These meters are usually called "Leakage Testers" because they'll measure the UL-limited leakage currents of 0.75 mA (for two-prong) and 3.5 mA (for 3 prong). But they're really just clamp-on ammeters with a sensitive low range. This one has full-scale ranges of 200 mA, 2 A, and 200 A with a 2000-count display, so it can resolve 0.1 mA on its sensitive range. It's the Multicomp-Pro model MP790050 and it sells for about $55. I don't own one - mine is an older Hioki model that cost me about $275, but it has nice features like a signal output port, etc. that's useful for research projects.
 

rackmonkey

Well-known member
Joined
Sep 20, 2016
Messages
905
Location
Texas
Another LCR meter option is the popular DER EE-5000. At around $120, it’s been a solid performer for me for several years now. The accuracy claims are impressive, but I tend to be a little skeptical of them at this price point. However, measurements seem to jibe with my old Agilent and Leader tools. Same measurement frequencies as the meter Bill mentions.

Just another option to look into.

DE-5000 LCR Meter
 

Latest posts

Top