How to DIY?
Hi,
Like the O.P., I've also wondered this exact same thing, after seeing how commonly that configuration is used in commercial products. A long time ago, I found a formula/nomograph in a textbook for designing air-core inductors, and the coils I produced according to that turned out to have very accurate inductance as measured on a lab bridge (before the days of affordable digital meters!) Unfortunately, I have found no similar resource for designing iron-cored inductors. Does anyone have such info they could share or direct me to? Also, what exactly is meant by "air-gapped" core? Isn't the straight bar seen in the posted examples inherently air-gapped, or am I mis-understanding that term? Also, would it work to use a straight bar composed of transformer core laminations, or does it have to be the more commonly seen ferrite (pressed powder???) - I'm thinking DIY here, and don't have a ready source to salvage ferrite bars. Of course, I now have means to measure inductance readily, so the design calculations needn't be quite as precise (if, for example - the magnetic properties of the core material aren't exactly known.)
Thanks, Wilf.
Hi,
Like the O.P., I've also wondered this exact same thing, after seeing how commonly that configuration is used in commercial products. A long time ago, I found a formula/nomograph in a textbook for designing air-core inductors, and the coils I produced according to that turned out to have very accurate inductance as measured on a lab bridge (before the days of affordable digital meters!) Unfortunately, I have found no similar resource for designing iron-cored inductors. Does anyone have such info they could share or direct me to? Also, what exactly is meant by "air-gapped" core? Isn't the straight bar seen in the posted examples inherently air-gapped, or am I mis-understanding that term? Also, would it work to use a straight bar composed of transformer core laminations, or does it have to be the more commonly seen ferrite (pressed powder???) - I'm thinking DIY here, and don't have a ready source to salvage ferrite bars. Of course, I now have means to measure inductance readily, so the design calculations needn't be quite as precise (if, for example - the magnetic properties of the core material aren't exactly known.)
Thanks, Wilf.
You can measure it yourself 😉
https://www.youtube.com/watch?v=Z37962AcH0k
Alan Yates' Laboratory - Inductor Saturation Tester
in crossovers there is no DC, so amps rating are misleading, that's useful for DC bias applications which changes the BH operating point>
With audio AC signals the operating volts*seconds across the winding is what saturates inductors. looking at the BH curve tells us this. The shape of the BH curves can be altered to allow more volt*sec. this is done by gapping the magnetic path. to calculate the inductance simply obtain the Al for a particular core. this can provided by the OEM suppliers, or by measurement , or estimating knowing materials and dimensions.
Typical geometries useful for commercial chokes have been ferrite rods, ferrite bobbins, and iron laminations formed in bar stock. These shapes are seen as a magnetic path with a very large air gap, not to worry as the Al value takes this into account in the L calculation. Ferrites typically have less core losses for HF than iron which may be seen either as a plus or con in the final XO design. magnetics distributers typically don't deal with DIYers unless your gonna build 100s or 1000s.
Al of a suitable core can measured by knowing the inductance and the # turns an working the Eq. here Inductor Design with Magnetics Ferrite Cores
With audio AC signals the operating volts*seconds across the winding is what saturates inductors. looking at the BH curve tells us this. The shape of the BH curves can be altered to allow more volt*sec. this is done by gapping the magnetic path. to calculate the inductance simply obtain the Al for a particular core. this can provided by the OEM suppliers, or by measurement , or estimating knowing materials and dimensions.
Typical geometries useful for commercial chokes have been ferrite rods, ferrite bobbins, and iron laminations formed in bar stock. These shapes are seen as a magnetic path with a very large air gap, not to worry as the Al value takes this into account in the L calculation. Ferrites typically have less core losses for HF than iron which may be seen either as a plus or con in the final XO design. magnetics distributers typically don't deal with DIYers unless your gonna build 100s or 1000s.
Al of a suitable core can measured by knowing the inductance and the # turns an working the Eq. here Inductor Design with Magnetics Ferrite Cores
thinking about calculating saturation 1st consider volt*seconds. assume voltage is limited to some peak value from the amplifier, so seconds are related to the frequency. Lower frequencies are longer times at one cycle of the voltage waveform, Eg it's the area of one half cycle.
a lowpass using a series inductor drops little Volt*sec across it, why? its function is to pass low frequencies.
an inductor used in a high pass has little Volt*sec across it, why? a series cap limits it.
a lowpass using a series inductor drops little Volt*sec across it, why? its function is to pass low frequencies.
an inductor used in a high pass has little Volt*sec across it, why? a series cap limits it.
Last edited:
I beg to differ. Consider a woofer XO containing a series inductor. Frequencies much lower than fc will pass the filter with minimal change, and thus the full LF current will flow through the inductor. This causes an inductance change in a cored inductor, just like a DC bias does, only now, the change is temporary instead (as long as the current flows) of permanent. In this case, a strong LF current through the inductor will cause the filter parameters to be slightly modulated by the LF signal, and some distortion. So knowing the L versus I of any inductor is essential data for assessing its performance in a circuit, and sadly, I have not seen anything like that from the typical audio stuff manufacturers.
Specifying volts.second or current is equivalent for an ideal inductor with Rdc=0 by definition. With a real inductor, the DC resistance also determines how much current will ultimately flow due to an applied voltage, and this will always be less than with the ideal inductor. Because the current determines the magnetization of the core, a current rating makes more sense with any practical inductor, regardless if it sees AC or DC.
Last edited:
I thought we were talking about core saturation
AFIAK the DC ratings are related to the coils Rdc heating,
so the only inductance modulation is due to core material changing temp. ~ convince yourself its super easy to test.. get a DC supply , a good size current load, and inductance meter.
AFIAK the DC ratings are related to the coils Rdc heating,
so the only inductance modulation is due to core material changing temp. ~ convince yourself its super easy to test.. get a DC supply , a good size current load, and inductance meter.
We are. The B-H curve describes the core's response to a magnetizing field. The magnetizing field (H) is produced by the windings and is proportional to the current through the windings. It causes the magnetic domains in the core to align. This alignment is what causes the magnetic field (B). So the B is the response of the core to the H, the latter is proportional to the current through the windings. Now, when most of the available domains are already in line, there is not much left to align further when the H field (current through windings) is jacked up further. As a result, the core will no longer respond to an increase of H with an increase in B. This is saturation, and it causes the differential (= small signal) self-inductance at the bias point to collapse. It happens due to the momentary current through the windings, regardless of where it came from. So both AC and DC can saturate the core, it happens as soon as the current through the windings gets high enough. Saturation is not a point, rather, it is the smooth transition between the steep part of the B-H curve and the flat part of the B-H curve. When a manufacturer of inductors specifies a saturation current, it is usually the current at which the inductance has decreased to a percentage of its rated inductance.
Saturation of the core by a DC bias will manifest itself by a decrease of the measured small-signal self inductance. When the inductor saturates when an AC voltage is applied, the current will be distorted, and the differential self inductance will vary over the AC cycle. It is exactly the same process, only with the AC case the current through the windings, and thus the magnetization of the core, is not constant in time.
Now put yourself into the position of a coil manufacturer. The materials have a maximum temperature at which they operate. If the coil heats up to the maximum temperature while the core has not saturated yet, some headroom in the B-H curve goes to waste, and the inductor can be equipped with thicker windings in order to milk out the extra capabilities of the core and avoid wasting material (and thus $$$). On the other hand, if the coil is to operate in a high efficiency SMPS (by far the largest market for inductors), the inductor heating should be kept as low as possible (heat = wasted energy), and thus the Rdc needs to be kept low. The core will then saturate at a lower current than the current needed to overheat the windings.
So the manufacturer usually gives two current ratings, one at which the core will start to become saturated, and the maximum RMS current which will not cause the inductor to overheat. The latter is normally the largest for the reason stated above. See for example here:
http://www.bourns.com/data/global/pdfs/SRE6603.pdf
This is just elementary inductor data, but as I said earlier, audio stuff manufacturers typically don't even provide that. They just give an inductance, and the Rdc, but they tell you nothing about how the core behaves.
youre comparing oranges to apples
Ive designed DC/DC converters, usually we never care about Idc rating of inductors (core losses are greater than DC anyway), it's all about the stored energy in the gap. you cant compare small gapped inductors designed for tiny fly-back converters to what a XO inductors need to do. Idc/Isat rating just doesn't scale like what you imply. I'm saying test one find Isat> Idc
funny you picked tiny large value shielded inductors to try an make a point.
there is no such thing as "elementary inductor data", haha
all inductors are designed for a purpose you cant mix and match specs and end up with a universal current rating.
Last edited:
No, I'm comparing inductors to inductors. Regardless of the application, the physics that govern them are the same. AC current can cause an inductor core to go into saturation, just like DC current can. I tried to explain that. If you think otherwise, explain.
I picked the datasheet just to show that a datasheet of a common and inexpensive part is actually more detailed than the information given by the audio stuff makers. It was the first hit that Google gave me, so I didn't even have to search hard to make my point.
Explain why Bourns are wrong when writing up the datasheet I linked?
I agree that it depends on the design of the inductor, but if you manage to make resistive loss the limiting factor for using an inductor (regardless of application), you're wasting core material. This is also illustrated by the datasheet I linked. It serves as an example, not as a universal truth, in case you didn't get that. Explain why I am wrong.
When the inductor is designed properly, this is in support of the above. Although DC should really be called "resistive" here, because the AC current causes the windings to heat up just as well.
Explain why it is not possible to list essential parameters of an inductor in a table. While you're at it, also explain that to the magnetics manufacturers who do so all the time in their data sheets.
While you may not have noticed, we actually agree on that one. Audio inductors are usually only specified by their inductance value, their Rdc, and a power rating. Whats the latter? I think it is a nonsensical way to spec anything other than an air core inductor. And even then, the data are incomplete.
While sinking your teeth in the above issues, I recommend that you take a break every now and then to work on your attitude.
good thing they make boutique air cores, they have saved so many speaker engineers.
BTW they aren't hiding anything by not giving the huge current numbers you seem to think you need.
I can advise testing a few, would that ease your phobia.
BTW they aren't hiding anything by not giving the huge current numbers you seem to think you need.
I can advise testing a few, would that ease your phobia.
Last edited:
ever wonder why amplifiers don't spec max voltage or current either?
this is audio not engineering by catalogs
Is it so bad that I can't laugh at notions and conclusions some folks have round here
if I took them all so seriously, i'd end up just as crazy as them.
Last edited:
We're talking about components here. Surely the transistors and ICs inside them aren't specced for voltage and current, now are they?
I guess I should just stop wasting time feeding you.
If you have data showing a normal audio cored inductor becoming saturated in a reasonable design, Im all ears.
maybe you test a bunch, slap your own labels on em , and introduce a new brand.
maybe you test a bunch, slap your own labels on em , and introduce a new brand.
Last edited:
Ferrite materials exhibit Barkhausen noise.
This is audible in speaker inductors, speaker magnets and power filters.
Dan.
This is audible in speaker inductors, speaker magnets and power filters.
Dan.
got a ref? graphs, measurements? in practical, common audio applications
it is a physical effect but I haven't seen it shown to be a problem in real audio applications - some of our lowest audio noise interfaces are transformer based
I've looked and the absence of evidence make me think its a audiophile legend
it is a physical effect but I haven't seen it shown to be a problem in real audio applications - some of our lowest audio noise interfaces are transformer based
I've looked and the absence of evidence make me think its a audiophile legend
Audible only under controlled laboratory excitation signals i'd imagine. It is likely buried far into the system noise floor for any practical application.
Measurement test of cored inductors, showing saturation in a normal operating range:
Inhalt HOBBY HiFi 03/08
Inhalt HOBBY HiFi 04/08
Inhalt HOBBY HiFi 05/08
Inhalt HOBBY HiFi 06/08
Hi Sonce
do we have to buy something to get access?
did they test laminated steel cores ( what I use )
do we have to buy something to get access?
did they test laminated steel cores ( what I use )
Fantastic post, Dan!
https://en.wikipedia.org/wiki/Barkhausen_effect
All those little magnetic domains in the core creating clicks as they move around!
MIT Physics Demo - Barkhausen Effect | MIT Video
Sounds EXACTLY like a loose speaker connection to me. Order me some air coils immediately! 😀
I really don't know enough about this to know if AlNiCo or Laminated steel or ferrites are the best option if we must use cored coils. And speaker magnets.
- Status
- Not open for further replies.