sourcing audio inductors in 2022 (for equalizers)

[z11111]

It seems like sources for audio inductors have come and gone over the years. Some were readily available from Digikey / Mouser, some people were doing custom winds in Europe and elsewhere, etc... Personally, I have pulled inductors out of other equipment such as $10 consumer stereo graphic EQs.

The main reason I'm asking is because I ended up with a couple of Auditronics "Son of Grand" channel strips from a friend who has the full console and they are in various degrees of functionality and are missing several inductors for the EQ section. The values for this revision are 1.2 hy, 400 mh, 100mh, and 25 mh. Values for earlier revision of the same channel strip are 1.2 hy, 400 mh, 75 mh, and 20 mh. I actually just read through a pretty detailed old thread from 2005 here on the forum of people discussing the inductor EQ on this board.

I know a lot of people say to just get into winding your own, which is something I am obviously interested in but really don't have the time to figure out setting up a rig for that and figuring out where to source cores and what types of cores (unless it is easier than I think it is?). I know CAPI sells a couple of values but they are pretty expensive and aren't the values I need anyway.

 

[NOON]

I am NOT an expert on inductors at all but I know enough about them to know that they are one of the most imperfect components compared to the ideal model. They have significant resistance and capacitance which affects how they behave in an EQ circuit and they can get very non-linear with core saturation etc. This is all what makes them 'interesting' and why we still like them rather than just using a predictable gyrator opamp circuit. To get your EQ circuit to behave like the original you need to know a lot more about the inductor than just the Henries. Exactly what you need to know about it is beyond me and hopefully someone more qualified can chime in.

 

[z11111]

I am NOT an expert on inductors at all but I know enough about them to know that they are one of the most imperfect components compared to the ideal model. They have significant resistance and capacitance which affects how they behave in an EQ circuit and they can get very non-linear with core saturation etc. This is all what makes them 'interesting' and why we still like them rather than just using a predictable gyrator opamp circuit. To get your EQ circuit to behave like the original you need to know a lot more about the inductor than just the Henries. Exactly what you need to know about it is beyond me and hopefully someone more qualified can chime in.

Sure, makes sense to me. These inductors seem to be of similar style to what API and other pro audio companies were using back in the 1970s. These strips are from around 1978 based on capacitor date codes.

 

[john12ax7]

Cinemag sells multi tap inductors for a variety of EQ uses. Might be tough now with stock and lead times.

You could also look at AML which carries Carnhill, I recall they had common multi tap as well.

 

[z11111]

Cinemag sells multi tap inductors for a variety of EQ uses. Might be tough now with stock and lead times.

You could also look at AML which carries Carnhill, I recall they had common multi tap as well.

Yeah I’ve tried to contact Cinemag a few times this past year and they just don’t respond haha

 

[gyraf]

Get a P30/19 pot core in the nicelygood T38 material variant. This has A(L) at 28.000nH/sqrTurns and lovely misbehavior. Get a bobbin and a mounting-clip-plus-baseplate set.

Use calculator: Inductance Calculator – Power Magnetics

This one tells you that, given an A(L) of 28.000

25mH = 30 turns
100mH = 60 turns (the previous 30 plus 30 turns more)
400mH = 120 turns (the previous 60 plus 60 turns more)
1.2H = 207 turns (the previous 120 plus 87 turns more)

..in parenthesis if you do a multi-tap inductor

Wind by hand on the bobbin, whatever wire you have at hand. You can count in-head to 207, surely.

Then assemble with baseplate and mounting clip, rub the two halves together some to get good magnetic contact, and you're good. Remember that these un-gapped core materials come with relatively large tolerances on A(L), so you may want to measure A(L) first with known-number-of-turns then recalculate turn numbers accordingly if you need precision for stereo work etc..

/Jakob E.

 

[salomonander]

If you dont want to do this yourself, Sowter do custom inductors for a (somewhat) reasonable price. You can also try Don Audio for custom inductors.

 

[Bo Deadly]

This one tells you that, given an A(L) of 28.000

25mH = 30 turns
100mH = 60 turns (the previous 30 plus 30 turns more)
400mH = 120 turns (the previous 60 plus 60 turns more)
1.2H = 207 turns (the previous 120 plus 87 turns more)

Is this for real? I know zero about winding magnetics but I'm shocked at the thought of getting anywhere near the mH range with 30 turns.

 

[EthanSpaulding]

example of pot core part numbers that Gyraf mentioned:
https://www.digikey.com/short/p92bd3wd$26.70 total without wire.

The missing inductors are toroids. Those are not impossible to wind by hand with the help of a shuttle, or just brute patience. With the same formula's you can derive the turns. Measuring the geometry of the current toroids could maybe get you closer to a toroid core part number. Then affixed with zip ties, if I'm seeing correctly.

 

[gyraf]

Is this for real? I know zero about winding magnetics but I'm shocked at the thought of getting anywhere near the mH range with 30 turns.

It's real alright - it's just modern hi-perm materials when done right. A quick test here, 30 turns and no force-clip, shows 23.2mH, pic related. As you can see from Ethan's link above, it's not the most inexpensive material - but it sure works nice, i.e. sounds nice at boundary conditions...

/Jakob E.

 

[EthanSpaulding]

The mounting hardware for the P30/19 is as expensive as the pot core.
Here is a link to a full setup in RM8 T38. 12.5uH AI. Total price $6.57 for one set.

(notice in above posted cart, I put in for 10 bobbins, which is handy to have extra of)

Curious to any thoughts on RM8 vs a P30/19 core.

https://www.digikey.com/short/r3ppr838

 

[gyraf]

Agree, RM8/T38 is also worth looking at - but mainly for high'ish frequencies - it's tiny, so it saturates much earlier than the P30/19. For anything resonating above say 800Hz, it's great. Alternatively, run it so that absolute levels across the inductor is reduced to an amount the material will handle..

/Jakob E.

 

[thomasdf]

Anybody tried winding toroid MPP core ? They're pretty cheap in the US or AliBaba (if you buy hundreds) , but I can't find them in EU ... ? Seems a bit more difficult to wind, but for Pultec style EQs it seems more period (and tone) correct.

 

[gyraf]

can't decode "MPP"?

 

[thomasdf]

Sorry ! It's Mollypermalloy Powder
https://en.wikipedia.org/wiki/Molypermalloy_powder_core
I found this : MS-141060-2

AL = 56, seems like a lot ? edit : erf, maybe not.
I'm surprised that the size of the wire does not matter ?

 

[nielsk]

It is very easy to wind an inductor to come up with a desired value, and very difficult to wind an inductor that sounds great. There are some EQ designs (Neve 1073 for example, once again I wonder what one would sound like with great inductors) that sound acceptable with a mediocre inductor, others (like the Auditronics) that REALLY need excellent sounding units.
The obvious example is any of the Pultec EQs, they are tons of them that sound anywhere from barely OK to absolutely bad (some from the most famous companies), and just a tiny handful that sound great. The only real difference is the inductors. Too bad most folks have never got to use a good one, and are under the impression that a crap one is "just what they sound like".
Just take one of the bad inductors apart and you will get an idea of what is involved in winding one:
About 100' of hair thin wire wound perfectly flat and even around a tiny doughnut shaped core, half way up then all they way round then back then all the way down, to make many, many layers.
I wish I knew of a source of singe quantity high quality audio pye wound mollypermalloy torroidal inductors, they only way I know of to obtain them is to have quantity made, and that involves a hefty set-up fee.
Every once in a while you can find old stock Triad units, usually potted in a red compound, and possibly find a catalog entry that notes what it actually is.
Other than that it means buying broken EQs that have a good inductor of the value you need.

 

[thomasdf]

Totally agreed, I've heard a lot of crappy Pultec clones. Many more than good sounding ones, actually.

While I've never heard them myself, I've had lots of great feedback about the Cinemag inductors as well as Vintage Windings.

I wonder how they wind the old ones in various directions ? Was it automated, or done by hand ?

edit : some interesting reading : http://vintagewindings.com/tech swag/Pultec-Facts-&-Fiction.pdf keep in mind the guy sells inductors, so it may be a bit biased

Wind by hand on the bobbin, whatever wire you have at hand. You can count in-head to 207, surely.

/Jakob E.

A few questions come to mind :
-what's the role of the wire size, except varying diamter = varying length for the same result = varying R = varying Q in the EQ circuit?
- in the case of multi tap, do the inductor need more and more turns when the wire gets further and further away from the core ?

 

[abbey road d enfer]

Totally agreed, I've heard a lot of crappy Pultec clones. Many more than good sounding ones, actually.

The most common mistake is to use cores with a high AL. Typically they dont have a gap and saturate early. In addition they have a very large tolerance range.
Using AL630 and AL1k requires many turns for the higher nominal values, but they produce good results.

-what's the role of the wire size, except varying diamter = varying length for the same result = varying R = varying Q in the EQ circuit?

Most EQ's don't require a very high Q; most EQ's resort to adding series resistor to tame Q. Of course it takes some calculations to get the DCR right but it's not too difficult. Typically one designs for a slightly lower DCR than needed and compensate with an actual resistor.

- in the case of multi tap, do the inductor need more and more turns when the wire gets further and further away from the core ?

No. The turn number does not change much with distance from the core. Actually the core is both inside and outside the winding in RM and P cores.
The typical final adjustment consists in rubbing core halves against each other. You can adjust about +/-2%.

 

[thomasdf]

Super enlightening, thanks !! What about toroid cores ? No halves and wound thickness must have some role to play ?

 

[abbey road d enfer]

Super enlightening, thanks !! What about toroid cores ? No halves and wound thickness must have some role to play ?

Being gapless, toroid cores saturate in a more abrupt way, but since they are usually bigger, they are operating at lesser induction.
OTOH, since the winding is continuous not too thick, the coupling factor is extremely good; 99.98% of the flux is concentrated in the core. It is quite important for transformers, not so much for inductors.