14
\$\begingroup\$

I am building a tube amplifier, and have decided to add a 5ch EQ to it. Historically, the ones used in guitar amplifiers were all passive RLC filter style systems, with large inductances of 0.5 - 2H for the 80Hz channel.

I know now days it would be smaller to use an active op-amp style, but I am doing this as a hobby and want to try making a passive one. I was slightly surprised to find Digikey does not carry any small inductors over ~100mH. My guess is this is because no one uses them for small current applications anymore with the advent of DSP or active capacitor based filters.

Any advice for creating large 1H inductors with <1mA of current that does not involve winding 300 turns through a 1" torroid?

Or does anyone know how they were historically made, really fine wire i would guess with many turns?

img

\$\endgroup\$
  • \$\begingroup\$ I have lots of these old inductors that I scarfed up as a kid. They are very much as Tim describes. Basically built like a transformer but with thinner steel sheets stacked up. Heavy. Some I have weigh in at about a kg, or so. I also have the ones used for matching up the plate impedance with the speaker impedance for driving speakers. Also heavy, though usually around half a kg, or so. I donated boxes (and I'm talking about 4'x4'x3' boxes here) of vacuum tubes I'd also collected to some radio clubs -- about 15 years ago. Hopefully, put to good use. \$\endgroup\$ – jonk Sep 5 at 6:51
  • \$\begingroup\$ You can just buy them at parts-express.com/cat/solid-core-inductor-crossover-coils/299 \$\endgroup\$ – Hilmar Sep 5 at 11:06
  • \$\begingroup\$ Hilmar, I need Henrys, not milliHenrys, so I don't think crossover inductors will work here. Good try though. \$\endgroup\$ – MadHatter Sep 5 at 11:42
  • \$\begingroup\$ digikey.com/product-detail/en/triad-magnetics/C-24X/237-1777-ND/… ? \$\endgroup\$ – kkm Sep 5 at 15:14
  • \$\begingroup\$ If you do end up winding your own and it's not toroidal, a coil winder isn't too hard to make. In a previous job we used a mixture of wood to take the weight, lego technic for the motion, and a thin steel rod for wire final guide. That was an electromagnet with a rather small gap and a couple of thousand turns for 1A; many core shapes would be easier \$\endgroup\$ – Chris H Sep 6 at 9:06
12
\$\begingroup\$

The primary of a smallish power transformer is on the order of 1 H.

Magnetizing current for such a transformer is on the order of 0.35 A, which means that its inductive impedance is

$$Z = \frac{V}{I} = \frac{120\ \mathrm V}{0.35\ \mathrm A} = 343\ \mathrm{\Omega}$$

This means that the inductance must be

$$L = \frac{Z}{2\pi f} = \frac{343\ \mathrm{\Omega}}{6.2832 \cdot 60\ \mathrm{Hz}} = 0.909\ \mathrm H$$

Inductors in that range, for low frequencies, use the same construction techniques as transformers, but with just the one winding.

\$\endgroup\$
  • 1
    \$\begingroup\$ I did not consider steel laminate. Would that also have a higher effective inductance per turn? Still probably bigger then a toroid with a ton of turns, but worth considering. \$\endgroup\$ – MadHatter Sep 5 at 4:29
  • \$\begingroup\$ Yes, much higher than ferrite. \$\endgroup\$ – Dave Tweed Sep 5 at 10:59
  • \$\begingroup\$ @MadHatter, compare a PC 300W switching supply has an 1" toroid ferrite transformer working at 50kHz; a similar power laminate 60Hz transformer from an old color TV has a 5lbs steel core. Higher frequency = farther away from flux saturation = smaller core. But the laminate quickly incurs eddy losses as frequency goes up; ferrite does not. The permeability of ferrites is much lower however: typical 5 mH/m steel, 0.08 mH/m MnZn ferrite, 60 times worse. Try using ferrite in that old TV transformer, and it would weigh 300lbs. :) At your design frequency of 80Hz, steel is the clear winner! \$\endgroup\$ – kkm Sep 5 at 15:07
  • \$\begingroup\$ Smallish? 0.35A ×120V =42VAR , say it's 20% of full power (really poor figure), what comes out is 200VA or approx 3kg of steel \$\endgroup\$ – carloc Sep 5 at 18:45
  • \$\begingroup\$ Instead of cannibalizing a transformer you can just buy an inductor from DigiKey. \$\endgroup\$ – J... Sep 6 at 15:21
6
\$\begingroup\$

They were wound like transformers, as Dave Tweed mentioned. But they were of specific construction that is different from typical power transformers: the laminations were thinner, so they would be low loss at audio frequencies, and they may well have been gapped for linearity.

It may be worth it to look at the audio transformers available for tube amps (such as the ones that Hammond makes, available from www.tubesandmore.com), and either just use them as-is, or rewind them.

It may also be worth it to investigate using the biggest E cores that you can find, and possibly even gapping them with Kapton tape or similar (to do it really right you get the inner leg precision ground, but I'm assuming this is a hobby project, not for production).

\$\endgroup\$
  • \$\begingroup\$ If you go with E cores you may want to search on "gapping E cores for prototypes" or something similar -- I'm guessing that Kapton tape would do it, but someone out there undoubtedly knows. \$\endgroup\$ – TimWescott Sep 5 at 14:57
4
\$\begingroup\$

Pot cores. The winding was on a simple bobbin - much easier than a toroid - and it is likely to be nearer 3000 than 300 turns (of very fine wire).

The bobbin is then fitted between two ferrite cores which are tightly clamped to virtually eliminate the air gap. Some variants had a moveable slug for fine tuning over a few percent, rather like a radio IF transformer.

Signal levels had to be strictly controlled to limit harmonic distortion as the core started to saturate. (As with today's switching supply cores, there were different ferrites with different characteristics, allowing e.g. lower distortion if you didn't need the highest values of specific inductance).

Useful search term : Vinkor was one of the common makes. I may dig out some datasheets later on...

\$\endgroup\$
2
\$\begingroup\$

There is a lot of materials available for making custom transformers/Inductors. When I discovered this I felt quite liberated because I could finally just make what I needed.

At university we had an old machine to wound up these transformers. You mainly just need something that rotates and counts.

I understand why you do not want to custom wind toroids. It is a pain.

Specifically you have to go under the categories:

  • Coil formers (An easy-to-wind (still hard to count though) plastic assembly
  • Ferrite Cores (There are a lot of different ones, but for low frequency you are not too picky i believe. These will be secured to the coil former with clips)
  • Magnetic wire. (Chosen based on average current)
\$\endgroup\$
  • \$\begingroup\$ Thanks Bonnevie, I want flybacks etc on ferrite cores so I'm pretty familiar with transformer making... The issue is if you use ferrite without going to a massive core you need a lot of turns, I think is Dave pointed out, for this low frequency steel has a much better Al value. \$\endgroup\$ – MadHatter Sep 5 at 11:45
  • \$\begingroup\$ Ah yes, I thought there would be adequate ferrite cores. It is quite difficult to find solid steel cores compatible with coil former standards. I am guessing that for laminate steel cores it won't be possible to line up the 2 pieces so no air gap is created.. \$\endgroup\$ – Bonnevie Sep 6 at 5:44
1
\$\begingroup\$

"Historically" perhaps isn't the example you want. Historically, amplifiers had levels of distortion, noise and mains hum which we wouldn't accept today from the cheapest kid's radio. Not only that, all electrical/electronic devices broadcast significant electrical noise which would be picked up by other equipment.

Winding toroids is never fun. However only toroids can ensure your high inductance isn't receiving or transmitting noise. This really isn't somewhere you want to take shortcuts.

\$\endgroup\$
  • 2
    \$\begingroup\$ This is for a guitar amp, non-linearity, second and third harmonics, and a little bit of buzz is exactly what I want. \$\endgroup\$ – MadHatter Sep 6 at 11:30
  • \$\begingroup\$ I'm surprised to read this answer from somebody with a Strat in his profile picture... \$\endgroup\$ – leftaroundabout Sep 6 at 14:40
  • \$\begingroup\$ @MadHatter The guitar part is important then. Hi-fi nuts are also into tube amps, with a very different aim in mind. Still, don't discount the problems with mains hum and stray noise pickup. It might not sound too bad on its own, but try recording or gigging with a mic in front, and you can have real problems. As a soundman, I've had to work with a few guitarists with noisy amps, and to a man they all said they didn't realise how bad it was until they got on stage, which isn't a good time to find your instrument sounds crap. :/ \$\endgroup\$ – Graham Sep 6 at 14:52
  • \$\begingroup\$ @leftaroundabout I like Strats, but I'm under no misapprehension that the electronics on old kit was ever any good. Sometimes it went wrong in aurally-interesting ways, but mostly it just went wrong. Proper screening and earthing techniques and buffer preamps are very much good things. \$\endgroup\$ – Graham Sep 6 at 14:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.