Basics of Magnetic Amplifiers

[@TheTrueCBaer]

"Like with any Circuit there is a big difference between the proof of concept and the useful high performance implementation" yeah tell that my boss

[@gerarddugas6224]

Your insightful

[@TheAnthroheart]

Yup

[@gregkocher5352]

I graduated in 1979 with an Electrical Engineering degree. We had exactly 1 hour of vacuum tube review, lol. I was fairly well prepared in basics to work on Mag Amp controls in steel mills. I had to pick up control principles on the job unfortunately. I was privileged to see the evolution as mag amps were replaced by analog electronic drives, then digital drives and master computer controls were added as the overall system manager. And PLCs rapidly replaced the extensive relay panels. I think it was a great time to be an EE. As I worked in old mills there were many mixed mag amp/plc/digital drives/and master computers that were still in use to 2006. I got pretty good at digging up old vintage parts and figuring replacement parts. The upgrades we installed were over decades. It was challenging and satisfying. I even had cranes designed in 1890s lol! Mag Amp mfgrs like GE, Westinghouse, Allis Chalmers, Jocelyn Clark, and more. Your article is very interesting! PS, it wasn't just power, it was used for regulating dc motor speed, torque, and tension loops. I toured a WW2 submarine mueeum and the mag amps were clearly present.

[@gerarddugas6224]

Sorry you had such high expectations of college. TEST is real life. You passed with money as the reward. I did something parallel in expectations and experience. CHEERS to you!🎉🎉🎉

[@anonymous.youtuber]

Some forty years ago, I encountered a magnetic amplifier in an industrial CO2 laser used for metal sheet laser cutting. It was used to regulate a high voltage supply that powered the electrodes on both sides of the laser tube.

[@r.kellycoker1981]

Fifty years ago I was introduced to magamps while in the Navy. They were mechanically shock proof and used in Nuclear reactor safety circuitry. They never gave any problems , unlike the transistorized circuits. Thank you for your informative video and bringing back memories.

[@louco2]

Only you to present such a span of topics without missing a beat, Tank you Fez!

[@Scrogan]

I’ve seen these in schematics, not intended for power regulation or amplification, but rather as an electrically variable inductor for passive LC filtration. There’s no other (non-electromechanical) method of varying inductance with an electrical signal, and varactors don’t have as large a range and are probably more temperature sensitive. I can see this being especially useful for RF circuits, where the required coil size is much smaller, and you don’t have the ability to make active filters and virtual inductors without expensive high-end semiconductors. Not just for VCAs and VCFs, but for VCOs too. I don’t know how linear they are compared to JFETs for amplifiers and active filters, but you can always use an extra winding for feedback.

[@uwezimmermann5427]

I have not come across a magnetic amplifier myself - as far as I can remember - but I first learned about them with regard to old radio transmitters from about 100 years ago. It seemed to be the only feasible way to control the amplitude modulation at kW power levels back then.

[@operatorenabla8398]

woah man, I was looking for some understandable thorough explanation of mag-amps for weeks and best I could find was a 50s or even older thesis... thanks for the vid!!!!!!!

[@willthecat3861]

Look better. There is a lot on magnetic amplifiers.

[@paradiselost9946]

try the other name... "saturable reactor" which is far more self explanatory than "magnetic amplifier"...

[@macspillers4731]

I hadnt seen nonlinear magnetic saturation used for amplification, but one of the coolest things Ive seen is M0BMU's electromechanical VLF receiver that uses magnetic saturation to make an rf mixer for listening to SAQ at ~17 kHz

[@olivierconet7995]

I finally understand 💡how this old PC-AT power supply was regulating the secondary voltage outputs. Thanks a lot !!

[@jonathanjohnston716]

Industrial/Commercial welders used the magnetic amplifier into the 1980's - my personal experience was troubleshooting and repair of the Miller 330 AP-B/Goldstar and Aircrafter 330. The former controlled the DC bias current via a 200W rheostat, the latter through a triac. Heavy, reliable machines!

[@Stelios.Posantzis]

Great video and I hope there is a sequel on magnetic audio amplifiers, preferably sans crossover distortion. Magnetic amplifiers are preferred in cases where reliability, robustness, longevity and efficiency matter much more than cost of building, volume and weight of the device.

[@m.e.8273]

An incredible book around this topic, if you want to see a lot of interesting topologies and design your own transformers suitable for magnetic amplifiers, is the book "Wzmacniacze Magnetyczne" written by Gabler, Haškovec, and Tománek!!! It can be found in most Warsaw-pact languages even, I have a Polish translation. It is from the time when active research was still done in the subject, so you'll find a lot of alternative information and topologies that may not have been published on the internet yet. I really recommend anyone to get it if you're interested diving into the material :-)

[@PENelektronik]

In Polish radars, in the 80s/90s, in modulator systems for controlling the magnetron, transformers with a so-called "rectangular hysteresis loop" were used. The core of the pulse transformer was "under-magnetized" by an additional winding with regulated current. This has nothing to do with signal amplification, but there is an analogous shift in the core hysteresis characteristic.

[@briankirk4685]

A mag amp is a “static variable inductance in series with a load”. This was the description in an instrumentation manual. I saw two basic uses for these. In instrumentation to amplify DC signals. When the instrumentation was designed the choice was tubes or mag amps. Tube type DC amplifiers are a challenge and much less reliable than mag amps. The other use was in static exciters to provide up to 50 amps to the field winding of a turbine generator. The alternative was an amplidyne In addition to the control and gate windings, most mag amps also have a feedback winding and a bias winding. Adjust feedback and bias to set zero and span

[@МихаилМихаил-б6я]

Магнитные усилители широко применялись в системах Г-Д для управления электроприводами. В электрических экскаваторах. Там применялись двухтактные дифференциальные преобразоватетипа ПДД 1,5 на экскаваторах ЭКГ-4,6, ЭКГ-5. И ПДД 1,2 на экскаваторах Э - 25 серии Э-2503, Э-2505, Э-2508, Э-2508СА. И на металлообрабатывающих станках для управления электроприводами. Были еще и злектромашинные усилители типа ЭМУ которые выполняли те же функции.

[@paradiselost9946]

people always seem amazed by these things... i find it more amazing how much technology has been left by the wayside... that our education system lacks in outlining the basics... the name "saturable reactor" is a far more... self explanatory name. you just apply a dc magnetisation current to "saturate" the core, at which point theres no lenz reactions or "inductive reaction" to an AC current. the coil acts like a plain wire with only DC resistance. another closely related device is the "peaking transformer"... using core materials of two different saturation curves, to produce pulses from a sinusoidal or other waveform. it pays to collect as many old books as possible... the ignitron, kenotron, thyratron, and pliotron may be obsolete, but often, those books that deal with them break down many otherwise complex subjects into simple and easily understood basics... which then makes it far easier to digest other seemingly unrelated topics... we got where we are by riding on the shoulders of extremely intelligent men. its a bit like opening an old desk fan the other day... oh my, it has an inductor. which were also common in ceiling mount speed controls. early on, it was cheaper and more reliable than any capacitor of the time. but now its always a capacitor used as they are so cheaply made, and the dielectrics so much better. ptfe and polyethylene versus waxed paper, etc...

[@yveslesage8525]

Magnetic amplifiers were used in the french nuclear submarines (called SNLE in France) for amplifying the current delivered by power nuclear sensors because of their reliabilty. There was a feedback scheme for improving the linearity.

[@6WTF_MAN9]

Круто! Так редко про магнитные усилители вспоминают))) Молодец что такой материал подготовил. Я к стати использую магнитный усилитель для управления двигателем вентиляции, тиристорный слишком шумный, магнитный работает достаточно мягко.

[@johnopalko5223]

The Fred A. Leuchter Associates electric chair design used a saturable reactor to regulate the output current to 5 amperes. The instruction manual described it as a magnetic amplifier. There was a 6A circuit breaker "to protect the load" which I always thought was a particularly eerie turn of phrase.

[@danstiurca7963]

This is a VERY specific piece of technical information, sir. One does hope none of the loads were...accidentally damaged.

[@bpark10001]

I built a magnetic amplifier using a pair of 1 amp Variac cores. Each was wound with half the number of turns that the 120V original winding had. Cores were stacked & 10,000 turn control winding was wound through both. I used this to control the current to 8 fluorescent lamps on wall of discotheque, which were modulated to the music. It is important that the 2 cores be matched & the number of turns on the power windings match to the turn, or AC will back up into the control circuit.

[@ricksampson6780]

I believe the term "magnetic amplifier" is somewhat ambiguous. Rather, a "saturable reactor" is more definitive, cheers.

[@paradiselost9946]

yes. as it actually explains what they do... "saturate" a "reactor"... and its the name i know them as. they have a cousin, the peaking transformer... another "obsolete" bit of tech... best counterpart i can think of is a schmitt trigger.

[@ricksampson6780]

@@paradiselost9946 Yes, I agree with your comments.

[@willthecat3861]

The U.S. Navy, in the mid 1950's, at least... used magnetic amplifiers to very successfully replace syncho systems that had previously used control transformers, and electronic amplifiers. There were many benefits over the purely 'electronic' systems of the day. Like being lighter, and more robust in harsh environments shipboard. It provided syncho isolation, and very fast, and stable, transient response. Remember, this was in the vacuum tube, zinc plated, world of the 50's

[@johnathancorgan3994]

The US submarine I served aboard in the 80s was built in the 70s and had mag amps at the heart of many of the nuclear plant control systems.

[@davidprocopio9021]

@@johnathancorgan3994 RC Div, USS Truxtun 83-87. We had them all over, but haven't seen one since!

[@kenwallace6493]

Mag amps were used extensively in Navy ships to control gun turrets. When they became unstable, the triple 9" or 12" guns could shake the whole ship and send the fire control tech running for the anti-hunt pots.

[@sergepetrov8598]

What an example! Control theory students, if exposed to that, would get a big chunk of motivation to study.

[@power-max]

12:27 I think it is still commonly used in cheaper and lower power ATX power supplies for PCs. It generally goes by the name of 'group regulation' although I could be mistaken, it is possible that in such cheap PSUs there is no form of regulation of all auxilary rails and reliant on tight enough coupling and low enough resistance of components to ensure stable enough voltage under most load conditions that the rails remain within tolerance.

[@sergepetrov8598]

AFAIK there is no saturation in group regulation inductor.

[@billbynum2210]

Yes in a power supply for a backup battery charger! A very old design for an older aircraft.

[@maxheadrom3088]

That's how old telephones worked, isn't it? Thanks! Nice channel, btw!

[@mineown1861]

An informative video with follow-up reading , you are too kind . An interesting video as always , thank you .

[@danstiurca7963]

Thanks, fez. Given the historical uses of these amplifiers, I'm now probably on multiple 3-letter agencies watchlists for trying to learn more. That TI slup129 App Note is pretty insane, btw, it still doesn't make sense after multiple readings.

[@ic7481]

I have used equipment with saturable reactors (which I guess are magnetic amplifiers) , but this is for very high power (+500kW switching), used to vary inductance very quickly.

[@ChiefBridgeFuser]

Mag amps were popular in nuclear submarines for control because of the extreme reliability. Not sure if newer designs still use them.

[@johnathancorgan3994]

Yup--it was hard to imagine the kind of damage we'd have to take that would knock those out of commission.

[@aduedc]

You may be able to use Variable Inductor (VarInductor), much like Variable Capacitor (Varactor), to tune VCOs and filters. Specially for high power applications.

[@paulcohen1555]

MANY PC power supplies had a magnetic amplifier on the 3.3V rail, with a core made of a special material.

[@sergepetrov8598]

Mine had a core consisting of two same size rings: one some kind of ferrite, and another one seemingly just plastic. It was widely used in ATX 3V3 rail before transitioning to separate buck converters from 12V rail, and a small TO92 transistor controlled several Amperes of output, deriving it from 5V winding.

[@paulcohen1555]

@@sergepetrov8598 The special core material (With a square magnetic loop) is VERY BRITTLE and that's why it's enclosed in a protective plastic case.

[@gary.richardson]

​@@paulcohen1555was that brittle material Beryllium?

[@marimarmarimar25]

PS All right I've seen your page and the other comments, now you put here the basics, it's ok and I realy like all.

[@HuyenTran-pk5yr]

If you haven't seen a Mag Amp in action, take apart an ATX pc power supply, it's a 3v3 coil

[@xDR1TeK]

Interesting work. Thank you.

[@peterwathen3463]

Magnetic Amplifiers were used in Aircraft instruments (Auto Pilots) Smiths SEP 2 , still flying in RAAF older aircraft in the late 80s when I worked as an Aircraft instrument maker. They were bullet proof. Pun intended.

[@bobi_lopataru]

I see this having a really nice use in a homemade variable power supply for vacuum tubes. The one you made is perfectly good as-is, but why not make it more efficient by using such a mag-amp instead of an ordinary linear regulator? I'll try this out sometime!

[@marimarmarimar25]

I expected you show us an audio amplif working with only transformers and control windings in cc. So, when it wil be the next episode? Be engineer, not only teacher! ;) Good luck!

[@fmk326]

Bueno esto ya hay quien lo ha hecho y es posible, pero yo llevo tiempo intentando hacerlo, un amplificador de audio magnético y aún no lo he conseguido, Puedes darnos aquí el creador del video u otra persona con esos conocimientos, los detalles necesarios para hacer un amplificador magnético útil?, Aunque sea de baja potencia, 1watio en 8 ohmios? O 5watios en 8 ohmios?, No se bien como se modula la señal de audio , supongo que se añade a la entrada el audio en el circuito de control, pero estoy confundido porque una señal de audio es alterna AC y el circuito de control es DC, entonces por donde se inserta el audio por el circuito de carga AC?, Alguin que de un poco de luz respecto al amplificador magnético de audio o un esquema fijado o funcional detallado?

[@WECB640]

Really excellent content. Keep up the good work OM. 73

[@TriunfoGim]

12:00 yes, it is still used in PC's power supply, aka cross regulators (is the big core inductor plenty of wires with almost all output voltages at the output side of the P.S.U.

[@colinstamp9053]

I think these were (at one time at-least) commonly used to dim airport runway lights.

[@paulcohen1555]

@@colinstamp9053 And because the bulbs were connected in series, the magnetic controller was used to BYPASS burnt bulbs.

[@sergepetrov8598]

@@paulcohen1555 Could you please elaborate on bypassing?

[@paulcohen1555]

@@sergepetrov8598 When the bulb on the secondary burnt out, the transformer core went quickly into saturation and presented low impedance (resistance) to the loop.

[@sergepetrov8598]

@@paulcohen1555 So my imagination draws each lamp having personal transformer, with all primaries in a common loop. How did dimming work then? BTW techie replacing bulbs is protected from high loop voltage by design. No need to turn lights off in nasty weather.

[@operatorenabla8398]

noob question: could crossover distortion be avoided by applying a low DC bias to the AC side of each transformer? Of course with bypass capacitors (could we still call them "coupling" here?) to block it from getting where we don't want it. Also an unrelated question: how to I check the saturation characteristics of a certain transformer? The datasheets I find online don't provide them. Should I look for the datasheets of inductors with the same cores like you suggested? For example I have some EI14 1:1 isolation transformer in transit and would love to try out a mag amp. I wanna build it for a unique kind of guitar distortion effect.

[@FesZElectronics]

Usually, the saturation behavior will not be described in the transformer datasheet, but rather the core datasheet - the exact core material and geometry is what influences this; I am actually working on a couple of videos on this topic, hopefully they will be ready in the not so distant future

[@alexloktionoff6833]

@@FesZElectronicsplease please more videos about mag amps, especially practical ones.

[@alexengineering3754]

very interesting. i can imagine that old dimmer switches work that way but i am not old enought to know.

[@johnwest7993]

Thank you.

[@trcwm]

Excellent! ❤

[@n.shiina8798]

Seasonic S12iii uses magamps to regulate its 5 and 3.3V rails. i havent seen any other PSUs using it for both minor rails

[@sergepetrov8598]

Maybe they have an engineer who is good in magamps and likes them.

[@jluke6861]

greate video. Thank you.

[@roliveira2225]

Excellent!

[@Stefanev]

Thanks and 👍

[@greggorr314]

First [community] theater I worked in used saturable core realtors to control the stage lights. The console had an array of slider potentiometers hooked up to the dimmer cabinet, delivering a 0-10vDC voltage to each lighting channel. Buzzed like a hornets' nest when the lights were brought up. Cool retro'. Doubtless the waveform was ratty as h311.

[@filipc3010]

Nice explenation. Mr. Ben-Yaakov uploded the video about usage of this variable inductor idea to provide lower losses in power converters working with much higher input voltages (Buck)/smaller load. Probably variable frequency is still an easier solution.

[@sergepetrov8598]

And that explanation shows a nice picture on how "change inductance using DC" works without AC backfeed.

[@_wave64_]

3:07 in this setup, how much of the inductance decrease is caused by the power supply circuit actually shorting out the secondary side as you decrease the series resistance? At 250mA, there is only ~12 ohms on the secondary.

[@_wave64_]

10:22 I think this whole experiment is wrong - the current simply passes through the 20 ohm resistance and the power supply - try disconnecting the power supply and simply short the leads together - the light bulbs will come up again.

[@sergepetrov8598]

@@_wave64_ If so, how would one control inductance from electrical input, and in continuous fashion? It would need a variable resistor, but usual electronics and mechanics are out of scope.

[@_wave64_]

@@sergepetrov8598 With high resistor values and high voltage, or with a current source/sink (vacuum tube anode). The problem in these experiments is that (besides explaining the theory correctly) he is using low resistance values that are simply shorting out the secondary side - so what we see is not the core saturation, but rather just a simple impedance transformation.

[@sergepetrov8598]

@@_wave64_ My bad not to notice that coil was fed from voltage source with resistor, not from current limiter mode. Got used to CV/CC power supplies. If windings are located separately on a ring, leakage inductance can be huge. In my experiment some months ago, shortening secondary resulted in ~~ 50% drop in primary inductance. Here (pri and sec) wires look the same, unfortunately.

[@batica81]

Fascinating subject, and well presented, as always. I wonder could this principle be used to make a "voltage controlled" low pass filter? Changing the inductivity of the filter L component should give a different response. Also, could saturating that core from your example be done by a simple permanent magnet moved into proximity?

[@sergepetrov8598]

You are right, it's all possible.

[@wimhey]

Yes i've seen it in the seventies for a speed regulator for elevators.

[@mr.cunamis]

Hi can you make a video about Parametric Amplifiers? (They are used as LNA at really higb frequencies). I think it would be interesting. Thanks

[@ShredEngineerPhD]

9:54 I don't get why the circuit does not behave symmetrically w.r.t. the bias current… Could you elaborate?

[@ghlscitel6714]

There are still magnetic voltage regulators (Magnetischer Spannungskonstanter) in the market, at least in Germany.

[@sergepetrov8598]

Well, afaik magamps were widely used in Wunderwaffe "Fau" missiles during WW2.

[@ghlscitel6714]

@@sergepetrov8598 What have they done in these units?

[@sergepetrov8598]

@@ghlscitel6714 Afaik missile included at least INS and some flight program, so needed control circuits. And German engineers prefered magamps to vacuum tubes for some reason.

[@Cynthia_Cantrell]

@@sergepetrov8598 Vacuum tubes tend to be very delicate devices, easily damaged by vibration - such as that would be seen in military equipment - whether it was being launched or the target of explostive devices. Mag amps on the other hand don't care about vibration, they are much more robust in high-shock environments.

[@willthecat3861]

@@sergepetrov8598 Think of late 1930's electronic power amplifiers, and then contrast that with robust, predictable, powerful, magnetic amplifiers... especially for servo control.

[@yellowest_ball]

heell yeahhh Some Sony devices had a magnetic amplifier for its feedback between the main and secondary of the switch mode power supply, with unusually shaped cores.

[@Tadesan]

Those poor poor waveforms. Lol Thanks for a great video!

[@jayantibhakat621]

👍👍

[@analoghardwaretops3976]

Will an ML4818 (Fairchild) phase shift modulator provide accurate current control to such similar loads ...with proper f.b. loops and some additional circuitry from its outputs to generate a load frequency synced "ac controlled signal" ? ...instead of the normal "dc".....perhaps even core saturation may be tightly controlled.

[@Cynthia_Cantrell]

Getting an ACCURATE current level is relatively easy with all the high-gain opamps available. The tricky part is the control loop design. That inductance vs. current function is highly non-linear, which means the GAIN is highly non-linear. This means you have to be extra careful about loop stability, and make some compromises with response time vs. control level. If you're not careful, and design a circuit to respond quickly when the amp is nearly saturatied, it could easily have too much gain to remain stable at low saturation levels.

[@analoghardwaretops3976]

@@Cynthia_Cantrell well I shouldn't have used the word "accurate". but "reasonable".. ...On second thoughts..with the 4818...the magnetic flux could be better controlled with its phase shift control circuitry..thus core saturation is controlled .

[@BT-schannel]

From what I understand.. under the right conditions inductors behave like transistors..

[@sammin5764]

🌟

[@djisydneyaustralia]

Haha turn on subtitles and watch the opening , he mentions saturation of the transformer and then talks of magnetic ampli fires as per subtitles haha I guess that’s the result with sustained saturation

[@masudsamarin6166]

Can I make a variac by this method?

[@FesZElectronics]

Not really; with a variac, you change the turns ratio, not the specific inductance...

[@sergepetrov8598]

It will work at least for resistive loads.

[@chilifinger]

In this episode: Speedy Gonzales invents the 'Mag Amp'.

[@SignalProduct-be2js]

was it used before BJT transistors were invented?

[@Cynthia_Cantrell]

Definitely. Think gun control circuits for WWII ships.

[@grumble2009]

The first US patent for MagAmps is from 1901 - that predates the earliest vacuum tubes.

[@ahmedalshalchi]

This shows that working with magnetic is just working in mine field of non linearity that noway but blow up something ....

[@action4free369]

Nice 😘

[@fmk326]

Please more info and details for "audio" mag amps, or funtional scheme, please, please please please³ 😢😅at the orden and whit the soulder on 😊 Thanks of the Spanish pacient man

[@willthecat3861]

Seems a significant amount of people never heard of magnetic amplifiers. IME, either their engineering school sucked, or they did computer engineering and spent most of their time playing video games and drinking beer. IMO... if you don't understand D.C. currents, superimposed on A.C. currents, in inductors and transformers, you don't understand either.

[@niallflynn1833]

Quantum computing modules.....

[@pnachtwey]

Magamps were used in many places in the navy. They are robust.