In the eighties I serviced a CO2 laser that used a magnetic amplifier to control the output power of a high voltage transformer that powered the laser. The current through the control winding was regulated by an eimac triode. The grid voltage of the triode was regulated by a transistor circuit. The combination of those three technologies made it interesting. Sometimes it pays to be aware of how “old” stuff works.
That is so interesting particularly to me and I'll tell you why: in the late 80s early 90s I was a grad student and often used a Spectra-Physics argon laser in our lab. It was run by a large three phase power supply and water cooled, all to produce probably less than 5w of light. From your post now I wonder if it had a magnetic amp in the power supply - although I think I would have noticed that - I'm sure we had the schematic. And for a laugh - about 5 years ago I was visiting the university with my long retired grad studies supervisor - in the center of the department lobby they had a display of antique equipment - the center piece was our laser :) You are so right about "it pays to be aware of how “old” stuff works".
The nuclear submarine I was stationed on in the 80s was built in the early 70s, and had mag amps for a lot of the control circuity in the propulsion plant. Very old school, but rugged as hell and designed to withstand the rigors of combat. This was a nice trip down memory lane.
Really! Wow - that's really interesting. I guess weight would not have been a problem , and rugged - particularly compared to tubes which may have been the only other high power option back then. So glad you enjoyed a trip down memory lane from this video!
@@h7qvi Good point. Now that I think about it, the copper and steel are probably immune to radiation and even the plastic insulation would only degrade with really high doses - and I'm sure there are plastics or similar insulators know to withstand very high level of radioactivity.
Your video was probably suggested for me by TH-cam because I was investigating fluxgate current sensors and compasses and I do think that I understand these better after this video. That being said, it would be great for everyone if you would someday make a video about them! Nice video!
Wow - I am amazed that they managed to link the two - although maybe since I mention flux a number of times they connected that to fluxgate. I'm glad you liked the video and found it useful. I was actually debating whether to mention how the two transformers go non-linear on the opposite sides of the the AC cycle but in the end decided it would over-complicate the explanation and I am always struggling with videos getting too long. That would have been so close to the fluxgate sensor. I will keep it in mind for a future video - just added it to my list of topics - maybe along with Hall Effect sensing ....
watching the current spikes phase up and back on the AC waveform reminds me of a DC drive, and the action of those coils reminds me of the saturable reactors some of the early DC drives used
Saturable reactor - thats exactly it! - the purpose built version of of this usually with the DC coils being on the outer leg of a fairly typical IE transformer core E. The spikes can be huge - undesired in this case but once commenter pointed out how they were used to trigger high current tubes way back before high power semiconductors.
Back in the late 80s a friend bought an old Burroughs computer that was being scrapped from a defunct business. $10 or something like that, more or less take it away and it's yours. It had core memory in it. The carriers were about an inch square of white ceramic material with gold plated pins. The array was a square of uniform tiny cores that were small enough to need a magnifying glass to see the feed through holes. They really were a work of art. I have one here somewhere. The idea of using different signal levels to increase data density should be familiar to many people. It's exactly the same concept as employed by MLC SSDs to increase the NAND storage capabilities. The shape of the spike in the waveform will be familiar to people that have done work with magnetic media, floppy drives, tape drives etc.
I would have killed for that computer! Really tiny cores - must have been the final development of that technology. I actually have some vintage core memory too - one sample from the US and one from the USSR. Both bigger than yours. Amazing how they would have woven them - as you said, a work of art. Yes - multi-level NAND storage - same idea. I don't think the vintage cores ever were multi-level though - although maybe someone will correct me on that!
Very interesting! I haven't tried building a mag amp before. However, the projects I work on use battery chargers operating on this principle. They call the device doing the magnetic amplification a "saturable reactor." It has a three-leg core, like an E-Core transformer. The center post of the core has the DC winding for controlling the current; and the outer two legs of the core have the AC windings which pass the current to the load. The charger has a typical power transformer and rectifier for charging the battery; with the sat reactor limiting the current reaching the transformer. There is a small circuit which monitors the battery voltage and charger current. It provides a small DC control current to the sat reactor for controlling the charger. Look up LaMarche A12B battery charger. Good company; still making a product like things should be made. It's so cool to see the old school technology making up the core of the charger, with a modern monitoring which can display all the info via Modbus to a PLC. Definitely the best of both worlds. Very neat demonstration of magnetic core memory concepts, too. Always enjoy your projects. Thanks and please keep them coming!
Thanks! Thats so interesting - I have never seen the a "real" saturable reactor as you describe other than in photos. And I will look up the LaMarche A12B - I'll bet it is nice and reliable. So cool that it has a modern interface. And I will keep them coming! Its too bad we are so far apart - when I look at your videos I always think it would be neat to do a project together!
@@ElectromagneticVideos It is interesting! They have done a good job of keeping the modern electronics in a "supervisory" role, as well. Even if that stops working, the charger still performs its core functions. Yep, Alabama is a fair piece of travel from Canada. I also think it would be good to meet up some day and share projects. With my work, travel is a constant thing, though. I do have my e-mail available on my channel page, if you go to the about page.
@@davida1hiwaaynet Yes - sometime when one of us is near the others home base, would be a great thing to get together. I would be fascinated to see your workshop with all the vintage machinery!
I have an old heavy TIG welder that has a very interesting amperage control. Imagine a three-phase transformer, with three cores, the core on the left will be the input from the 240 volts, the core on the right is the output for the tig, something like 24 ark volts? Imagine the middle core with nothing around it, the magnetic fields will flow through it taking the path of least resistance and did not making it to the end core. The electronic control for this TIG welder used a separate coil around the middle core that cuts off the magnetic field keeping the magnetic field from taking the path of least resistance. It does it without actually passing much amperage through that coil of wire. The output core on the right has a variable output depending upon how much magnetic resistance you impart on the middle core. Of course in this scenario there's nothing actually three phase about this transformer just imagine it as a way of picturing what it would look like. I might not have done a good job of describing what's going on. Sorry if I didn't describe it well enough. It's a pretty cool old school setup if you ask me, no Giant fet's or igbts. The high frequency start is a whole different monster on that machine. One that was particularly in danger of overheating.
That cool. So the general setup you describe is typical of a classic old welder transformer. The center leg of the core allows as you suggested allows some but not all of the field to take that leakage path. This has and effect similar to having a separate inductor in series with the 240V input that limits the current input and therefore also the low voltage output current. This has the effect of making it a crude constant current (or perhaps max current) source output so when striking the arc you don't effectively short the system a blow a fuse on your 240V supply. In simple cheap arc welders like the one I have been using as a current source, the leakage field on the center leg (and hence current out) is adjusted by cranking a piece of steel in or out of a gap in that leg. It sounds like your more elaborate TIG welder might actually be using saturation by passing DC though the center coil to regulate the amount of leakage flux. Is there some automatic regulation? That would work well with saturation control. I have the small modern electronic power supply welders for sale. I wonder if they are easier to use since they presumably have precise current control?
@@ElectromagneticVideos interesting, I actually hadn't considered that the center coil could have DC through it, it appears to be quite a lot of winding on the center coil, I had somehow imagined that they effectively controlling an AC load on the center coil instead of supplying DC to it. There does not appear to be a DC power supply on the control board. Now you've really got me wondering. First time around, I must have sat there and stared at it for like 15 minutes before got my head around how that transformer was supposed to work, a gap in the Middle with a steel plate would have been a lot easier to wrap my head around. As it is that part of the welder works fine, it was the high frequency that went bad. It was a solid state system that used a high frequency transformer wound with lits wire... Which was also my first introduction to lits wire. Lol the transformer totally burnt up and I tried to rewind it with wire that wasn't litz wire, as you can imagine it wouldn't pass as much current and didn't give me as much high frequency Arc. The replacement transformer is no longer in production and hundreds of dollars if you can find one. I plan to to change out the high frequency start with an arc gap system I bought off of AliExpress. Until then it's just a stick welder.
I have worked with magnetic amplifiers as a means of measuring high DC currents (approximately 400A). A pair of inverse wound cores over the main power bar will give an output proportional to the current in the main conductor. Very robust and reliable and ideal for rough environments such as railway trains.
So using it as a current sensor? I had no idea the concept was used that way! So from what you say, it sounds reasonably linear presumably when beyond a small minimum current? Railways - so to measure the current going into the drive motors in a locomotive?
@@ElectromagneticVideos I am going back to 1980 before microprocessor controls etc. The design consisted of two back to back cores made using mu metal so the magnetizing current was low, They were excited with a 400Hz square wave from an inverter. They had decent linearity over quite a wide range and were ideal for the application Probably now replaced with hall effect transducers on new designs.
@@apk55 Interesting! I didnt know mu metal was used for anything other than shielding. I'll bet your right about Hall effect replacing it - although one viewer pointed out fluxgate sensors which are two winding saturation based sensitive magnetic sensors.
As in a constant voltage transformer? I used one for my first computer years ago. The one I had got really hot from all the circulating currents. Worked great though!
Really interesting stuff! I remember reading about a high-tesla setup and it said that full field was achieved only after about 30 minutes. I didn't understand why, but I think I understand now that I see the lag effect and the likely carefully applied current and voltage waveforms to coerce a coherent and maximum magnetic field.
Thats intersting! I'm surprized it would take that long but perhaps with a huge effective inductance it could have a really long effective time constant. I have never worked with high fields like that but have seen those photos of a frog being levitated with super high field strength.
Very interesting. I saw a few videos on the subject, but yours is the best! I have two questions though: 1) Could this be used as an amplifier at HF (radio frequencies) I mean, could the saturation be changed fast enough? 2) Could this concept be used to change the inductivity of the toroid coil, making it a "voltage controlled variable inductor"?
Thank you so much! Your questions: 1) Yes - it was apparently used in the early days of radio with Alexanderson alternator to modulate some voice transmissions, but the problem is its limited frequency response as pointed out. My guess is it was just barely fast enough. 2) Yes! If you kept the AC signal voltage low which it usually is, you could nicely vary the effective inductance. I dont know of any examples of this, but I'm sure it has been used somewhere!
Wow - EW stuff on a sub. You might enjoy a previous comment from someone who also worked on a sub where magnetic amps controlled propulsion (look for @johnathancorgan3994) . Its interesting they were used on subs. I wonder if it has to do with them being more mechanically rugged than something like a thyratron in the days before high power semiconductors. And also less vulnerable to EMP than semiconductors.
@@ElectromagneticVideos the one i referred to was in 1980, there were a few fets available but had reliability problems. this system had a 100khz inverter driving about 15 output boards with a sine wave. magamps used 2 E34 ferrite cores for controlling the outputs of around 150W on some bourds. the SONAR CRT deflection took +- 15V at 7.5A. It was very reliable but hard to make. What started me on this is a guy that I worked with who was in charge of the transformer winding dep of a large enectronics manufacturer, he told me about 50Hz ones and I saw an opportunity with the new SMPS circuits I was working on.
I had never heard the term Magnecitor. Just googled and an found only a few fleeting mentions of it come back. I'm guessing its some sort of magnectic amp or you wouldn't have mentioned. I would be really interested to hear how it worked.
@@ElectromagneticVideos i saw the circuit of one i could make out on a label inside the box of an old generator on the smallenginemechanic channel some months ago. Iirc, it was the type of magnetic pwm where the ac is applied to a control coil through a diode. Search for Onan magneciter, there's a few manuals
@@h7qvi I'll do some looking - I can certainly see something like that being a nice rugged and reliable regulator for those hot environments with lots of vibration.
Something I’ve been concerned with is the frequency stability issues on the output. I’ve read that these are very frequency sensitive in the sense that a small change in frequency in the AC will be magnified. So for example, trying to use this to amplify in an oscillator application is a non-starter... I had hoped I could use this in the amplification stage of an oscillator, but to no avail due to the frequency issue. Obviously this can’t simply replace a transistor (which, as a semiconductor device, exhibits ‘negative resistance’ characteristics - like in a Colpitts oscillator). Thoughts, maybe I’m completely off base with my assumptions? Great videos by the way. 🤗
It is frequency dependent because inductance is used to block the AC current. Change the AC frequency and the reactance will change allowing more or less current to pass. However this change in reactance is a different mechanism than the saturation used for amplification. Combine the two and you end up with messy combination of both. I had never thought of using a magnetic amp to make an oscillator but its theoretically possible since amplification is achieved. You would need an external AC power source which after passing though the magnetic amp could be rectified to produce DC which is an amplified version of the control DC voltage. And then with suitable feedback filtering send the amplified output back to the control input. The impractical part of this is that you would need an AC supply, and the frequency of your magnetic amp oscillator loop would have to me much lower than the AC supply frequency. I'm not sure if a circuit like this would be much practical use but would be a neat demonstration. I'm quite intrigued with your idea - maybe there is a better way of doing it that what I have presented in my thoughts. If you - or anyone reading this - has a better idea how one might do it, please comment!
It certainly is something you rarely hear about these days. Now that I think about it, I don't know where I first heard about about magnetic amps ..... certainly a neat way of controlling loads.
Very good video! I'm only about 15 minutes into it. I'm hoping that you relate this to NASA's computer for going to the Moon the first time. I watched the whole documentary on that computer alone. It's cool to see a macroscopic version. Lol
Glad you liked it. Unfortunately I didn't mention the Apollo computers with the woven core memory (no easy code changes there!) - its always a struggle to mention what I want to mention and keep the video short enough for people to watch.
Yes - I'm always surprised they are virtually unknown in Canada or the US - really too bad for Cliff Electronics - they make it so easy and safer to test or experiment. In case anyone reading this wonders what we are talking about its the red box I have the bulb and toroid connected to 120V. They seem to be available from Amazon, Ebay and Newark.com
This is a really good video for being able to see how a transformer works. I particularly like how you're using a battery with two transformers so that there's no AC potential across the battery. Do you think you could accomplish the same thing in a simple EI transformer if you wrapped the opposing wires around the top and bottom of the E, and have it control the power transferred through the core, the middle of the E? If you get what I'm saying... I'm not sure if that would work or not, I'm trying to picture how you could do something similar with one transformer, you couldn't do it with a single troidal transformer.
The E config almost as you describe (or maybe exactly as you describe if I misunderstood) is actually a standard version of this called a saturable reactor: two DC control windings, each on the outer leg of the E and the AC winding in the middle leg. The two control windings are connected in series to make their AC voltages cancel and you use a battery or DC voltage just like I did ion them
@@ElectromagneticVideos Yeah I watched some videos where they were explaining old memory working on this principle. Thought - "Its inductivity in play , coil over coil - too complicated to understand " . Hard when you don't know the basic principle which you managed to explain in simplest terms and examples 👍
Glad you liked it. I would guess they used many more turn on the DC control winding so they could use a higher more practical voltage like 12 or 24 volts. I used a really low voltage here to emphasize how little is needed to control a much higher powered AC device.
the 14awg wire from the first transformer is wound identically to the second one when they are placed the same and must be connected with the same polarity, so in parallel?
From your your more recent comment, you seem to have figured it out! The trick is to have the control winding wired in series such that the AC voltage across each of them opposes each other. That way, the battery uses to control the amp will not be subjected to an AC voltage.
With 2 identical transformers hooked up as you have shown, is it possible to make a self running oscillator through Ferro magnetic saturation and positive feedback?
Thats a very interesting question. A few quick thoughts: We do have amplification, so with feedback we should be somehow able to make an oscillator. One issue is the the amplified output is at 60Hz, while the control voltage is much slower than that - essentially DC, and at a lower voltage. So maybe we could feed the output into a step-down transformer to go from 120V to maybe 6V, and then put that through a full wave bridge rectifier to make low voltage DC. I'm not sure if the pulsating nature of the DC would be a problem so one might need a large capacitor to smooth things. Also not sure what the loop phase shift would be and if one would have to add some huge capacitors or inductors to setup the conditions for oscillation. The interesting thing is that the engineers who designed magnetic amp feedback control circuits would instantly know the answer since they would have had to make sure the system was not prone to oscillation. You have certainly made me think - might be fun to experiment a bit with sometime.
@@ElectromagneticVideos Thanks for reply, a YT video showed where 2 transformers where hooked back to back with a feedback network and apparently oscillate ,could be another fake tho. Patent US3015073 Mag amp prw gain 50,000
I think that the primaries are the same (230v, 160VA). It does not matter if the secondary or secondaries are lower or higher in voltage and current (if they are not used) example; 19v, 24v, 30v, 35v, 40v, 55v, 15v, 18v etc?
Exactly - you are not connecting anything to secondaries so it doesn't matter what they are. However, be sure the ends of the wires from the secondaries are covered/insulated so they dont short!
I'd really like if you could display a schematic for a few seconds before each change to the wiring. It's hard to visualize the wires on the bench. For example, when you said "add a light bulb into the mix" I have to guess that it went in series with the primary of one of the coils. It's also not easy to tell whether the two primaries are wired in parallel to the mains, or just one primary. Oh, I see you've linked the schematic from one of the comments: th-cam.com/channels/V2-Qbq9o0EOQWPLAWkVH2w.htmlcommunity?lb=UgkxYiE7LI0tCuI2siiZf4A_UTw5m2SzSKy6
You know, I shoudl do that. I also find it frustrating when people dont do that - now I am guilty. Yes _ did link it cause someone else asked. So - point well taken.
Nothing special about the battery. I used a "D" size which I think is the same a your r20. I used alkaline because they generally are longer lasting than zinc-carbon.
The problem with me is that everything has to be done first because I cannot work alone because I am a person with physical disabilities. My mother helps me and the time she works with me is limited.I have 220v at the outlet. I found some transformers 160VA 230v 30v 5,333A. Is it good to connect the primary ones in parallel and the bulb in series with the primary ones like yours or should I connect them in series?Will it work with a 100w bulb (made of 4x25w bulbs in parallel) or is it good with a 60w bulb because I have that too?How is it better?Does it matter if the 14 awg wire is solid or twisted because I found only twisted or i need a thicker wire(awg13)?
So sorry to hear you have a disability that makes it hard to do interesting experiments like this - totally understand how you have to have it all planned ahead. Since you transformers are 230V, connect the primaries in parallel. Try a 60W bulb since that is simplest. If the transformers are built to have a large inductive magnetizing current, it may glow dimly (before you saturate the cores) but that is unlikely. The larger 100W bulb is less likely to glow before being turned on since it needs more current to do so. 14AWG solid or twisted - both are fine! Good luck! A word or caution - is your mother comfortable doing stuff like this? Does she understand how to do it safely? Like I put at the end of all my videos, no experiment is worth risking any injury. If you mother is not familiar enough with this sort of thing, it would be better to find a friend or neighbor who is who can help and knows how to deal with electricity safely! Best wishes and stay safe!
With how many volts do you feed the circuit?the I have 220v at the socket. you think it can be fed with a max 300v 3kw autotransformer?My toroidal transformers are 230v on the primary and 30v on the secondary, what voltage should I apply to the primary and how many turns should I add to each transformer? In your case, the 2 primaries are connected in parallel, I think. I saw a circuit on google with the primaries connected in series .How they must be tied to the end?your light bulb is incandescent (120v or 230v)or led and what light source i can use?
You need to use a voltage that keeps the AC peaks below magnetic saturation and the is the typically the design voltage of the transformer. In my case the transformer was wired with a 120V primary so I powered the system with120V. In your case, apply 230V or something close to it like 220 or 240. You can always use a lower voltage. The voltages are high to be careful!. I would try around 20 turns and see is that works - all depends on the low voltage control source you are using an how much current it can produce - that current times number of turns needs to be enough to put the core into saturation. The primaries can be either in series or parallel - since the saturation results in asymmetric AC currents in each coil, my preference is parallel. The transformer primaries are like a crude switch or lamp dimmer so should be in series with the load. I would try a with an incandescent bulb first - not all LEDs respond well to lower voltages if the core is not fully saturated.
They are standard power transformers. The core is a long strip of laminated silicon steel wound into a doughnut shaped core. Ferrite may work - but probably not as good a core for this type of application.
They only need to carry an amp or two. Since battery voltage used is very low, the winding needs to be thick to have low resistance. I used AWG #14 wire which was fine.
The transformers I used were Hammond 1182M30 which come with various secondary voltages. Lower secondary voltages would be better, and any similar transformers should work. I have posted the schematic. Let me know if this link works. (Os transformadores que usei foram Hammond 1182M30 que vêm com várias voltagens secundárias. Tensões secundárias mais baixas seriam melhores e quaisquer transformadores semelhantes deveriam funcionar. Postei o esquema. Deixe-me saber se este link funciona. ) th-cam.com/channels/V2-Qbq9o0EOQWPLAWkVH2w.htmlcommunity?lb=UgkxYiE7LI0tCuI2siiZf4A_UTw5m2SzSKy6
About transformers, I don't have them, but I have to buy them and how many VA (volt ampere) do I need for a 100w incandescent light bulb.Are they good with 100VA ;230v ;30v; 3,333A?Or do I need a higher power and how much?I find more models with different values and i dont know what to buy.help.Or should change the light bulb an what power.Does the noise from the transformers appear when connecting and disconnecting the battery or is it continuous (that is, as long as the battery is connected)?please help me .reply
I used HAMMOND 1182M30 transformers. (Available from Newark, RS and similar suppliers). They can be configured for both 230V buy connecting the primary in series. They are 160VA and should be fine for a 100W bulb. There always is some hum from the transformer. I dont recall how noisy it was when fully saturated.
Where do you get your hands on the cores? I've looked for them before and the only thing it seems like I can find is cores that are made for much higher frequency, and usually smaller. I'd like to find one that's somewhere in the neighborhood of 1000 to 3,000 VA
Most of the main elecronics suppliers (Newark, Digikey, Mouser) have toroid transformers. These came from Newark - here is the link. canada.newark.com/hammond/1182m30/transformer-toroid-60v-160va/dp/54X7521 Be sure to switch to the Newark site for whatever country/currency you are in. They are not cheap - I picked up a bunch on clearance at 75% off. Well worth scanning Newark periodically for discounted items ..... I'm not sure how big they make them. These were made by Hammond which makes transformers of all sizes so if Newark does not have them big enough, you might look at Hammond's website directly. Shipping: they are heavy, but Newark, Digikey, Mouser usually offer free shipping for order over $100 or $200. For heavy items, its often cheaper to order something else as well to get into the free shipping order amount.
@@chadhiggins8397 Looks like a good price for the core, but you would have to wind it yourself. Doing just the 16 turns I did was a pain. Much more with thicker enameled wire would be even worse. My 180W toroids had about 500 turns for 120V. So unless you are only dealing with really low voltage I sure wouldn't try winding them. So what are you planning to make?
@@ElectromagneticVideos probably mostly want to do experimental stuff, maybe build my own low frequency inverter? I love the versatility, you can use the same transformer for a few different things by adding another winding or two.
@@ElectromagneticVideos one of the things I wanted to try was winding it with one primary winding with a center tap, and then use two small inverter generators that are designed to be paralleled, attaching each 120 volt generator to one end of the transformer and the other ends to the center. It should make 240 volts all the way across the transformer and the transformer should keep the generators paralleled, each generator thinking that it's connected to another generator in parallel when really it's in parallel but 180° out of phase. The transformer itself even with only one generator working would put out 240 volts across the ends if you connected to generator to the center and one end. Of course that'd be an auto transformer. My thought is that as long as you had a 240 volt load you wouldn't need a transformer that was all that large at all compared to the load. But even a reasonably size transformer would allow you two run something like a 120 volt refrigerator and a 240 volt well pump. It's probably cheaper just to buy a 240 volt inverter generator. Lol 🤣 but I'd still love to see if it work.
In the eighties I serviced a CO2 laser that used a magnetic amplifier to control the output power of a high voltage transformer that powered the laser. The current through the control winding was regulated by an eimac triode. The grid voltage of the triode was regulated by a transistor circuit. The combination of those three technologies made it interesting. Sometimes it pays to be aware of how “old” stuff works.
That is so interesting particularly to me and I'll tell you why: in the late 80s early 90s I was a grad student and often used a Spectra-Physics argon laser in our lab. It was run by a large three phase power supply and water cooled, all to produce probably less than 5w of light. From your post now I wonder if it had a magnetic amp in the power supply - although I think I would have noticed that - I'm sure we had the schematic.
And for a laugh - about 5 years ago I was visiting the university with my long retired grad studies supervisor - in the center of the department lobby they had a display of antique equipment - the center piece was our laser :)
You are so right about "it pays to be aware of how “old” stuff works".
The nuclear submarine I was stationed on in the 80s was built in the early 70s, and had mag amps for a lot of the control circuity in the propulsion plant. Very old school, but rugged as hell and designed to withstand the rigors of combat. This was a nice trip down memory lane.
Really! Wow - that's really interesting. I guess weight would not have been a problem , and rugged - particularly compared to tubes which may have been the only other high power option back then. So glad you enjoyed a trip down memory lane from this video!
And you know, just thinking - also also probably somewhat immune to EMP although tubes are quite immune too.
@@ElectromagneticVideos more tolerant of high radiation
@@h7qvi Good point. Now that I think about it, the copper and steel are probably immune to radiation and even the plastic insulation would only degrade with really high doses - and I'm sure there are plastics or similar insulators know to withstand very high level of radioactivity.
Your video was probably suggested for me by TH-cam because I was investigating fluxgate current sensors and compasses and I do think that I understand these better after this video. That being said, it would be great for everyone if you would someday make a video about them!
Nice video!
Wow - I am amazed that they managed to link the two - although maybe since I mention flux a number of times they connected that to fluxgate.
I'm glad you liked the video and found it useful. I was actually debating whether to mention how the two transformers go non-linear on the opposite sides of the the AC cycle but in the end decided it would over-complicate the explanation and I am always struggling with videos getting too long. That would have been so close to the fluxgate sensor. I will keep it in mind for a future video - just added it to my list of topics - maybe along with Hall Effect sensing ....
this channel is going to boom. The presentations are great. Great work!
Thank you so much! It will be interesting to see how much of an audience there is for this wort of video!
watching the current spikes phase up and back on the AC waveform reminds me of a DC drive, and the action of those coils reminds me of the saturable reactors some of the early DC drives used
Saturable reactor - thats exactly it! - the purpose built version of of this usually with the DC coils being on the outer leg of a fairly typical IE transformer core E.
The spikes can be huge - undesired in this case but once commenter pointed out how they were used to trigger high current tubes way back before high power semiconductors.
I didn't expect the battery to do anything, this was for sure very interesting to me and I learned something from you today.
Ricardo I'm so glad you found it interesting! It sure is a really unusual way of amplifying!
Back in the late 80s a friend bought an old Burroughs computer that was being scrapped from a defunct business. $10 or something like that, more or less take it away and it's yours. It had core memory in it. The carriers were about an inch square of white ceramic material with gold plated pins. The array was a square of uniform tiny cores that were small enough to need a magnifying glass to see the feed through holes. They really were a work of art. I have one here somewhere. The idea of using different signal levels to increase data density should be familiar to many people. It's exactly the same concept as employed by MLC SSDs to increase the NAND storage capabilities. The shape of the spike in the waveform will be familiar to people that have done work with magnetic media, floppy drives, tape drives etc.
I would have killed for that computer! Really tiny cores - must have been the final development of that technology. I actually have some vintage core memory too - one sample from the US and one from the USSR. Both bigger than yours. Amazing how they would have woven them - as you said, a work of art. Yes - multi-level NAND storage - same idea. I don't think the vintage cores ever were multi-level though - although maybe someone will correct me on that!
Very interesting! I haven't tried building a mag amp before. However, the projects I work on use battery chargers operating on this principle. They call the device doing the magnetic amplification a "saturable reactor." It has a three-leg core, like an E-Core transformer. The center post of the core has the DC winding for controlling the current; and the outer two legs of the core have the AC windings which pass the current to the load.
The charger has a typical power transformer and rectifier for charging the battery; with the sat reactor limiting the current reaching the transformer. There is a small circuit which monitors the battery voltage and charger current. It provides a small DC control current to the sat reactor for controlling the charger.
Look up LaMarche A12B battery charger. Good company; still making a product like things should be made. It's so cool to see the old school technology making up the core of the charger, with a modern monitoring which can display all the info via Modbus to a PLC. Definitely the best of both worlds.
Very neat demonstration of magnetic core memory concepts, too.
Always enjoy your projects. Thanks and please keep them coming!
Thanks! Thats so interesting - I have never seen the a "real" saturable reactor as you describe other than in photos. And I will look up the LaMarche A12B - I'll bet it is nice and reliable. So cool that it has a modern interface.
And I will keep them coming! Its too bad we are so far apart - when I look at your videos I always think it would be neat to do a project together!
@@ElectromagneticVideos
It is interesting! They have done a good job of keeping the modern electronics in a "supervisory" role, as well. Even if that stops working, the charger still performs its core functions.
Yep, Alabama is a fair piece of travel from Canada. I also think it would be good to meet up some day and share projects. With my work, travel is a constant thing, though. I do have my e-mail available on my channel page, if you go to the about page.
@@davida1hiwaaynet Yes - sometime when one of us is near the others home base, would be a great thing to get together. I would be fascinated to see your workshop with all the vintage machinery!
I have an old heavy TIG welder that has a very interesting amperage control. Imagine a three-phase transformer, with three cores, the core on the left will be the input from the 240 volts, the core on the right is the output for the tig, something like 24 ark volts? Imagine the middle core with nothing around it, the magnetic fields will flow through it taking the path of least resistance and did not making it to the end core. The electronic control for this TIG welder used a separate coil around the middle core that cuts off the magnetic field keeping the magnetic field from taking the path of least resistance. It does it without actually passing much amperage through that coil of wire. The output core on the right has a variable output depending upon how much magnetic resistance you impart on the middle core. Of course in this scenario there's nothing actually three phase about this transformer just imagine it as a way of picturing what it would look like. I might not have done a good job of describing what's going on. Sorry if I didn't describe it well enough. It's a pretty cool old school setup if you ask me, no Giant fet's or igbts. The high frequency start is a whole different monster on that machine. One that was particularly in danger of overheating.
That cool. So the general setup you describe is typical of a classic old welder transformer. The center leg of the core allows as you suggested allows some but not all of the field to take that leakage path. This has and effect similar to having a separate inductor in series with the 240V input that limits the current input and therefore also the low voltage output current. This has the effect of making it a crude constant current (or perhaps max current) source output so when striking the arc you don't effectively short the system a blow a fuse on your 240V supply.
In simple cheap arc welders like the one I have been using as a current source, the leakage field on the center leg (and hence current out) is adjusted by cranking a piece of steel in or out of a gap in that leg. It sounds like your more elaborate TIG welder might actually be using saturation by passing DC though the center coil to regulate the amount of leakage flux. Is there some automatic regulation? That would work well with saturation control.
I have the small modern electronic power supply welders for sale. I wonder if they are easier to use since they presumably have precise current control?
@@ElectromagneticVideos interesting, I actually hadn't considered that the center coil could have DC through it, it appears to be quite a lot of winding on the center coil, I had somehow imagined that they effectively controlling an AC load on the center coil instead of supplying DC to it. There does not appear to be a DC power supply on the control board. Now you've really got me wondering. First time around, I must have sat there and stared at it for like 15 minutes before got my head around how that transformer was supposed to work, a gap in the Middle with a steel plate would have been a lot easier to wrap my head around. As it is that part of the welder works fine, it was the high frequency that went bad. It was a solid state system that used a high frequency transformer wound with lits wire... Which was also my first introduction to lits wire. Lol the transformer totally burnt up and I tried to rewind it with wire that wasn't litz wire, as you can imagine it wouldn't pass as much current and didn't give me as much high frequency Arc. The replacement transformer is no longer in production and hundreds of dollars if you can find one. I plan to to change out the high frequency start with an arc gap system I bought off of AliExpress. Until then it's just a stick welder.
I have worked with magnetic amplifiers as a means of measuring high DC currents (approximately 400A). A pair of inverse wound cores over the main power bar will give an output proportional to the current in the main conductor. Very robust and reliable and ideal for rough environments such as railway trains.
So using it as a current sensor? I had no idea the concept was used that way! So from what you say, it sounds reasonably linear presumably when beyond a small minimum current? Railways - so to measure the current going into the drive motors in a locomotive?
@@ElectromagneticVideos I am going back to 1980 before microprocessor controls etc. The design consisted of two back to back cores made using mu metal so the magnetizing current was low, They were excited with a 400Hz square wave from an inverter. They had decent linearity over quite a wide range and were ideal for the application
Probably now replaced with hall effect transducers on new designs.
@@apk55 Interesting! I didnt know mu metal was used for anything other than shielding. I'll bet your right about Hall effect replacing it - although one viewer pointed out fluxgate sensors which are two winding saturation based sensitive magnetic sensors.
This is a great channel, you explain reasonably complex topics pretty well. Subscribed.
Thank You! I really appreciate that - hope you enjoy my future videos!
I know it by the name of saturated core regulator, it was commonly used in external voltage stabilizers for TVs.
As in a constant voltage transformer? I used one for my first computer years ago. The one I had got really hot from all the circulating currents. Worked great though!
Really interesting stuff! I remember reading about a high-tesla setup and it said that full field was achieved only after about 30 minutes. I didn't understand why, but I think I understand now that I see the lag effect and the likely carefully applied current and voltage waveforms to coerce a coherent and maximum magnetic field.
Thats intersting! I'm surprized it would take that long but perhaps with a huge effective inductance it could have a really long effective time constant. I have never worked with high fields like that but have seen those photos of a frog being levitated with super high field strength.
Very interesting. I saw a few videos on the subject, but yours is the best! I have two questions though:
1) Could this be used as an amplifier at HF (radio frequencies) I mean, could the saturation be changed fast enough?
2) Could this concept be used to change the inductivity of the toroid coil, making it a "voltage controlled variable inductor"?
Thank you so much!
Your questions:
1) Yes - it was apparently used in the early days of radio with Alexanderson alternator to modulate some voice transmissions, but the problem is its limited frequency response as pointed out. My guess is it was just barely fast enough.
2) Yes! If you kept the AC signal voltage low which it usually is, you could nicely vary the effective inductance. I dont know of any examples of this, but I'm sure it has been used somewhere!
EW I used mag amps a lot through my carier, one application was a PSU I designed for the EW equipment on a SUb.
Wow - EW stuff on a sub. You might enjoy a previous comment from someone who also worked on a sub where magnetic amps controlled propulsion (look for @johnathancorgan3994) . Its interesting they were used on subs. I wonder if it has to do with them being more mechanically rugged than something like a thyratron in the days before high power semiconductors. And also less vulnerable to EMP than semiconductors.
@@ElectromagneticVideos the one i referred to was in 1980, there were a few fets available but had reliability problems. this system had a 100khz inverter driving about 15 output boards with a sine wave. magamps used 2 E34 ferrite cores for controlling the outputs of around 150W on some bourds. the SONAR CRT deflection took +- 15V at 7.5A. It was very reliable but hard to make. What started me on this is a guy that I worked with who was in charge of the transformer winding dep of a large enectronics manufacturer, he told me about 50Hz ones and I saw an opportunity with the new SMPS circuits I was working on.
Magnecitors on AC generators regulate the field winding current. Also used on old diesel locomotives for the same thing in the generator or motors.
I had never heard the term Magnecitor. Just googled and an found only a few fleeting mentions of it come back. I'm guessing its some sort of magnectic amp or you wouldn't have mentioned. I would be really interested to hear how it worked.
@@ElectromagneticVideos i saw the circuit of one i could make out on a label inside the box of an old generator on the smallenginemechanic channel some months ago. Iirc, it was the type of magnetic pwm where the ac is applied to a control coil through a diode.
Search for Onan magneciter, there's a few manuals
@@h7qvi I'll do some looking - I can certainly see something like that being a nice rugged and reliable regulator for those hot environments with lots of vibration.
Something I’ve been concerned with is the frequency stability issues on the output. I’ve read that these are very frequency sensitive in the sense that a small change in frequency in the AC will be magnified. So for example, trying to use this to amplify in an oscillator application is a non-starter... I had hoped I could use this in the amplification stage of an oscillator, but to no avail due to the frequency issue. Obviously this can’t simply replace a transistor (which, as a semiconductor device, exhibits ‘negative resistance’ characteristics - like in a Colpitts oscillator). Thoughts, maybe I’m completely off base with my assumptions? Great videos by the way. 🤗
It is frequency dependent because inductance is used to block the AC current. Change the AC frequency and the reactance will change allowing more or less current to pass. However this change in reactance is a different mechanism than the saturation used for amplification. Combine the two and you end up with messy combination of both.
I had never thought of using a magnetic amp to make an oscillator but its theoretically possible since amplification is achieved. You would need an external AC power source which after passing though the magnetic amp could be rectified to produce DC which is an amplified version of the control DC voltage. And then with suitable feedback filtering send the amplified output back to the control input.
The impractical part of this is that you would need an AC supply, and the frequency of your magnetic amp oscillator loop would have to me much lower than the AC supply frequency. I'm not sure if a circuit like this would be much practical use but would be a neat demonstration.
I'm quite intrigued with your idea - maybe there is a better way of doing it that what I have presented in my thoughts. If you - or anyone reading this - has a better idea how one might do it, please comment!
Cool! Did not know about this before!
It certainly is something you rarely hear about these days. Now that I think about it, I don't know where I first heard about about magnetic amps ..... certainly a neat way of controlling loads.
This great video earned my subscription!
Wow! I'm glad this video made such an impression! Hope you enjoy some of my other videos as much as this one!
Very good video! I'm only about 15 minutes into it. I'm hoping that you relate this to NASA's computer for going to the Moon the first time. I watched the whole documentary on that computer alone. It's cool to see a macroscopic version. Lol
Glad you liked it. Unfortunately I didn't mention the Apollo computers with the woven core memory (no easy code changes there!) - its always a struggle to mention what I want to mention and keep the video short enough for people to watch.
Cliff Quicktests are great.
Yes - I'm always surprised they are virtually unknown in Canada or the US - really too bad for Cliff Electronics - they make it so easy and safer to test or experiment. In case anyone reading this wonders what we are talking about its the red box I have the bulb and toroid connected to 120V. They seem to be available from Amazon, Ebay and Newark.com
Thank You for sharing!
Your welcome!
This is a really good video for being able to see how a transformer works. I particularly like how you're using a battery with two transformers so that there's no AC potential across the battery. Do you think you could accomplish the same thing in a simple EI transformer if you wrapped the opposing wires around the top and bottom of the E, and have it control the power transferred through the core, the middle of the E? If you get what I'm saying... I'm not sure if that would work or not, I'm trying to picture how you could do something similar with one transformer, you couldn't do it with a single troidal transformer.
The E config almost as you describe (or maybe exactly as you describe if I misunderstood) is actually a standard version of this called a saturable reactor: two DC control windings, each on the outer leg of the E and the AC winding in the middle leg. The two control windings are connected in series to make their AC voltages cancel and you use a battery or DC voltage just like I did ion them
@@ElectromagneticVideos cool, that's what I was thinking! I had no idea that had a name!
@@ElectromagneticVideos I have seen some welders with variable inductor control, do you know if that's how they do it?
I never understood this. Now I do get it . Thank you 👍
I'm so happy to hear that - thanks - I never know if my time-limited video explanations are understandable!
@@ElectromagneticVideos Yeah I watched some videos where they were explaining old memory working on this principle. Thought - "Its inductivity in play , coil over coil - too complicated to understand " . Hard when you don't know the basic principle which you managed to explain in simplest terms and examples 👍
@@cocosloan3748 That you sooooo much! I'm really thrilled to have succeeded in making it as simple as possible!
Great.
Good knowledge
May be the under water submarine used only 1.5 volts to 3 volts batteries and doing something similar.
Glad you liked it. I would guess they used many more turn on the DC control winding so they could use a higher more practical voltage like 12 or 24 volts. I used a really low voltage here to emphasize how little is needed to control a much higher powered AC device.
the 14awg wire from the first transformer is wound identically to the second one when they are placed the same and must be connected with the same polarity, so in parallel?
From your your more recent comment, you seem to have figured it out! The trick is to have the control winding wired in series such that the AC voltage across each of them opposes each other. That way, the battery uses to control the amp will not be subjected to an AC voltage.
With 2 identical transformers hooked up as you have shown, is it possible to make a self running oscillator through Ferro magnetic saturation and positive feedback?
Thats a very interesting question. A few quick thoughts: We do have amplification, so with feedback we should be somehow able to make an oscillator. One issue is the the amplified output is at 60Hz, while the control voltage is much slower than that - essentially DC, and at a lower voltage. So maybe we could feed the output into a step-down transformer to go from 120V to maybe 6V, and then put that through a full wave bridge rectifier to make low voltage DC. I'm not sure if the pulsating nature of the DC would be a problem so one might need a large capacitor to smooth things. Also not sure what the loop phase shift would be and if one would have to add some huge capacitors or inductors to setup the conditions for oscillation.
The interesting thing is that the engineers who designed magnetic amp feedback control circuits would instantly know the answer since they would have had to make sure the system was not prone to oscillation.
You have certainly made me think - might be fun to experiment a bit with sometime.
@@ElectromagneticVideos Thanks for reply, a YT video showed where 2 transformers where hooked back to back with a feedback network and apparently oscillate ,could be another fake tho. Patent US3015073 Mag amp prw gain 50,000
Super lessen
Thanks!
I think that the primaries are the same (230v, 160VA). It does not matter if the secondary or secondaries are lower or higher in voltage and current (if they are not used) example; 19v, 24v, 30v, 35v, 40v, 55v, 15v, 18v etc?
Exactly - you are not connecting anything to secondaries so it doesn't matter what they are. However, be sure the ends of the wires from the secondaries are covered/insulated so they dont short!
that was awesome
It is neat! Glad you liked it!
I'd really like if you could display a schematic for a few seconds before each change to the wiring. It's hard to visualize the wires on the bench. For example, when you said "add a light bulb into the mix" I have to guess that it went in series with the primary of one of the coils. It's also not easy to tell whether the two primaries are wired in parallel to the mains, or just one primary.
Oh, I see you've linked the schematic from one of the comments: th-cam.com/channels/V2-Qbq9o0EOQWPLAWkVH2w.htmlcommunity?lb=UgkxYiE7LI0tCuI2siiZf4A_UTw5m2SzSKy6
You know, I shoudl do that. I also find it frustrating when people dont do that - now I am guilty. Yes _ did link it cause someone else asked. So - point well taken.
The dry cell is a special battery or it work with 1,5v same size alkaline or zinc-carbon battery or 1,2v r20 size acumulator?
Nothing special about the battery. I used a "D" size which I think is the same a your r20. I used alkaline because they generally are longer lasting than zinc-carbon.
The problem with me is that everything has to be done first because I cannot work alone because I am a person with physical disabilities. My mother helps me and the time she works with me is limited.I have 220v at the outlet. I found some transformers 160VA 230v 30v 5,333A. Is it good to connect the primary ones in parallel and the bulb in series with the primary ones like yours or should I connect them in series?Will it work with a 100w bulb (made of 4x25w bulbs in parallel) or is it good with a 60w bulb because I have that too?How is it better?Does it matter if the 14 awg wire is solid or twisted because I found only twisted or i need a thicker wire(awg13)?
So sorry to hear you have a disability that makes it hard to do interesting experiments like this - totally understand how you have to have it all planned ahead.
Since you transformers are 230V, connect the primaries in parallel. Try a 60W bulb since that is simplest. If the transformers are built to have a large inductive magnetizing current, it may glow dimly (before you saturate the cores) but that is unlikely. The larger 100W bulb is less likely to glow before being turned on since it needs more current to do so.
14AWG solid or twisted - both are fine! Good luck!
A word or caution - is your mother comfortable doing stuff like this? Does she understand how to do it safely? Like I put at the end of all my videos, no experiment is worth risking any injury. If you mother is not familiar enough with this sort of thing, it would be better to find a friend or neighbor who is who can help and knows how to deal with electricity safely!
Best wishes and stay safe!
With how many volts do you feed the circuit?the I have 220v at the socket. you think it can be fed with a max 300v 3kw autotransformer?My toroidal transformers are 230v on the primary and 30v on the secondary, what voltage should I apply to the primary and how many turns should I add to each transformer? In your case, the 2 primaries are connected in parallel, I think. I saw a circuit on google with the primaries connected in series .How they must be tied to the end?your light bulb is incandescent (120v or 230v)or led and what light source i can use?
You need to use a voltage that keeps the AC peaks below magnetic saturation and the is the typically the design voltage of the transformer. In my case the transformer was wired with a 120V primary so I powered the system with120V. In your case, apply 230V or something close to it like 220 or 240. You can always use a lower voltage. The voltages are high to be careful!.
I would try around 20 turns and see is that works - all depends on the low voltage control source you are using an how much current it can produce - that current times number of turns needs to be enough to put the core into saturation.
The primaries can be either in series or parallel - since the saturation results in asymmetric AC currents in each coil, my preference is parallel.
The transformer primaries are like a crude switch or lamp dimmer so should be in series with the load. I would try a with an incandescent bulb first - not all LEDs respond well to lower voltages if the core is not fully saturated.
Are these toroidal transformers with an iron core or can they also be ferrite? I suspect that if they work at low frequencies, they are made of iron.
They are standard power transformers. The core is a long strip of laminated silicon steel wound into a doughnut shaped core. Ferrite may work - but probably not as good a core for this type of application.
How thick should the windings with 20 turns be?
They only need to carry an amp or two. Since battery voltage used is very low, the winding needs to be thick to have low resistance. I used AWG #14 wire which was fine.
Bom dia , me manda o esquema por favor, gostei , quero enteder melhor como funciona.
The transformers I used were Hammond 1182M30 which come with various secondary voltages. Lower secondary voltages would be better, and any similar transformers should work. I have posted the schematic. Let me know if this link works. (Os transformadores que usei foram Hammond 1182M30 que vêm com várias voltagens secundárias. Tensões secundárias mais baixas seriam melhores e quaisquer transformadores semelhantes deveriam funcionar. Postei o esquema. Deixe-me saber se este link funciona. ) th-cam.com/channels/V2-Qbq9o0EOQWPLAWkVH2w.htmlcommunity?lb=UgkxYiE7LI0tCuI2siiZf4A_UTw5m2SzSKy6
About transformers, I don't have them, but I have to buy them and how many VA (volt ampere) do I need for a 100w incandescent light bulb.Are they good with 100VA ;230v ;30v; 3,333A?Or do I need a higher power and how much?I find more models with different values and i dont know what to buy.help.Or should change the light bulb an what power.Does the noise from the transformers appear when connecting and disconnecting the battery or is it continuous (that is, as long as the battery is connected)?please help me .reply
I used HAMMOND 1182M30 transformers. (Available from Newark, RS and similar suppliers). They can be configured for both 230V buy connecting the primary in series. They are 160VA and should be fine for a 100W bulb.
There always is some hum from the transformer. I dont recall how noisy it was when fully saturated.
Where do you get your hands on the cores? I've looked for them before and the only thing it seems like I can find is cores that are made for much higher frequency, and usually smaller. I'd like to find one that's somewhere in the neighborhood of 1000 to 3,000 VA
Most of the main elecronics suppliers (Newark, Digikey, Mouser) have toroid transformers. These came from Newark - here is the link. canada.newark.com/hammond/1182m30/transformer-toroid-60v-160va/dp/54X7521
Be sure to switch to the Newark site for whatever country/currency you are in.
They are not cheap - I picked up a bunch on clearance at 75% off. Well worth scanning Newark periodically for discounted items .....
I'm not sure how big they make them. These were made by Hammond which makes transformers of all sizes so if Newark does not have them big enough, you might look at Hammond's website directly.
Shipping: they are heavy, but Newark, Digikey, Mouser usually offer free shipping for order over $100 or $200. For heavy items, its often cheaper to order something else as well to get into the free shipping order amount.
Thank you for the link.
@@chadhiggins8397 Looks like a good price for the core, but you would have to wind it yourself. Doing just the 16 turns I did was a pain. Much more with thicker enameled wire would be even worse. My 180W toroids had about 500 turns for 120V. So unless you are only dealing with really low voltage I sure wouldn't try winding them. So what are you planning to make?
@@ElectromagneticVideos probably mostly want to do experimental stuff, maybe build my own low frequency inverter? I love the versatility, you can use the same transformer for a few different things by adding another winding or two.
@@ElectromagneticVideos one of the things I wanted to try was winding it with one primary winding with a center tap, and then use two small inverter generators that are designed to be paralleled, attaching each 120 volt generator to one end of the transformer and the other ends to the center. It should make 240 volts all the way across the transformer and the transformer should keep the generators paralleled, each generator thinking that it's connected to another generator in parallel when really it's in parallel but 180° out of phase. The transformer itself even with only one generator working would put out 240 volts across the ends if you connected to generator to the center and one end. Of course that'd be an auto transformer. My thought is that as long as you had a 240 volt load you wouldn't need a transformer that was all that large at all compared to the load. But even a reasonably size transformer would allow you two run something like a 120 volt refrigerator and a 240 volt well pump. It's probably cheaper just to buy a 240 volt inverter generator. Lol 🤣 but I'd still love to see if it work.
its ok with230 v, 150VA?
Should work!
I'll build you plug and play winder.....
If I was going to wind more toroids I sure would taker you up on that offer :)
how much i want onr