I tried Busting 4 Weird Electronics Myths!
video description
Date: 2025-05-10
Related videos
Comments and reviews: 20
d614gakadoug9
It isn't the pressing that makes the foil sausages lower in resistance, it is simply the fact that the total cross-sectional area of the conductor is greater because there's more foil in the same volume.
An ideal voltage source has zero output impedance. That means they are very unlikely to produce equal current when connected in parallel. If they are current limited then the invariable result is that one will go into current limit, it's voltage will fall and the supply with the next highest voltage will start to take up some of the load. Sharing can be improved by adding a small resistance in series with each supply to the common node. While that does improve sharing it degrades regulation. _Diodes will not work! _ They can prevent one supply from driving current into another and will slightly increase the output impedance of each supply but they will not cause the supplies to share well. They do degrade regulation.
Making multiple constant-voltage supplies share equally without degradation in regulation is difficult and typically requires additional active circuitry and supplies that in some fashion can be controlled by such circuitry.
Constant current supplies can be paralleled with no problem, each delivering current according to its own setpoint. An ideal current sources has infinite source impedance.
The impedance of the AC power distribution system is designed to be extremely low at mains frequency. Again, it is trying to behave as an ideal voltage source. The inherent impedance at higher frequencies in the audio range will be considerably higher, but it you consider the full audio spectrum (20 Hz to 20 kHz) you have a big problem with impedance. Things that use carrier current communications typically modulate some frequency of at least 100 kHz. Ferrite core coupling transformers are used.
Those big inductors in the Ethernet modules are to keep the load that is plugged into them from excessively attenuating the high frequency signal. Switchmode power supplies are notorious for this because they almost always have a good high-frequency capacitor right across the line.
When you emulate the mains by using a local power transformer the leakage inductance of the transformer will significantly raise the impedance at moderate frequencies.
The laminations used in motors will be very lossy at the frequencies used for modern switchmode converters. The steel is far too thick to keep eddy currents down to a tolerable level.
reply
It isn't the pressing that makes the foil sausages lower in resistance, it is simply the fact that the total cross-sectional area of the conductor is greater because there's more foil in the same volume.
An ideal voltage source has zero output impedance. That means they are very unlikely to produce equal current when connected in parallel. If they are current limited then the invariable result is that one will go into current limit, it's voltage will fall and the supply with the next highest voltage will start to take up some of the load. Sharing can be improved by adding a small resistance in series with each supply to the common node. While that does improve sharing it degrades regulation. _Diodes will not work! _ They can prevent one supply from driving current into another and will slightly increase the output impedance of each supply but they will not cause the supplies to share well. They do degrade regulation.
Making multiple constant-voltage supplies share equally without degradation in regulation is difficult and typically requires additional active circuitry and supplies that in some fashion can be controlled by such circuitry.
Constant current supplies can be paralleled with no problem, each delivering current according to its own setpoint. An ideal current sources has infinite source impedance.
The impedance of the AC power distribution system is designed to be extremely low at mains frequency. Again, it is trying to behave as an ideal voltage source. The inherent impedance at higher frequencies in the audio range will be considerably higher, but it you consider the full audio spectrum (20 Hz to 20 kHz) you have a big problem with impedance. Things that use carrier current communications typically modulate some frequency of at least 100 kHz. Ferrite core coupling transformers are used.
Those big inductors in the Ethernet modules are to keep the load that is plugged into them from excessively attenuating the high frequency signal. Switchmode power supplies are notorious for this because they almost always have a good high-frequency capacitor right across the line.
When you emulate the mains by using a local power transformer the leakage inductance of the transformer will significantly raise the impedance at moderate frequencies.
The laminations used in motors will be very lossy at the frequencies used for modern switchmode converters. The steel is far too thick to keep eddy currents down to a tolerable level.
reply
arjovenzia
Ive never heard of a motor being used as an inductor. but thats because my favorite thing about using inductors in my designs is they are about the only electronic component you can realistically build/tune yourself (yeah, fencing wire can make a high wattage low value resistor, and there are all sorts of mad HV cap designs for tesla coils etc, my favorite being glass jars and salt water, but they arnt exactly practical. If you have an LCR meter, which you would need to use a motor anyway, its trivial to add or remove windings on any generic inductor to 'dial it in'. another useful source of 'inductors' is a standard mains transformer, ive used them for ripple filters in valve amps, the value isnt critical and they can take the voltage.
I once designed a 12v fluorescent tube driver for camping that used a rod ferrite, adjusting the ferrite position with a screw adjusted the inductance and the amount of current drawn. this was pre-LED/solar panel era, and being able to adjust an 8w fluro down to about 2w was pretty nifty. full brightness for meal prep and such, but a dim background light so you dont trip over stuff while sitting around the fire was quite a feature. and it could run all night without killing your battery, unlike an incandescent. the only problem was you had to wind it up to full power to start the lamp, which was annoying if you had developed your 'night eyes', 8w can be quite blinding when your pupils are dinner plates.
reply
Ive never heard of a motor being used as an inductor. but thats because my favorite thing about using inductors in my designs is they are about the only electronic component you can realistically build/tune yourself (yeah, fencing wire can make a high wattage low value resistor, and there are all sorts of mad HV cap designs for tesla coils etc, my favorite being glass jars and salt water, but they arnt exactly practical. If you have an LCR meter, which you would need to use a motor anyway, its trivial to add or remove windings on any generic inductor to 'dial it in'. another useful source of 'inductors' is a standard mains transformer, ive used them for ripple filters in valve amps, the value isnt critical and they can take the voltage.
I once designed a 12v fluorescent tube driver for camping that used a rod ferrite, adjusting the ferrite position with a screw adjusted the inductance and the amount of current drawn. this was pre-LED/solar panel era, and being able to adjust an 8w fluro down to about 2w was pretty nifty. full brightness for meal prep and such, but a dim background light so you dont trip over stuff while sitting around the fire was quite a feature. and it could run all night without killing your battery, unlike an incandescent. the only problem was you had to wind it up to full power to start the lamp, which was annoying if you had developed your 'night eyes', 8w can be quite blinding when your pupils are dinner plates.
reply
InductorMan
An interesting commercial ) application of motor-as-inductor is the Tesla Roadster's battery charger. Yup, that's right, the 17kW internal 240VAC/70Arms battery charger of the Roadster uses the induction motor as the inductor when charging the traction battery. This was design content inherited from the AC Propulsion T0 (the IP from which the Roadster was derived, and is frankly silly in several ways.
The efficiency is pretty low, but more critically, it means that the entire motor needs to be double insulated from the vehicle chassis for electrical safety. The motor uses an insulative drive coupling, is mounted on insulated standoffs, and is, very unfortunately, air cooled through an insulating duct.
Anyone who has worked with cooling of high power density devices knows that direct air cooling REEAALLLY sucks. Blows Well, it's very not good. Water cooling with a separate radiator is massively better in most ways, with lower pumping power loss, more flexible packaging, etc etc.
But it did work, some several thousand of them were sold, it launched an EV empire, and catapulted a very (ahem) interesting person to world fame. So, in that one application at least, we can't really denigrate the motor-as-inductor.
reply
An interesting commercial ) application of motor-as-inductor is the Tesla Roadster's battery charger. Yup, that's right, the 17kW internal 240VAC/70Arms battery charger of the Roadster uses the induction motor as the inductor when charging the traction battery. This was design content inherited from the AC Propulsion T0 (the IP from which the Roadster was derived, and is frankly silly in several ways.
The efficiency is pretty low, but more critically, it means that the entire motor needs to be double insulated from the vehicle chassis for electrical safety. The motor uses an insulative drive coupling, is mounted on insulated standoffs, and is, very unfortunately, air cooled through an insulating duct.
Anyone who has worked with cooling of high power density devices knows that direct air cooling REEAALLLY sucks. Blows Well, it's very not good. Water cooling with a separate radiator is massively better in most ways, with lower pumping power loss, more flexible packaging, etc etc.
But it did work, some several thousand of them were sold, it launched an EV empire, and catapulted a very (ahem) interesting person to world fame. So, in that one application at least, we can't really denigrate the motor-as-inductor.
reply
Bicketybam68
In the late 80's, a friend owned a Mazda Capella here in Australia. While out and about one day, one of the fuses broke and we were stranded. Not wanting to wait for a tow-truck or catch a cab (the years before mobile phones) we would have to make a trek to the nearest phone booth and call for help and we both didn't want to do that. We were smokers (we know, horrible) and cigarette packets had a type of tinfoil paper inside around the smokes. Using a trick his uncle (a mechanic) had showed him, we ripped out that paper, folded it up around the old fuse and inserted that. It WORKED. We got the car started and got to drive home. Of course he was never going to keep it there, but it worked enough to get us home and again to get his car to his uncles the next day to fix it properly.
reply
In the late 80's, a friend owned a Mazda Capella here in Australia. While out and about one day, one of the fuses broke and we were stranded. Not wanting to wait for a tow-truck or catch a cab (the years before mobile phones) we would have to make a trek to the nearest phone booth and call for help and we both didn't want to do that. We were smokers (we know, horrible) and cigarette packets had a type of tinfoil paper inside around the smokes. Using a trick his uncle (a mechanic) had showed him, we ripped out that paper, folded it up around the old fuse and inserted that. It WORKED. We got the car started and got to drive home. Of course he was never going to keep it there, but it worked enough to get us home and again to get his car to his uncles the next day to fix it properly.
reply
bumbixp
Power supplies that are meant to operate in parallel have a larger voltage droop on purpose (i. e, the higher the current the lower the output voltage. It makes it easier to balance the load since the supply with a lower load would have a lower voltage it's force to take more load since the voltage must be the same where the supplies are connected together. It is inherently stable.
If the supplies have no droop at all then even an infinitesimally higher output voltage on one supply would cause it to take all the load (the voltage regulator will try to raise the voltage by the infinitesimal amount, which causes the load to rise until it has all the load and can finally raise the voltage. This is unstable.
reply
Power supplies that are meant to operate in parallel have a larger voltage droop on purpose (i. e, the higher the current the lower the output voltage. It makes it easier to balance the load since the supply with a lower load would have a lower voltage it's force to take more load since the voltage must be the same where the supplies are connected together. It is inherently stable.
If the supplies have no droop at all then even an infinitesimally higher output voltage on one supply would cause it to take all the load (the voltage regulator will try to raise the voltage by the infinitesimal amount, which causes the load to rise until it has all the load and can finally raise the voltage. This is unstable.
reply
dedr4m
Parallel PSUs: Worth noting is that a lot of server PSUs of the redundant types mainly do one of two things where either by feedback performs current sharing actively, or by dropping the voltage at max load so the next PSU can take over, however it has to be designed to do so. Useful when I want to combine a couple 350W to get 700W (I stick to no more than 600W usable with 50W margin. Another way is to have a linear slope proportional to current so that as more current is drawn the less voltage on the output so that the second PSU can begin taking over, though I've not seen that type that often.
I haven't tried mixing different server PSU current sharing methods as I'm not sure if there'll be fireworks.
reply
Parallel PSUs: Worth noting is that a lot of server PSUs of the redundant types mainly do one of two things where either by feedback performs current sharing actively, or by dropping the voltage at max load so the next PSU can take over, however it has to be designed to do so. Useful when I want to combine a couple 350W to get 700W (I stick to no more than 600W usable with 50W margin. Another way is to have a linear slope proportional to current so that as more current is drawn the less voltage on the output so that the second PSU can begin taking over, though I've not seen that type that often.
I haven't tried mixing different server PSU current sharing methods as I'm not sure if there'll be fireworks.
reply
Daniel-k8y1r
I used to design brushless DC motors for an OEM. You hit the nail on the head with your mention that inductors are meant to be energy reservoirs while motor windings are not. The motor laminations are made from electrical steel, alloys which typically contain silicone and used for motors and transformers. When designing a motor, you use these alloys because they build and release the magnetic flux induced by the coils very quickly. You don't want the laminations storing magnetic energy, if the laminations are still holding that energy when the coils switch you're going to have a harder time inducing the new field and reduce the magnetic flux you can build up before the coils have to switch again.
reply
I used to design brushless DC motors for an OEM. You hit the nail on the head with your mention that inductors are meant to be energy reservoirs while motor windings are not. The motor laminations are made from electrical steel, alloys which typically contain silicone and used for motors and transformers. When designing a motor, you use these alloys because they build and release the magnetic flux induced by the coils very quickly. You don't want the laminations storing magnetic energy, if the laminations are still holding that energy when the coils switch you're going to have a harder time inducing the new field and reduce the magnetic flux you can build up before the coils have to switch again.
reply
TheRailroad99
Be careful with the parallel PSUs. While it worked with your examples, it can easily lead to a cloud of magic smoke.
Especially with higher wattage supplies (like ATX PSUs) and especially SMPS which have a tendency to oscillate, this is a very bad idea.
Two LDOs in parallel are most likely fine (but even those can oscillate, two SMPS most likely will oscillate.
Source: I build and program test equipment. I have seen lots of circuits, and especially lots of bad circuits. Most of them work fine, but looking at the statistics after 10000 DUTs tells very quickly which circuits are designed robustly with manufacturing tolerances in mind, and which ones just barely work in theory
reply
Be careful with the parallel PSUs. While it worked with your examples, it can easily lead to a cloud of magic smoke.
Especially with higher wattage supplies (like ATX PSUs) and especially SMPS which have a tendency to oscillate, this is a very bad idea.
Two LDOs in parallel are most likely fine (but even those can oscillate, two SMPS most likely will oscillate.
Source: I build and program test equipment. I have seen lots of circuits, and especially lots of bad circuits. Most of them work fine, but looking at the statistics after 10000 DUTs tells very quickly which circuits are designed robustly with manufacturing tolerances in mind, and which ones just barely work in theory
reply
barryobrien1890
For the fuse you are only supposed to put a single layer and hold it in with the glass element. Power line communication actually uses pretty complex modulation and error correction, so it's only good for digital data, and completely fails if you plug a cheap Chinese phone charger anywhere in the house. Finally APC actually modulates the H bridge on the inverter transformer as an inductor (using leakage inductance) to boost the voltage and charge the UPS battery. The down side is it gets really hot modulating a steel transformer at khz. It's great if you don't mind cooking the battery and buying a new one after the warranty runs out
reply
For the fuse you are only supposed to put a single layer and hold it in with the glass element. Power line communication actually uses pretty complex modulation and error correction, so it's only good for digital data, and completely fails if you plug a cheap Chinese phone charger anywhere in the house. Finally APC actually modulates the H bridge on the inverter transformer as an inductor (using leakage inductance) to boost the voltage and charge the UPS battery. The down side is it gets really hot modulating a steel transformer at khz. It's great if you don't mind cooking the battery and buying a new one after the warranty runs out
reply
ve2um
Rather than tin foil, i use individual strands of electric wire or small enameled magnet wire that, before use, i calibrate using a CC power supply with two values: Hold current that the wire can support before releasing smoke (i physically touch the wire before test so the smoe is skin oil, and obviously Blow current.
Useful for testing and on low voltage circuits. Do not use on mains.
Inspired by ooold E. F. Johnson 12v mobile VHF rigs that used a classic 3AG fuse on the radio's rear panel and an hidden thin wire fuse, inside the radio, in series with the main fuse for protection against guys that replaced the 3AG by a tin foil fuse.
reply
Rather than tin foil, i use individual strands of electric wire or small enameled magnet wire that, before use, i calibrate using a CC power supply with two values: Hold current that the wire can support before releasing smoke (i physically touch the wire before test so the smoe is skin oil, and obviously Blow current.
Useful for testing and on low voltage circuits. Do not use on mains.
Inspired by ooold E. F. Johnson 12v mobile VHF rigs that used a classic 3AG fuse on the radio's rear panel and an hidden thin wire fuse, inside the radio, in series with the main fuse for protection against guys that replaced the 3AG by a tin foil fuse.
reply
lucaboschian4205
Not only powerlines can be used for sending signals (like the adaptors for ethernet) but radio experimenters used mains aerial antennas (here in Italy they were called tappo luce. The idea behind it is to use the house wiring as a massive receiving antenna. Putting a high voltage capacitor of correct capacitance, you can isolate the radio signals (kHz-MHz range) from the mains voltage (50-60 Hz.
To be fair, basically anything was used as antenna. I personally heard of metal bed frames, pipes (older ones were very commonly made of copper, even cranes at construction sites.
reply
Not only powerlines can be used for sending signals (like the adaptors for ethernet) but radio experimenters used mains aerial antennas (here in Italy they were called tappo luce. The idea behind it is to use the house wiring as a massive receiving antenna. Putting a high voltage capacitor of correct capacitance, you can isolate the radio signals (kHz-MHz range) from the mains voltage (50-60 Hz.
To be fair, basically anything was used as antenna. I personally heard of metal bed frames, pipes (older ones were very commonly made of copper, even cranes at construction sites.
reply
nastyscaper
You sent
Hello Scott!
Loved the new video. I was wondering if you're planning to turn it into a series. If so, here’s a myth for you:
You can transmit data (like Ethernet, something that is high speed) through any connector, as long as the connections are solid. For example, Ethernet can be passed through both an RJ45 and a DSUB connector.
Some students in my robotics club keep saying that and say we tested it, it works but i keep explaining if you look at it through an eye diagram, you'll see the signal get affected making this not usable. Thanks!
reply
You sent
Hello Scott!
Loved the new video. I was wondering if you're planning to turn it into a series. If so, here’s a myth for you:
You can transmit data (like Ethernet, something that is high speed) through any connector, as long as the connections are solid. For example, Ethernet can be passed through both an RJ45 and a DSUB connector.
Some students in my robotics club keep saying that and say we tested it, it works but i keep explaining if you look at it through an eye diagram, you'll see the signal get affected making this not usable. Thanks!
reply
greatscott
professional servers usually have 2 power supplies that run parallel. And since they are sized to handle the load alone the issues you mentioned simply don't matter.
But they do go into alert state if one fails anyways, because that means there is no more redundancy. The alert is loud enough so you can easily find them in a loud data center.
The one I have at home has 2 1000w power supplies. But to even draw anywhere near that you would need some pretty beefy accelerator cards in there in addition to the 12 hard drives 2 CPUs and 2-24 RAM modules.
reply
professional servers usually have 2 power supplies that run parallel. And since they are sized to handle the load alone the issues you mentioned simply don't matter.
But they do go into alert state if one fails anyways, because that means there is no more redundancy. The alert is loud enough so you can easily find them in a loud data center.
The one I have at home has 2 1000w power supplies. But to even draw anywhere near that you would need some pretty beefy accelerator cards in there in addition to the 12 hard drives 2 CPUs and 2-24 RAM modules.
reply
rewolff2
That motor is built for say 50A. So the puny 1A of your DCDC will NOT saturate it.
The motor is expected to run at say 10000 RPM. Or 166 rev per sec. With say 4 ERPM/RPM that comes to 666 Hz on the stator.
Your DCDC is running at a much hiigher frequency. The stator is suitable for frequencies up to say 1000Hz or maybe a bit more (if meant to run in BLDC mode as opposed to FOC/sinusoidal. So my guess is that that's the main cause for the efficiency drop: You're losing energy in the stator due to the higher frequency.
reply
That motor is built for say 50A. So the puny 1A of your DCDC will NOT saturate it.
The motor is expected to run at say 10000 RPM. Or 166 rev per sec. With say 4 ERPM/RPM that comes to 666 Hz on the stator.
Your DCDC is running at a much hiigher frequency. The stator is suitable for frequencies up to say 1000Hz or maybe a bit more (if meant to run in BLDC mode as opposed to FOC/sinusoidal. So my guess is that that's the main cause for the efficiency drop: You're losing energy in the stator due to the higher frequency.
reply
slipspectrum9253
LOL I honest to god have fixed blown fuses in my car using the paper/foil from a cigarette pack. The point is NOT to work as a fuse, but rather to bypass the fuse so I could still use my wipers. In car electronics, it’s often not a hard fault, but some transient condition like an iced over windshield or some such thing, that popped a fuse. Today, cars have PTC fuses and other more graceful ways of dealing with transient faults, but back in the day of glass fuses and fusible links, you had to do what you had to do!
reply
LOL I honest to god have fixed blown fuses in my car using the paper/foil from a cigarette pack. The point is NOT to work as a fuse, but rather to bypass the fuse so I could still use my wipers. In car electronics, it’s often not a hard fault, but some transient condition like an iced over windshield or some such thing, that popped a fuse. Today, cars have PTC fuses and other more graceful ways of dealing with transient faults, but back in the day of glass fuses and fusible links, you had to do what you had to do!
reply
AC9BXEric
Audio on the mains power is not practical. The power frequency is in the audio range. There's also some voltage leakage concerns if the capacitor gets bigger, so there's a limit to how low audio frequencies you can use. But above about 50kHz (or 60k if you're targeting 1000 times line power) things get better. Then there's the transformer which gets problematic as frequency increases. Carrier current radio however in the AM broadcast band works quite well. 600kHz for example on top of 60Hz power is a good fit.
reply
Audio on the mains power is not practical. The power frequency is in the audio range. There's also some voltage leakage concerns if the capacitor gets bigger, so there's a limit to how low audio frequencies you can use. But above about 50kHz (or 60k if you're targeting 1000 times line power) things get better. Then there's the transformer which gets problematic as frequency increases. Carrier current radio however in the AM broadcast band works quite well. 600kHz for example on top of 60Hz power is a good fit.
reply
southernflatland
Back in the 90's, my late father had a problem with the distributor in his truck while we were at the store, miles away from home. The carbon center conductor had apparently cracked or burned up or something.
So what did my dad do He just pulled the foil paper out of his cigarette pack and wadded it up into a cylinder shape and stuck that in place where the carbon conductor went.
He drove that thing around like that for at least 6 months, probably longer than that knowing him haha!
reply
Back in the 90's, my late father had a problem with the distributor in his truck while we were at the store, miles away from home. The carbon center conductor had apparently cracked or burned up or something.
So what did my dad do He just pulled the foil paper out of his cigarette pack and wadded it up into a cylinder shape and stuck that in place where the carbon conductor went.
He drove that thing around like that for at least 6 months, probably longer than that knowing him haha!
reply
Geekeh_UK
I always thought the tinfoil thing was that you were supposed to wrap it around the outside of the blown fuse. But then I may just be thinking of watching MacGyver do it all the time, and I think he used to use a gum wrapper.
Either way I don't think the idea was that it was supposed to serve as an actual, appropriately valued replacement fuse, but rather that you'd pray the house didn't burn down, until you could source an appropriate replacement the next day or whatever.
reply
I always thought the tinfoil thing was that you were supposed to wrap it around the outside of the blown fuse. But then I may just be thinking of watching MacGyver do it all the time, and I think he used to use a gum wrapper.
Either way I don't think the idea was that it was supposed to serve as an actual, appropriately valued replacement fuse, but rather that you'd pray the house didn't burn down, until you could source an appropriate replacement the next day or whatever.
reply
db1tau75
Power Lines dampen different frequencies to different levels. First of course you cannot use the 50Hz of the AC Power, you have to start much higher. High datarates need high frequencies, so some modems measure the frequency response of the line and adjust the pre-amplifier and filters accordingly. If you play around with an SDR (Hack RF one or RTL SDR) you want to ban all PLC from your house because it causes a lot of RF emissions.
reply
Power Lines dampen different frequencies to different levels. First of course you cannot use the 50Hz of the AC Power, you have to start much higher. High datarates need high frequencies, so some modems measure the frequency response of the line and adjust the pre-amplifier and filters accordingly. If you play around with an SDR (Hack RF one or RTL SDR) you want to ban all PLC from your house because it causes a lot of RF emissions.
reply
rachid746
might be my lame electronic/electric knowledge but I have a question regarding the powerline communication system: if you broadcast your internet connection through the power line, what stops it from being read by anyone on the same network (let's say, in your neighborhood) if they connect their own devices to read signals from the powerline Don't be too harsh on me, I'm just a curious noob: )
reply
might be my lame electronic/electric knowledge but I have a question regarding the powerline communication system: if you broadcast your internet connection through the power line, what stops it from being read by anyone on the same network (let's say, in your neighborhood) if they connect their own devices to read signals from the powerline Don't be too harsh on me, I'm just a curious noob: )
reply
Add a review, comment
Other channel videos
