That's really interesting, you're right we just focus on electronically doing everything now. It has become a blind spot now. Nice bit of outside the box thinking!

As an electro engineer: Rotating converters was the business in the old days. You still need somehow to regulate the rpm of the converter. Sorry, this is heavy, bulky and the carbon brushes for heavy loads? They got a 2V voltage drop, each! This is a efficiency nightmare with very limited use everywhere. You could rather use dc brushed motor only and easy regulate the rpm. If you realy got heavy loads on a budget, you could go with the silicon solution with used parts and knowhow.

@j.goebbels2134

2 жыл бұрын

I am trying very hard to source a controller for a Motenergy me-1306, 24-70V, 12KW motor. I refuse to believe that the only thing that can drive it is a $500 controller shrouded in mystery by Kelly, which people say can still burn up after a few minutes. Please help if you can.

Love it. Waiting for part 2. Makes me think about the fact that many power tools are now going bushless in favor of commutator and brushes.

@ThinkingandTinkering

3 жыл бұрын

tomorrow mate and yes they are but then they are still quite low power

Really cool. I've been thinking for a long time how to mechanically switch on and off motor phases. I always liked carburetors over annoying electronic fuel injection. This is a great thing and I think it's possible to drive the commutation from the motor itself. Will need some kind of a kick start though maybe. Great stuff, thanks for sharing this.

@FBPrepping

2 жыл бұрын

Fuel injection was developed to improve only one thing: manufacturers´ profits.

@petedude2lu3

Жыл бұрын

have you ever met an annoying carburetor?

A big advantage of an electronic solution is feedback - either by measuring back EMF, measuring the current in two or more phases, or with an encoder. Feedback is essential for moving BLDC motors at slow speeds with high torque like in electric vehicles. It will be interesting to see your approach to this problem in a mechanical way, keep up the good work!

@Buzzhumma

3 жыл бұрын

Yeah the variable load is defo a problem but some genius out the box thinking may solve it . Maybe silicon is a solution to that problem if its just feedback ?

You've just helped solve a problem that I've been working on for quite awhile. Thank you.

@ThinkingandTinkering

3 жыл бұрын

Glad to help mate

Good idea Robert. Your rotary controller will need to handle a lot of current so it will need to be large. I connected two identical stepper motors together wire for wire and found I could generate enough current by twisting the shaft of one motor to drive the other motor. No external power supply other than my own twisting force from my hand. I had to get the speed just right, fast enough so I generated enough power but slow enough to allow the second motor time to react.

what a fascinating concept , i love electronics but a good scientist should look at many possible solutions to a problem, lets hope there is more of this !

Finally, a bloke that makes sense, this video saved my sanity, if I hear more "lost in flux" nonsense from some dummy i will totally lose it! Thank you so much for being an engineer-level technical man that knows how to communicate.... cheers! MarkMannM2

This is why I follow this channel, brilliant Robert...

After having two dc to ac inverters for my solar setup explode on me I’m also going the rotary converter way. To bad if it’s less efficient, it’s repairable! When a silicon inverter dies it takes out all the electronics upstream in the inverter and downstream! By adding a flywheel the rotary converter can cope with capacitor start devices better than an electronic inverter!

Youre partly right. In many cases relays and contacts are used to handle high power output. This is still the case. It's more robust, is forgiving when mishandled, has low loses and is cheap compared to electronics for the same current. However for many applications it's just not possible to use contacts. I don't have experience with electric car motors but i do have experience with stepper motors. These need a constant current for the RPM range the system is designed for. If that is a high RPM you need a high voltage to maintain current while winding impedance is going up at high rpm but low voltage when RPM is low. This would cause huge losses in an analog system. To avoid that PWM is used to control average voltage. maybe 8 to 500 kHz. Drive systems for trains are simular. For some you can actually hear the PWM when traveling by railway. Car's won't be much different. Thats very hard to do with any kind of mechanical system. Both high frequency and pwm current control. Bottom line : You need electronics for such applications.

Another cost effective brushless motor sweet spot is around 1HP 750 w for electric bicycles....millions produced....around $15 USD on ebay Has lots of features like regen braking if required, drives low cost display....As u say big cost jump for water cooled full size vehicle electronics

@ThinkingandTinkering

3 жыл бұрын

it is what I was thinking about mate

Thank you! I'm refitting my 4Runner with an electric motor, the ICE having seized after the oil pump failed. I'll be installing a 30hp to 50hp motor as funds become available. I'm looking at a similar control with mine, only a little different. As with the logarithmic dominos I'm controlling a big motor with smaller devices. I may encase my controller in sulpherhexafluoride... Oh I have a 5hp electronic controller and your conversion blends with that perfectly.

Well done, it's good to question everything and keep the discussion going

One might draw some inspiration from "rotory converters". They were used in the past to mechanically turn DC to AC, or AC to DC, and can be used with poly phases

@ThinkingandTinkering

3 жыл бұрын

nice one mate - cheers

Amazing! Looking foward to part 2. Im wondering whether this could be used in conjunction with relays somehow, but yet again thats swaying towards an electronic solution. Thank you for a very informative video, might try building my own.

@ThinkingandTinkering

3 жыл бұрын

part 2 gives build details mate and that will be up tomorrow

The rotation of the mechanism is not the same as the motor. You have to divide the rotation of the controller by the number of windings in the motor divided by the number of phases. To run a 3-phase motor, with 24 coils, at 5000 rpm, the controller would have to run at 40,000 rpm. It's a good idea, but in all practicality, the switching mechanism would have to be internal to the motor. Then you just need a PWM to control the speed, and maybe a parallel ceramic capacitor on each coil to smooth the pulses and limit the back emf. The capacitor would also work like that Constantinescu torque converter that Robert built, and it would help with startup. I love it when a video gives me ideas. 😂😘

Robert, We hate it. Show it to us please! LOL 🤣. Not bad! How come no one ever made a bike car? Like you spin the commutator with pedals and it just goes.

Great reasoning. Thanks for sharing. Love all your videos.

This is well worth further explorations. A+ As usual ROB.

Pretty darn cool...thanks for sharing!

@ThinkingandTinkering

3 жыл бұрын

cheers mate

so basically you can use low inexpensive brushless mottor with cheap esc (5 EUR of costs) with this mechanical controller - and voila you have electromechanical controller for big motors for cents :)

Tha's a great solution, Rob. Real "outside the box" thinking.

I like your idea! Brilliant!

I beyond love this. I'll be making an electric Dune buggy soon, so now seeing this I know it's going to be elecreo-steampunk themed!

@paulmaydaynight9925

3 жыл бұрын

cycloidal gearbox or magnetic ^_~

@lubbock2704

3 жыл бұрын

@@paulmaydaynight9925 I'm thinking magnetic but I'll try to add in electromagnets and see if I can get some extra functionality

RMS, and the inverters on the electric cars also have to have extensive cooling systems to protect the electronics. Mechanical contactors are cheap and offer high current control. I totally agree with you assertions.

@ThinkingandTinkering

3 жыл бұрын

cheers mate good to know

I seeeeeee Dishwasher parts. I'm already excited. That heater or the right is one of my biggest bug bearers with Bosch

@ThinkingandTinkering

3 жыл бұрын

lol - it was the one you sent me mate - turns out it is a brushless dc motor! - who would have thought - cheers

@karlmyers6518

3 жыл бұрын

@@ThinkingandTinkering I have probably 100 of them . I've been saving them up along with the drain pumps. I couldn't get them to run on my dc power supply though.

@ThinkingandTinkering

3 жыл бұрын

well you can now lol

Great start, I have had the same thought myself, the relatively low amperage of reasonable cost speed controllers has stopped me making an electric car conversion. I thought it would be good to have a mechanical smoother sine type wave somehow? I note the motors driven seem a bit jumpy.

Great Philosophy here! Awesome!

You can also use that for a DC to AC inverter. Don’t know if you realize that yet

@ThinkingandTinkering

3 жыл бұрын

I didn't mate - nice one - cheers

@Buzzhumma

3 жыл бұрын

Yeah a perfect sinewave !

Totally thumbs up. thank you for this video. Two thing this mechanical commutator is missing, positioning sensing and limiting the voltage/current. 1) Strange motor vibrations will happen if the rotor's rotation can not match the stator's field rotation. Maybe this issue can be managed by avoiding trying to accelerate too fast, 2) Not controlling voltage at low RPM will lead to excessive currents at slow speed even under light loads. electronic speed controller use PWM and inductance to control the voltage and current, but I'm not seeing an easy way to mechanically control the voltage/current at low low rpm. Anyone have ideas?

@ThinkingandTinkering

3 жыл бұрын

cheers mate and to be honest this idea - in it's basic form - is as old as the hills - it's really just replacing the commutator -I am sure solutions exist

@kreynolds1123

3 жыл бұрын

@@ThinkingandTinkering I love the idea. Just identifing potential issues and trying to come up with solutions 😁 Here is Some out of the box thinking. Maybe if the commutator had capacitors on its rotor between input and output. Then with a limited charge stored on each cap, the capacitors provide a higher average impedence at low switching/commutated freqency, while at higher freqencies the capacitors provide a low average impedence path. The capacitors would need to be appropriately sized for the application.

@ThinkingandTinkering

3 жыл бұрын

I like that idea mate - nice share - cheers

this is really good im probably going to try this, the only down side is is there going to be arcing in the commutator when you put huge power through the instant cut of huge power causes a backup like in points might need so mitigation

There was a very serious inventor from NZ that was making motor/ generators who when showed a "replication" of his machine that had FETs he told the guy 'stick to mechanical switching'...Robert Adams talking to JL Naudin

Awesome thanks for the chosen article. Nickxon on your right

I remember seeing a mechanical controller on a golf cart an basically it was a dial with a negative and a number of positive terminals at certain increments that went 12,24,36,48v to increase/decrease speed

Sure, SiC gets more expensive as you begin to increase the power demands, that is a given. While a mechanical commutator will be cheaper (somewhat) compared to the 'same capability' SiC solution, the reason its not used is the fact that it is so inefficient compared to a 'proper' brushless driver when you begin to scale up to large systems. Not only will the friction and resistance of the brushes reduce efficiency, but the fact that you can't feed a proper Sin or Trapezoidal waveform to the coils will inhibit the efficiency, and max speed/torque of the motor. Block commutation, which is the most rudimentary form of driving a BLDC motor, and used in 95% of cheap hobby ESCs, is still more efficient at driving brushless motors considering it is able to 'step' the voltage applied to coils vs a brushed system which is full on-off. When you compare this to Sin, Trapezoidal, and even the holy-grail FOC drive, brushed systems clearly loose in nearly every regard. The cost of a proper brushless controller only looks bad when you think you can drive the same motor with a brushed system and still hit similar performance targets. If Brushed systems were a viable solution, they would far more prevalent in industry, as you'd bet they would do whatever they could to lower the cost. Not to mention computerized control lets you tweak drive parameters on-the-fly to tune for performance, efficiency, etc.. A brushed system isn't nearly as configurable. Same reason why DSP(Digital Signal Processing) has largely replaced Analog filters and systems in 95% of applications.

Interesting idea. We could apply this for steppers, but timing would be for example 1/4 turn is 1 step, so we get simple manual precision control without any electronics.

Agree about low power handling capabilities of semiconductor devices. Have yo heard about Ignitrons? How bout triggered spark gup? Also we are using traveling wave tubes even today and there are some immensely powerful Radio tube used today.

I am over joyed with this. I have a use for it and need to gather the other parts but you will here from me again. Limited funds will slow me down but it will happen. Thank you so much.

A fine example of higher education leading to alternatives rather than improvements, although possibly an improvement for some people.

congratulations! this is what I have been trying to get folks interested in for 40 years, and the flack from it has been monstrous. now I dont feel so much and out sider knowing that you Robert, thought so too. I also have for those that that are seeking, a what i call three phase Sudo parametric rotary speed controller that is mechanical and I feel repairable for very little with just a bit of time and scrap ironically driven with a variable electric drill ( you have no idea how much I smiled when you busted out the cordless variable speed electric drill) good show 👍 that rotary switch in the original black and white Frankenstein movie got me started thinking about it when I was a kid, do you know the one from the movie that was big and round and had sparking around its perimeter and made a very pronounced clacking noise with sparks flying out of it as each one of the switches fired has the internals rotated, I built on that idea, Incorporating a rolling contact concept. for longevity and build ability

@ThinkingandTinkering

3 жыл бұрын

mate - have you done some videos on this - if not - why not lol - love the idea myself

@MrAnderson4509

3 жыл бұрын

no have not done any vids, but you caught on fast, thought you might, we seem to be wired similarly and you want bypass the.... i am being redundant. due to down sizing and some health things my shop and gear is suspended in the perfect storm so to speak, I would love to have a channel going wanted to do that since about 2004, but I had allot of interference and allot of life happen, If I were down the block from you, i am quite positive we would play together and cause perhaps some unrest and heresy with proof of concept in the Scientific, DIY, and commercial industry while building our play things. some of the controler ideas we're actually blocked in the DIY maker space community is if you can believe that, it was incredibly important to them That no electro mechanical concept proof be made , And that everything should be based on an arduino, don't get me wrong I Love the Arduino, but microcontrollers Are not new to me, 8080A, 6502, 6800, 68000, and have their place, it's just that I meet resistance when I Carry myself with the concept that the time the technology was created Little Bearing on the practicality it offers to achieve a goal within a construct. whatever gets me there within my means and ethics.

@j.goebbels2134

2 жыл бұрын

what are my controller options to drive a Motenergy me-1306, 24-70V, 12KW motor?

@TheSorters

11 ай бұрын

Could a mechanical commutator also be used to create regenerative braking or for power generation? I am trying a hybrid drone generator, working with over 100 amps, and the electronic speed controllers are prone to failure at those high power levels. But, goal is to get power from BLDC driven by engine

Excellent, I like the idea of rejuvenating past electro mechanical system.

Maybe a very similar idea but even easier is to have magnets on the rotor and a set of reed switches on the stator? Still no brushes but a way to control the phase switching of a bigger motor. Plus speed control.

Hey thanks Robert👍

Very nice Robert agree mechanical is the way to go for high power applications this is a really great concept would like to see more can you make a single phase to 3 phase commutator to drive a 3 phase motor iv seen it done very crudely before but iv really never seen and nicely built version cheers :)

I have been thinking of this for a while. I was thinking about using magnetic switches to control relays that would drive motor. Not sure this would save power, but I think it would be very robust. You could drive very large motors for very little cost and still have no brushes.

Great solution to deal with a problem for those of us trying to operate on a budget. Do you have an idea to deal with the motor chatter when it first starts? I happens on silicon controller brushless as well. I see it a lot in brushless RC car applications.

Just goes to show that the old ways still have their uses. Electronics are fun but as you said they have limits before prices skyrocket.

@ThinkingandTinkering

3 жыл бұрын

for sure mate

@paulmaydaynight9925

3 жыл бұрын

@@ThinkingandTinkering thought.. they always say "you Can reduce eddy current effects by restricting paths for current flow:" i say what shape and placement of capacitors do you need to make to actually capture and store these unused localised (ringing?) high currents ... and so presumably lessen the unwanted heating effects on the (new capacitors in disguise) laminates as a side benefit, some uni math required for optimising the things i assume

That's awesome! I just have a good conditions washing machine motor in my garage

@ThinkingandTinkering

3 жыл бұрын

That is awesome too - lol!

Will be interesting to see how it copes with a very large inductive load? I expect to see a lot of arcing as it switches. I wait with anticipation on the results.

@ThinkingandTinkering

3 жыл бұрын

isn't a motor an inductive load? this is just mechanical commutation - pretty much standard for DC motors isn't?

@TheEmbeddedHobbyist

3 жыл бұрын

@@ThinkingandTinkering yes it is, but the car motor is a very large inductive load. It's normal to see brush arcing in DC motors but I've never seen a car motor being driven that way. Also as carbon is resistive, what sort of self heating do you expect to see? Hi current DC is a bigger to switch due to the arc's not wanting to stop by themselves. I've had large contactors weld their contacts closed when switching a large solenoid, fine when turning the solenoid on, but the arc when breaking was the cause of the problem.

@ThinkingandTinkering

3 жыл бұрын

I don't disagree agree mate - but large DC motors are going to be the place to look for solutions I would think

Mr. Smith is spot-on. I work in the utilities industry and large motors are controlled via frequency and use AC to avoid the DC brushless limitations described here. DC Brushless would be wonderful for example as backup water delivery directly from battery resources, as AC/DC inverters exhibit the same silicon limitations and expense.

@ThinkingandTinkering

3 жыл бұрын

thank you mate - I feel better about what I said with confirmation from a 'guy in the know' - cheers

What you built used to be called a motor/generator resolver. It was used in radar systems to accumulate azimuth sin/cosine values. It would seem it is necessary to reinvent the wheel as it were to accommodate what Tesla tried to tell us 100 years ago.

This is total madness, doesn't make any sense, and solves no problem but its still super bad ass.

This is more similar to a plug-in hybrid drivetrain such as the Prius synergy drive as it needs an external power train to turn at all than one that you can just connect a battery to and twist for a power delivery. A more effective way to make a non digital controller is to pair a potentiometer with plenty of mosfets and you got an analog controller which is my favorite alternative over the one that is popular in modern electric vehicles that we got now.

Good work. I was introduced to the concept of phase-locked A/C motors as control systems while training as a technician in the 1970s. I seldom saw them working in the field. Often wondered why the idea was not applied to e.g. brushess or stepper motors. Really cool! Suggest that a small, motorcycle-equivalent, lead acid battery in circuit would make your converted alternator work quite well.

There can a real advantage of avoiding electronics for some tasks. Reed switches would also serve well if they can be found with the necessary power rating

@ThinkingandTinkering

3 жыл бұрын

for sur mate

Nice, I am an expierimenting individual, like a mad dog with rabies I rack my mind and body building and creating something different and new, some of those self taught things had me performing live on stage all over my city. I have created 2 different things that have a hypothetic effect on people. But such things took years of messing around with electronics, chemical reactions, and bringing my visionary ideas to something physical that I can put in front of someone else...

Genius just genius Sir

Hi robert, great vid as usual, i went through this idea a while ago, i came up with a flat disc divided into quadrants(like a pie chart) but with a wide border to each quad, the dividing line remains a straight line but as you approach the edge of the disc, the relationship between contactor(ON) and the insulator strip(off) changes, the further out you go the more on time you get, but the off time always remains the same,, much like a pwm, still only a crude square wave generator, but variable (by sliding the contact further in or out on the disc surface), i was thinking of perhaps a solid coper disc with enamelled insulator strips. its much like the old distributor in an ignition system, the contacts in which can last a long time, although that is for high voltage not high amps.

@stevelloyd5785

2 жыл бұрын

In the distributor the high voltage arcs from rotor to pole without contact so not really like what you described. Also in a DC motor commutator the brushes are wider than the gap between segments and therefore are briefly in contact with two contact as the transition from one to the other occurs and therefore there is very little wear as that all happens. Your mechanical disc psuedo PWM must break the circuit and there will be considerable arcing and probably physical wear as the edges of the conductor (or insulator) chisel away at the brushes

For smooth acceleration sensors are needed for correct phase control, mechanically linking the smaller motor to the bigger motor may work if they are designed to match each other? Also the direction the motor spins can be random without sensing the back emf. Just pointing out potential issues. A brushed motor controlled by voltage works very well whilst the brushes are good but needs maintenance and of course the voltage needs to be controlled efficiently. The biggest motor I have is 1000w ebike dd, to upgrade the controller for serious power is expensive and also they are prone to shorting out fets if there is an issue with timing of the switching. A simple mechanical solution with easily replaced parts would be interesting.

@ThinkingandTinkering

3 жыл бұрын

they are good thoughts mate and as I said this is really proof of concept level - but I am pondering the issues you raise for sure

also on electronic control at high cost due to power, I am wondering if a better material would work to make this system work, so have you guys tried using the materials you use to make your conductive paint and such to make semiconductors yet? (other than the graphitic carbon nitride) and make your own semiconductor components (I made a diode with liquid soap and cola soda of sugarless variety like diet coke with aspartame instead of sugar, by placing a drop of each on glass plate next to each other so the edges of the drops touched, forming a single drop with two halves basically, and then did tests using a multitester on the resistance mode, and found it both had a diode effect and when first making contact in the reverse direction acts a bit like a capacitor till it gets charged up then stops and blocks the flow, while the other way you never see that change in resistance and stopping flow, though not sure if higher voltage or amps would cause a salt exchange like how two electrodes in a tank of salt dissolved in water will cause the chlorine to cluster around one electrode while the other attracts the sodium ions so that you get chlorine gas that evolves out of the tank, which would ruin its specific chemical nature of course, and something similar could be going on in the phosphate based compounds in the cola to cause the phosphorus to migrate to the soap (phosphate free alternative most likely since everything is like that now and has been for years) so once those phosphate ions are dispersed evenly it just becomes a resistor and no longer that diode effect or something, BUT if this were dispersed in say carbon lattice and baked once done in a kiln it might allow a higher voltage tolerant (and higher temperature tolerant) semiconductor that with no liquids present would allow it to continue to work all from home made tech that costs pennies instead of 100s or even 1000s of dollars or pounds as the silicon alternatives you have looked up have cost, and expand the range of manufacturing you have at home since while the mechanical may be good enough and far better than silicon derivative, in this case, what about other instances? I have a couple tasks and projects that will take about a week before I can start experimenting but I have had some decent success with the geopolymer work (baked some at about 400C that was basically red clay and then tested the bits of one thatt exploded LMAO and while it is exuding salts probably from trapped citric acetic or oxalic acids used to alter the clay, it is rock solid still and only darkened from absorbing water, over night but was still resistant to emulsifying or what ever it is that clay does to become plastic and malleable when you get it wet thus making it suitable as a say roofing tile or brick that wont go all soft and runny when it rains LOL all with out having to hit the normal higher temps (also put in a second bit that was the same with one exception of having some baking soda added and that one did not break, still have to test to see if it was an air pocket, or the surface being abraded to get flat faces on it vs the other being smoothed down with a slick fingernail to get an almost glaze smooth surface as it dried that might have sealed the interior too well or something, with another batch of tests (was also a test to see if the angle iron legs and 3/16th inch thick plate steel bed table I have is enough to insulate the kiln from damaging say a wooden or plastic tile floor the feet are resting on, or even carpeted surface, and have to say the 2 hours or so I ran it all the way up even to 1000F was still not enough to allow the outside to get hot enough to be too hot to touch or even uncomfortable to touch, so... have to do another 8 hour or 10 hour run I guess LOL anyway.) anyway going to try a couple things to see about making more electrically conductive semiconductor materials that also have as low of thermal conductivity as possible for TEG building, and improvement with the same tech and a few others combined together, and thought this might be something later to do a video on to continue on your content creation. (not that it seems like you have a lack of imagination on things to work on LOL)

Rob Sir This is a very great idea and congratulations on the Build of a mechanical controller for a BLDC motor, Kudos !! Rob Sir the standing questions remains in my mind are 2, 1) We need to have a motor to drive the mech unit which will drive BLDC motor, what will control that motor ? or can it be a Diesel Electric design?!! 2) do we need to rotate the controller to a desired RPM of motor ? How would it control RPM of BLDC? in the video I see you driving that unit with a electric hand drill I could see the RPM on motor was high the RPM on shaft of the converted alternator motor were much slow in comparison is it because of power supply or we need to drive the Mech unit at double speed to get desired RPM of the BLDC? Strange things in my mind but yes I admit they did trouble me a night !! Love and very warm regards from India Rob Sir!! Great Video!!

This is exactly what I'm looking to do, I have a LG 3 phase 22,000 ac window unit I'd like to run off solar. I went down the rabbit hole of diy esc to go from dc to 3 phase ac but find I don't have the technical knowledge to build such a thing for the power requirements of such a high amp draw application. I have the same unit that runs off 220v ac so this build will be supplemental cooling to lesson the load, not be the primary source of cooling my shop. I got the unit for $100 so no big deal if I melt it down ;) The unit will only work during the day of course but hey, if I can get it to work I can get near free to me cooling. On another note, I wonder why this setup isn't used in place of an inverter. Solar ==> 3 phase dc motor (with mechanical commutation) ==> single or 3 phase generator. I'm going to give it a go.

Sir, you speak my language...silicon is too weak for high power stuff and I'm all about power plus I love stuff that's simple to operate without the programmings and electronic tinkerings

Great idea, but you should use a magnetic amplifier as a controller. But you need an inverter.

I think im in love you mate! Lol your so awesome ive been tginking about this alot!

@ThinkingandTinkering

3 жыл бұрын

lol - awesome - cheers. mate

You are right about silicon I think new transistor for handling high power are gallium based and is a relatively new tech so expensive. If you want a real controller there should be a project called Vesc a really advanced motor controller for 50 £ the problem is that should be barely strong enough to drive the inverter so a lot of people fry them

This channel and its content just keeps getting better!

@ThinkingandTinkering

3 жыл бұрын

wow - cheers mate

Mos has least voltage drop but Tesla motor drivers still uses NPN MOS driven switches(IGBT), I think because it's more robust. .7volts at 200 amps X 2 X 3 is 840 watts of heat/loss at 100 HP output level.

the power usage must be enormous compared with silicon

I agree that when you just want the darn thing to spin, brushes are better than fancy brushless controllers. Although it's not so much the power transistors that I dislike, but rather the microprocessor. It's basically a black box where you just have to hope it works, and if something goes wrong there's nothing you can do. I don't think there is any good analog solution for robotics where you need to move precisely from one position to another with smooth acceleration and deceleration, but for a car I'd much rather keep all the digital nonsense out of it. Electric cars should be a revolution of simplicity, but instead both manufacturers and buyers have gotten so used to complexity that they can't see the potential.

I do admire your attempt to solve a valid problem. But I am confident you will have a lot of problems as soon as you put any reasonable power through the system. Lots of heating via resistive and arcing losses. You are trying to collapse a highly inductive load at high current and it will arc over at every commutation. DC motors don't do it for a different reason

The real problem you will have was clearly visible when starting, the motor did not start immediately and jerked back and forth. In an electric vehicle the would be terrible. Also you have no load so although you say the motor synchronized with the commutator it was pretty rough and if it gets out of sync when load is applied it will stall and lock up or jerk about again. The true BLDC driver uses hall effect sensors or measurement of induced voltage to determine the most efficient commutation timing. Also the Bldc really only controls torque by means of PWM, and speed is a result of load vs torque. You might achieve some better results if you use an induction motor instead of BLDC.

I was expecting you to show us how to build a mechanical controller. Kindly do a video on that.

I knew it! Have you got a video explaining design details?

I'm not professional electrical person, but what I know and what I understand and what I can think of there has to be an analog way for bldc

do you have a detailed video of how you made this? I have been looking for a cheaper alternative for DC motor controls to help get people into EVs. Can you vary the brush timing easily on this and brush angle?

@j.goebbels2134

2 жыл бұрын

Yes! These controller prices are outrageous. Ideas on something that will drive a 12KW brushless DC?

Hi Robert in your video you used a drill to spin it, if I were to put that on a robot how would I spin that without needing another moter with another moter controller. I was also wondering where that electronic speed controller the you held up and said, it is only is 4 or 5 pounds is from. Thanks the kid

I save mosfets from anything and everything just to swap the ones from my controllers and testing for blown ones are as easy as testing diodes

The question that must be asked is, drawingdrawing or arrangement of contacts?

@ThinkingandTinkering

3 жыл бұрын

next vid mate

Rotary Mechanical Pulse Width Modulation ?.

Robert. Did you get around to showing how you built the communtator, or is it kind of self explanatory ? Thanks.

Hi and thank you.... The idea sound good Mechanical commutator instead of transistors mosfets so on... Having in mind that nothing last forever well mechanical its best i compared with electronic for a number of reasons.2 among many can do it myself and price which for me and guessing tons of people the main reasons...

Very good! Do you have a schematics? The motor being run by the drill is a 3 phase motor being a generator?

@ThinkingandTinkering

3 жыл бұрын

next video for construction details mate

@ajw6715

3 жыл бұрын

@@ThinkingandTinkering Excellent! Can't wait! Lol.

Hi Rob, I am very much for analogue technology. Your controller here seems a cheap option but is probably pretty hard to build, I could not realy see the function. My question is why not use a mag amplifier solution to control a motor, that surely must be a solution?

@paulmaydaynight9925

3 жыл бұрын

seems so, as does Keith's "rotary converters" for better using/reclaiming the available power passed through etc , a small hybrid of some sort using magnetic ink as its laminates lol. it clicked today that motors are basically moving derivative transformers, if nothings moving or 'flowing' no power OMG O_O 'ChipDipvideo Magnetic Amplifier'

If he is able to make it a little bit smaller and more efficient, without using a large drill as the power supply, then this could be an excellent idea for very large Motors

@gcewing

3 жыл бұрын

Drive it with a small brushless motor and small elecrronic controller. Then it's essentially a mechanical power amplifier.

@nil5221

3 жыл бұрын

@Greg Ewing brilliant lateral thinking! But the commutator, having pressure from the brushes, will itself be a fairly high load, too much (I think) for a e.g. computer fan motor. I was thinking perhaps the slim windscreen-wiper permanent-magnet motor from the rear of a hatchback, speed controlled by the electronics from a cordless drill? Something about that size will be the smallest you can go (I think)

Need more details on construction and how it works. Part 2?

@ThinkingandTinkering

3 жыл бұрын

yep - part 2

I get your point that you dont like the complexity of modern speed controllers and Im sure there are some niche applications for mechanical commutation for synchronous motors. Its just that i think that the mechanical commutation you have shown doesnt really solve the complexity problem. You still need to control the motor that drives the commutator which means you need a motor driver and some sort of feedback of the main motor shaft. Why? Because you cant start a synchronous motor from a standstill by just applying some frequency x to its input terminals. It will just wiggle a bit. What would make much more sense would be a DC motor with variable excitation (basically a universal motor). But you would still need a motor controller for rpm regulation. And btw, you can get can IGBT half bridges for 11€ that can do 60kW. 3 of these and you can control any ansyc or sync motor with less than 180kW for 35€. Add lets say 20€ for gate drivers(which arent cheap, i`ll admit), 20€ for some filter caps (which you will need anyway unless you want to get into trouble with your local radio regulation authority because of radio noise) and some random uC flashed with open source ESC firmware and your done. 180kW capacity, rpm control, no sparks, no wearout, no rf noise and legal troubles, convenient bus interface, probably some spare IOs for about 75€.

@ThinkingandTinkering

3 жыл бұрын

it's a small dc motor - a potentiostat will do it

@hellelujahh

3 жыл бұрын

Thanks for the prices, this is good to know

thanks

Nice👍💛

Can this be used as a torque ramp up? IE having 1 of those with a central motor, powering other BDC motors at the wheels?

@ThinkingandTinkering

3 жыл бұрын

I would think so mate - this is really just replacing the commutator - the practice and theory is as old as the hills

I agree with you silicone is good for low power but controllers are stupid expensive and should use other methods that we can repair.

High current electric switching is best done with mercury switches but I don't know if you can still buy them.

@ThinkingandTinkering

3 жыл бұрын

no - I don't know either mate

@Inertiatek

3 жыл бұрын

@Dan A: Heavy current switching is best done with Ignitrons - which can also be controlled by the presented system. You can even make a Mega Watt inverter with these Mercury tubes. Great video!

I recently looked at relays instead of using Mosfets in a project at work. Some results were relays in the 10 to 15 Amp contact rating were the same price as logic level power Mosfets. By the way the mosfets if used within their application specs last longer ...like forever.The ON resistance was negligable when the gate drive signal is right. Ps you dont need a very high level of understanding to replace a switch with a Mosfet.

@paulmaydaynight9925

3 жыл бұрын

true but how do you make mosfet 'down the garden shed' ,jeri probably knows im sure ^_~ making commercial stuff when someone else is paying and can write the costs off as a business expense is fine, everyone else not so much, especially now with the covid scalping excuse for corporate piggybanks needing upgrades to hold the extra swag..

@bretthorwood9396

3 жыл бұрын

@@paulmaydaynight9925 how strange

@paulmaydaynight9925

3 жыл бұрын

@@bretthorwood9396 yeah i know ,why would anyone want to diy a device when they can pay you right :-S

@bretthorwood9396

3 жыл бұрын

@@paulmaydaynight9925 very strange

@stevecummins324

3 жыл бұрын

Unijunction transistors can be used for switching applications including motor control . (ie not just relaxation oscillators) sure they could be used for commutation . And they can be DIYed out of galv sheet, and a blow torch.

I'm going to wind a rheostat using wire from an old electric heater Wire scooter it's always just one step before it's running wide open anyway

Heresy? ...... Yes! You have an electro-mechanical servo controlling angry pixies! ;0)

@ThinkingandTinkering

3 жыл бұрын

lol - I knew there was a reason lol

I guess this guy never heard of IGBT's. Where I work we have 500HP at 480 three phase VFD's to run our overhead gantry and trolly crane. No issues. This is even more complicated than using them as switches. This is using them as pulse width modulators simulating an ac signal.

this is purely linier 😉 analogue one and it realy work .... you need to change 3 carbon brushes after once may be in two years.....

So how can the speed of the motor be increased?