Axial Flux generators

[booster]

Don't they have to control the frequency by RPM, like a generator does, unless it is an inverter style?

[kite_rider]

By the way, the diagram you just posted looks a lot different from the description you provided in the past. I thought the output of this Axial Flux generator was 120V AC, but on closer inspection it looks more like a modified alternator.

The 48V DC system I want to build is pretty much exactly what your diagram shows! The 'Processor' and 'Bi-Directional Power Supply' boxes would be the external voltage regulator mentioned before.

This could pretty much be built with 'off the shelf' components available to a handy DIY person for a reasonable cost. Or so my theory goes..

[kite_rider]

Originally Posted by booster

Don't they have to control the frequency by RPM, like a generator does, unless it is an inverter style?

Yes, the frequency of the 'unregulated 3 phase output' that Avanti mentions would be directly proportional the speed of the rotor. The rotor is connected to the engine crankshaft in some way...

Also, per your comment about seeing 17V (max) on an 'unregulated' alternator; I would reply that it would have everything to do with the voltage applied to the electromagnet on the rotor. I bet if you created the right field on the roto's electromagnet you would see all sorts of voltages coming out of that device.

[avanti]

Originally Posted by booster

Don't they have to control the frequency by RPM, like a generator does, unless it is an inverter style?

Originally Posted by kite_rider

I guess I would ask why you would want to produce 120V AC in the first place? The advantage might be that you could use the engine as the 'generator' for your AC appliances. That sounds like it might work, but what about the 60 Hz cycle that is common for utility power? Seems like it would be complicated (or inefficient) to replicate this over the rpm range of the vehicle engine.

I think the general thought is that it's a much better design is to have a big energy bank (batteries) and then convert that energy to 120V AC via an inverter. You would need this system anyway for times when the engine is off and you charge batteries with DC.

This is a little hard to talk about, since there are not standard names for what is going on in this system.

Basically, what we have here is a purpose-built inverter-like device that is capable of generating 120VAC either from the battery OR directly from the unregulated 3-phase power coming from the axial flux generator. There are three wires from the generator to the controller--one for each phase. Not sure what the voltage is, but I think it is pretty high. The controller produces 120VAC under crystal control--it doesn't care what the frequency (and so the speed) of the generator is. Apparently, this process is relatively insensitive to speed (and of course, the output frequency is completely decoupled from the input frequency).

It seems to me that getting to 120VAC as directly and efficiently as possible is simply taking the "go to higher voltage" idea to its logical extreme, with the bonus of moving to AC at the same time. Plus, it takes the availability of 40VDC appliances out of the equation. Everything is already available at 120VAC.

[kite_rider]

Originally Posted by avanti

Basically, what we have here is a purpose-built inverter-like device that is capable of generating 120VAC either from the battery OR directly from the unregulated 3-phase power coming from the axial flux generator. There are three wires from the generator to the controller--one for each phase. Not sure what the voltage is, but I think it is pretty high. The controller produces 120VAC under crystal control--it doesn't care what the frequency (and so the speed) of the generator is. Apparently, this process is relatively insensitive to speed.

It seems to me that getting to 120VAC as directly and efficiently as possible is simply taking the "go to higher voltage" idea to its logical extreme, with the bonus of moving to AC at the same time. Plus, it takes the availability of 40VDC appliances out of the equation. Everything is already available at 120VAC.

OK, a few things to dissect here. First, an inverter is a complicated device and a generator is a simple device. A modern automotive alternator produces three phases of alternating current at a regulated voltage. It would be very easy to take a standard alternator and get three phases of 14.4V AC out of it. Per my other posts about changing the voltage regulator, you could get three phases of 24V or maybe even 58V AC out of it as well. It wouldn't be hard at all to build an alternator that put out three phases of regulated 120V AC, and THAT is what I thought this Axial Flux Generator was.

The problem would be that the AC frequency would be directly proportional to the roto speed and that would be variable. Appliances need a steady frequency to work properly. This is not an easy problem to fix!

So, if this device is just going to produce DC voltage that will go into a well built inverter, then there is nothing special about it. The most complicated part of the system is the inverter and thankfully there are many on the market already...

[avanti]

Originally Posted by kite_rider

The problem would be that the AC frequency would be directly proportional to the roto speed and that would be variable. Appliances need a steady frequency to work properly. This is not an easy problem to fix!

That is exactly the problem that this device DOES fix. As I understand it, there are two genuine innovations here:

1) The use of permanent magnets in a very specific configuration rather than electromagnets in the generator. The links I provided have very detailed descriptions of the advantages of this approach. This is what the term "axial flux" actually refers to.

2) A bespoke high-current controller that directly extracts the power from the unregulated phases (it doesn't have to be three) of the axial flux generator and uses it to synthesize true sine-wave 60 hz power in an efficient way, and in a way that is independent of alternator frequency. This is not exactly the same as any existing device, but with modern electronics it is not rocket science, either.

[kite_rider]

Hmmmm. The diagrams you provided show DC going into the inverter, which how an inverter works. There is no phase consideration with DC power at all. The battery input into the device would also be direct current.

Permanent magnets have an advantage that they are more efficient than the electromagnet, but have the disadvantage that you can not adjust the magnetic flux. They are used in hybid and electric cars for improved efficiency. I don't see how they have any possible effect on phase shifting the power from the crankshaft driven rotor to create a stable 60 Hz frequency.

"2) A bespoke high-current controller that directly extracts the power from the unregulated phases (it doesn't have to be three) of the axial flux generator and uses it to synthesize true sine-wave 60 hz power in an efficient way, and in a way that is independent of alternator frequency. This is not exactly the same as any existing device, but with modern electronics it is not rocket science, either."

Again, every diagram you provided shows a regular alternator (perhaps with the exception of having a permanent magnet) going through some sort of rectifier to produce DC and then inverting that DC current into "true sine-wave 60 hz power".

And yea, the modern electronics involved with efficiently creating a clean 120V AC at 60 hz from a direct current source kinda is rocket science!

[avanti]

Originally Posted by kite_rider

Again, every diagram you provided shows a regular alternator (perhaps with the exception of having a permanent magnet) going through some sort of rectifier to produce DC and then inverting that DC current into "true sine-wave 60 hz power".

No, it is not a "regular alternator", nor is there a rectifier involved as far as I can see. Note that triple arrows that appear in several places in the diagram. Yes, one box is labeled "DC power bus", but they mean something specific by that (each independent phase is in some sense "DC" in that there is no polarity reversal, but it is pulsating DC, and since it is coming from a single phase coil, it doesn't need to be rectified, at least as I understand it.

I am obviously not doing a good job at explaining this. But these systems are NOTHING like conventional alternator/inverter systems. I suggest you review the info at the links.

[kite_rider]

OK, well out of respect I re-read the links; but my previous statements stand.

I will modify my assessment of this 'axial flux generator' and call it a magneto rather than an alternator - but this is mostly on the suggestion that it might be using a permanent magnet rather than field coils (an electromagnet) on the rotor. For practical discussion purposes, it's an alternator that is voltage regulated to 400 volts. This 400V DC supply goes into an inverter and produces the 120V AC supply.

My suggestion would be to generate 48V DC rather than 400V DC for a host of safety and practical reasons for an RV build.

Best I can tell, you seem to think this device is some kind of dynamo https://en.wikipedia.org/wiki/Dynamo which I don't see at all. Even more confusing to me is that you seem to imply that this dynamo like machine produces an output that is somehow easier to invert than plain old DC voltage. That's where I get totally confused!

Anyway, it was a fun discussion and great thought exercise.

[avanti]

Originally Posted by kite_rider

I will modify my assessment of this 'axial flux generator' and call it a magneto rather than an alternator - but this is mostly on the suggestion that it might be using a permanent magnet rather than field coils (an electromagnet) on the rotor. For practical discussion purposes, it's an alternator that is voltage regulated to 400 volts. This 400V DC supply goes into an inverter and produces the 120V AC supply.

My suggestion would be to generate 48V DC rather than 400V DC for a host of safety and practical reasons for an RV build.

Best I can tell, you seem to think this device is some kind of dynamo https://en.wikipedia.org/wiki/Dynamo which I don't see at all. Even more confusing to me is that you seem to imply that this dynamo like machine produces an output that is somehow easier to invert than plain old DC voltage. That's where I get totally confused!

OK, I am going to try one last time. It is not my goal to badger you, but I see this as an important innovation, and I do not want to leave the record in a misleading state. They are NOT a kind of alternator. As I said, they have almost nothing in common with automotive alternators. In understanding them, it is important not to make a false analogy with alternators. This will just confuse.

Here is my best understanding of this technology in bite-sized pieces:

1) Axial flux generators (by definition) work via magnetic fields that project axially (i.e, parallel to the axis of rotation) from their rotor. This is in contrast with radial flux generators (such as automotive alternators) which project their flux radially from the edges of the rotor.

2) The rotors are completely passive. There are no brushes or slip rings. The rotor is the only moving part.

3) There are two kinds: "axially-symmetric inductive motors" which work like inductive motors in reverse, but have an axial configuration rather than conventional "squirrel cage" construction; and "axial-flux permanent-magnet machines", which I suppose could be considered to be a kind of magneto (although classic magnetos are radial flux, not axial flux.)

4) As I said, they have almost nothing in common with automotive alternators, since the latter have (a) rotating coils; (b) slip rings; and (c) internal rectification to produce a semblance of DC. None of these things apply here.

5) Similarly, they are nothing like dynamos, since dynamos have commutators to convert AC to pulsating DC.

6) Axial flux devices are quite flat, and are sometimes called "pancake generators". Often they are staked, kind of like the platters of a disk drive, with alternating rotors and stators.

7) The particular devices I am referencing have three phases, but this is arbitrary. There are versions used in wind generation that have more than a dozen poles.

Here is another diagram:


8 ) The stator coils each output an independent waveform. They are not connected together in any way within the generator. There are no rectifiers. I am not sure exactly what these raw waveforms look like (or even if they are AC vs pulsating DC). Doesn't really matter. Each of these waveforms is sent via its own wire to a central converter (labelled the "3 phase vector controller" in the above diagram). This controller generates the necessary field currents in the inductive case, and in both cases converts the three independent currents to an internal high voltage DC bus (200 or 400VDC). NOTE that there are THREE wires (plus ground) from the generator to the controller. This is key to understanding why this is different from an alternator, which only has one output wire. The high voltage DC bus is entirely within the ECU, so there is no safety issue.

9) The DC bus is fed into a pure sine wave inverter that produces 120VAC. It can be seen that there is no relationship whatsoever between the speed of the generator and the frequency of the AC output.

10) This setup is extremely efficient. Aura claims that their system runs at 75-80% efficiency, even at low speeds. An alternator might reach 75% at high speed, but reduces to maybe 10% at low speeds.

This technology is not pie in the sky. As I said, Great West was starting to ship B-vans with these systems when they folded. I saw one in a Sprinter. What interests me is a van that is almost entirely 120VAC, with all the DC and inverter conversion within the ECU, where it could be done very efficiently. No fat wires anywhere, and cheap, efficient AC appliances and standard power bricks for electronics.

[gregmchugh]

So, when the van engine is not running, the battery bank provides the DC power to the inverter instead of it being powered from the axial flux generator and there is no separate inverter used when the van engine is not running?

[avanti]

Originally Posted by gregmchugh

So, when the van engine is not running, the battery bank provides the DC power to the inverter instead of it being powered from the axial flux generator and there is no separate inverter used when the van engine is not running?

That is my understanding.

[gregmchugh]

And there is a single 120v AC “battery charger” used for battery charging when running the van engine or on shore power and a solar controller would connect directly to the battery bank as usual?

[avanti]

Originally Posted by gregmchugh

And there is a single 120v AC “battery charger” used for battery charging when running the van engine or on shore power and a solar controller would connect directly to the battery bank as usual?

It appears that the battery charger is built into the the one and only unit (labeled "ECU" in the diagram). Yes, solar would be conventional.

[Winston]

Originally Posted by Davydd

By leaving our inverter own all the time I estimate the "overhead" is a draw of about 72 amp hours off the batteries in a 24 hour period. That's about a 15 minute drive to replenish with our Delco second alternator producing 280 amps.

Originally Posted by kite_rider

Where we don't see eye to eye is that I think there are too many things that scream "we want 120V AC".

. . you already have a second parallel DC system with your second 12V alternator! What I don't like about the Nations system is that it's big, bulky, expensive, and has very thick cables.

Davydd, seems we have similar systems and overhead. Our 3.6 amp overhead equates to 86 ah/day - - surprisingly close to your figure. The difference being, for the most part, we don't use the second alternator relying, instead, on solar to meet our daily requirements with the 2nd alternator being a backup if the sun doesn't cooperate.

In general it seems we three agree that a 120VAC-centered system is the way to go, particularly with all those screaming AC appliances. Where we may not agree is that we feel that there remains a need for 12 volts DC. Now, maybe, you'd minimize this need and rely on the vehicle battery/system for that. But we can't ignore the need for a second DC system (conceding the point that we already have one) to receive and store the energy necessary to power the inverter. And since we have concluded that "we" need 12 volts, why not make this second DC system 12 volts? We're killing two birds with one stone.

You say "too large an alternator" and "too large wire size". Ok. But, so what? Yes, that Nations alternator hanging 'under' our ProMaster seems precariously low and, we admit, we're just waiting to whack it someday. And the required wire gauge is ungainly . . . but once installed, it's over (unless we succeed in whacking the alternator). For the minor inconvenience of that initial installation, we gain the huge benefit of having a virtually limitless supply of 12 volts.

By the way, we down-sized the wire gauge from our Nations alternator for the 15' (30' round-trip) run to the batteries (in the rear) to double-0. Just didn't want to fight with the heavier stuff. And since we ran a separate battery voltage sense wire to our Balmar regulator we really didn't care about the increased voltage drop.

But in fairness to the 48 VDC 'argument', there are some real advantages to a higher voltage system. First, unlike Davydd, we do not run our Nations alternator at its rated 280 amps, instead, at 125 amps. We've chosen this conservative number, in part, because it seems more than adequate to meet our charging needs but probably more to do with alternator temperature which, even at the 125 amp level, can rise dramatically, particularly when the vehicle is not moving. A 48 volt system would significantly reduce this heat.

Another, and more subtle, advantage of higher voltage relates to the idiosyncrasies of our Magnum 2812 shore power and Midnite Solar Classic solar charging systems. Ideally we would like chargers that 'put-out' everything they've got until a predetermined voltage is reached at which time they switch to a constant voltage. (That voltage, incidentally, is 13.36 volts - - more on this, maybe, in another post). Unlike our 2nd alternator, these two charging sources do not have battery voltage sense wires and therefore are making their 'decisions' based on their respective terminal voltages. When dealing with lithium, where 0.01 volts is important, even minimal voltage drops between charger and battery can be critical. Again, a 48 volt system would practically eliminate this issue.

Anyway, whether the smartest decision or not, we're currently 'stuck' with a 12 volt house electrical system and are not unhappy with it.

[kite_rider]

Avanti, I don't feel badgered at all. The last diagram you added to your post and the references to the wind power applications were interesting indeed and I agree with you that this type of generator is different from a traditional alternator.

Thanks for keeping the discussion going.

Also, even though this concept may not be seeing much love right now from the forum; it doesn't mean that it's not the right fit for the application.

For what it's worth, I also feel like I've presented ideas to this forum that might have been misunderstood or quickly discounted. In December of last year I suggested a 'unique' Li-Ion solution for a camper build that utilized a 52V solution (charged at 58V, but often called '48V') with NMC type Lithium batteries rather than the Iron Phosphate chemistry that is common today.

http://www.classbforum.com/forums/f2...uild-7217.html

The feedback was basically, '48V (really 52) was not worth the trouble' and 'NMC is not the right chemistry...'. Now Volta will offer a new system which is pretty much exactly what I had originally suggested and people seem to very excited about it. So go figure...

[booster]

Winston, how do you get the Nations down to 125amps? The largest turndown I have found in the Balmar is 50% (field not actual output controlled) and can be done one place as a fixed 50% reduction and another setting that allows anything from 50% to 100%.



We use an Ample Power regulator that allow two stages of fully settable output, so we are lower than 50% on our two in parallel (560 amps) of alternators, but if it dies we can't replace it because they are no longer made. We have them turned down to give 180 amps for normal charging as that is the most we can put into our AGM batteries long term without heating them up, and also have a 280 amp setting that we can use for a blast of quick charge for about 20 minutes before the batteries get to warm.

[avanti]

Is there any advantage to reducing the power of the alternator assuming one has the Balmar configured with temperature sensors both on the alternator and the battery?

[booster]

Originally Posted by avanti

Is there any advantage to reducing the power of the alternator assuming one has the Balmar configured with temperature sensors both on the alternator and the battery?

Mostly my opinion, but based on a long factory history with hot running electrical stuff.


Our rule of thumb for electrical stuff was that nothing should ever be thermal cycling off the thermal protection system. The heating and cooling thermal expansion of things puts a lot of cyclic stress on them, leading to cracks and breakage, especially in copper, which also oxidizes faster at higher temps. We would add cooling, or device capacity, to keep them within recommended running temps so they wouldn't thermal cycle unless there was a failure of something.


IIRC, Balmar even states that the regulator should not be thermal cycling off the alternator temp sensors, and the temp sensors are probably one of the higher failure rate parts we see for the Nations setups. I think they are just a thermal snap switch, so not designed for lots of cycles. It would be nice it the sensed alternator temp like they do battery temp, and then controlled the field output automatically like they do temp compensated charge voltage, so continuously variable output. That would give the highest continuous output you could get without overheating the alternator and would give no thermal cycling.

[avanti]

Originally Posted by booster

IIRC, Balmar even states that the regulator should not be thermal cycling off the alternator temp sensors, and the temp sensors are probably one of the higher failure rate parts we see for the Nations setups. I think they are just a thermal snap switch, so not designed for lots of cycles. It would be nice it the sensed alternator temp like they do battery temp, and then controlled the field output automatically like they do temp compensated charge voltage, so continuously variable output. That would give the highest continuous output you could get without overheating the alternator and would give no thermal cycling.

Interesting. I guess I had just assumed that the two temperature sensors worked the same way. So, you don't think that the alternator sensor is a thermistor? (It has been too long for me to remember).

I agree with you that it ought to modulate the current for steady-state operation. It would be an interesting project to use an arduino or some such to retrofit this behavior. One could spoof the battery temperature input to accomplish this.