Abate Holding Your Breath...Thompson's Design

I asked about that transformer. (Which is not an autotransformer is it? It has 2 coils.)

But Larkin did ask you to critique his architecture, and I pointed out that if you just did that we could move along. So why don't you both explain why it's necessary, or won't work, instead of this "you first".

Unless you want to make it a private conversation in which case you should do it via email.

--

Reply in group, but if emailing add one more
zero, and remove the last word.

Oh, and I'm curious about this dynamo vs alternator issue. Some people have said these things generate AC. If none of us have tested one, maybe we should assume the worst.

Is that the only reason you oppose the use of a transformer? You might have said so with less fuss.

Assuming the worst, I'd think to use a full wave bridge and no transformer.

--

Reply in group, but if emailing add one more
zero, and remove the last word.

message

a

First,

several:

these

same

dealing

at

reduce

really

the

so

voltage

battery

while

one

so

Say why an autotransformer won't work.

John

It doesn't matter if it's a transformer or an autotransformer. The point is that, at decent speeds, the alternator is a high voltage and a high impedance, practically a current source for low-resistance loads like lamps. A step-down transformer can boost the current into the load by better matching impedances. So can a rectifier-filter-buck switcher, done right.

A 2:1 step-down autotransformer might be convenient to get from Digikey, since you could use some standard part with dual primaries or a center-tapped secondary.

John

An autotransformer simply is a winding-use expression. Think of a simple 1:1 transformer, but connect the winding in series (minding the "dots" :-) I assumed "perfect", K=1, and manipulated only inductance. Input is end-to-end, output is across only one winding.

IF (and that's a big IF in my mind) the Magdowski dynamo model is correct, an auto-transformer buys you only something on the order of

20% output current improvement.

He claims, I asserted not. He won't/can't prove me wrong. But he will continue to be crude. (I don't care, it's sort of fun... wonder when he'll mentally crack ;-)

...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

Sort of interchangeable terms, but dynamos typically have permanent magnet "fields", while "alternator" usually means a field made up of a winding.

Yes. In automotive and some bicycle uses, followed by rectifiers.

Depends on your definition of "worst" :-)

I don't oppose the use of a transformer. Originally I thought Larkin's idea was a good idea... that's why I tested it... it would be an all-around easier device to regulate... but it fails, and I said so, and have been castigated for it.

Yep. Full-wave bridge, no transformer, chased by an off-the-shelf LiIon 3V charger chip to easily get the battery management (rolling your own would take too many chips... I did my first LiIon charge-discharge controller for CalMicro so long ago, pre-CAD, that I only have paper schematics :-), LED lighting... add input regulation to avoid over-voltage into the charger chip... at 3V it should be _really_ simple. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

message

largely a

First,

several:

generator

these

same

dealing

at

reduce

really

the

so

voltage

battery

having

monitor

while

one

3W, so

Because I say so. Prove me wrong. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

people

might

I, too, am unsure whether hub dynamos produce AC or DC. Larkin's design = with=20 the transformer assumes AC, while Thompson's assumes DC. If his "model"=20 includes diodes, then their specifications become very important to = overall=20 efficiency, since two diode drops may constitute 1/4 of the voltage. If = the=20 hub dynamo is a true alternator, then AFAIK it would contain diodes (as = in=20 an automotive alternator), and would very likely use three (or six) = phase.=20 But according to the following, the output IS AC, and it must be single=20 phase if there are only two terminals:

Perhaps even greater efficiency may be obtained by using a voltage = doubler=20 consisting of two Schottky diodes and two capacitors. If the dynamo = outputs=20

6 VAC then the doubler should produce about 16 VDC. However, the = performance=20 at low speeds would depend on the frequency of the AC and the size of = the=20 capacitors. The autotransformer is also subject to low frequency = limitation,=20 and in fact may present a heavy load when it saturates at low frequency, =

although not if the AC output is proportional to speed.

I think it would be advantageous to design a hub with a three-phase=20 topology, as it may reduce or eliminate the "cogging" effect that was=20 discussed in one of the articles. Another topology that might be=20 investigated is a three phase induction motor, which requires no = expensive=20 magnets and is very rugged. But it does require sophisticated = electronics=20 and a pre-existing voltage source to produce the rotating field. Or it = might=20 be possible to use a SEIG, which could be accomplished with a small PM=20 generator to power the electronics which can then run off of a small=20 capacitor during times of no movement.

Since we have broached the subject of wheel hub generators, it opens up = the=20 discussion to hub motors, which may provide enough power to assist the=20 cyclist on uphill climbs and then harvest the energy of downhill = coasting,=20 even providing some braking force.

Because of my problems with knees hips, and back, I can't really use an=20 ordinary bicycle, but one of these may fit the bill. But I may also try = a=20 dog-powered cart, or bike-jouring equipment:

My dog Muttley sure pulls hard enough!

(he helped pull my mower up the hill)

Paul and Muttley

Surely you realize that the only way to prove it _will_ work is by building it.

--

Reply in group, but if emailing add one more
zero, and remove the last word.

I mean it would be best to make it work with AC or DC.

That must have been another thread. I didn't see why it fails.

--

Reply in group, but if emailing add one more
zero, and remove the last word.

Possibly. I've said over-and-over, I'm suspicious of the Magdowski model. Simulation using the Magdowski model fails.

But we're pursuing a dead horse. Anyone with half-a-brain would go to LED lighting. Everything off-the-shelf for LED bicycle lighting is 3V powered. If you've got an old hub dynamo-based bicycle you're home free... only problem is that Ian doesn't have half a brain (likewise Larkin ;-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

Jim, I'm not getting into this stupid pissing match. But, erm, shouldn't = you=20 be working on your mid-January project, rather than trading verbal = volleys=20 with John? And, uh, what about your offer of free consulting, which I = took=20 you up on a month ago? I had hoped to get at least a "little" bit more = than=20 I paid for. The design is basically done, and I'm in the process of = testing=20 some of the subcircuits and getting bobbins and ferrite cores for the=20 transformers. And I realize that my project is much more than a few = hours=20 work. But what I WOULD like is at least a simple critique of my design = for=20 the high-isolation DC-DC converter, and perhaps your official blessing = (or=20 not) based on your PE credentials. I want to have some assurance that = there=20 will be no design-related problems in the future from the 480V 200A = mains=20 power jumping into the control circuitry and creating a big mess.

And if your advice and critique are sufficiently professional and = valuable,=20 I may be interested in additional consulting work for which I would pay = a=20 reasonable amount. Can you at least generate a report based on the = material=20 I provided?

Please?

Paul

(customer for whom I am doing this PI-200 project)

The "Position Sensing" thread was also hijacked. To see why it fails, run the Magdowski dynamo model into an auto-transformer whose rectified output has to be >= 6V. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

Dynamo outputs AC, DC is required to charge a battery.

As I've already stated, the CONCEPT works, efficiency can be brought up in several ways.

The limitation is the dynamo characteristics. I tried auto-transformers up to 100H inductance.

Besides, the output, according to Larkin is a CURRENT... not quite, but almost.

[snip] ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

Starts Monday. Right now I'm revamping my office layout to fit in some more file cabinets and storage.

I asked for a description of WHAT you were trying to do. Schematics and PCB layout did me no good, so I helped others.

Tell me WHAT your circuit is supposed to accomplish, and I'll give it a quick shot.

I didn't know that anyone here cares what I think ;-)

...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

Fails, with every circuit?

Ian said a couple of times he is using LEDs.

Quote:

The extra power going into the battery would be handy for occasional use of a 3W star LED "main beam" for unlit sections of cycleway.

--

Reply in group, but if emailing add one more
zero, and remove the last word.

Then why was he pushing charging his _6V_ SLA ?? ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

In British cycling terminology, the term "dynamo" is incorrectly applied to the lighting generator - a dynamo has brushes and generates DC while bicycle generators would be more correctly described as alternators.

Alternators come in a couple of principle types - permanent magnet and field coil excited type.

Permanent magnet alternators are usual on bicycles and some types of older/smaller motorcycles, field winding alternators are usual on newer/larger motorcycles and on cars.

Alternators produce AC - you can always be fed through a rectifier if you need to charge a battery.

AFAIK the misuse of the term "dynamo" is largely confined to Britain - the term "generator" is used more widely in connection with bicycles.

I had sent you a fairly complete project outline=20 ETI_Next_Gen_Test_Set_PI-200.doc with my initial contact email sent on=20 December 18. But that was rather comprehensive and included the entire=20 design, whereas I later condensed my requirements to just the = high-isolation=20 DC-DC converter, which mostly involved the transformer design and the = driver=20 to obtain about 6-9 VDC at 300 mA from a 12VDC supply. The schematics=20 included my design for that subcircuit, and I really just wanted your=20 opinion on how the board should be laid out to achieve the safety and=20 isolation of the 480 VAC mains.

My description of this portion of the design, from my email of January =

6,=20 2012:

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D I think the most reasonable approach at this point is to have you design = a=20 small, but important, component of the system. The document I sent you=20 described the unit and some of the components, and most of that has = already=20 been designed. But I am now working on the SCR gate drive, which = requires a=20 high isolation DC-DC converter. I am getting quotes on the transformer I =

need for this, but perhaps you can help. Here is what I have sent to = some=20 local transformer shops we have used:

I am designing a new SCR trigger board for a smaller test set (PI-200, = for=20 which you recently provided a transformer quote), but may also be used = for=20 the larger ones, and I will need a special transformer for a DC-DC = converter=20 for the gate drives. I will supply a 12 V P-P square wave to the = primary,=20 and the secondary will be a full wave bridge using Schottky rectifiers = to=20 produce a nominal 8VDC at about 400mA. I was planning to use a switching =

frequency of about 20-40 kHz, to keep the transformer size small. But I = need=20 to have isolation and safety for use with nominal 480 VAC mains. The = best=20 commercial DC-DC converters I could find are the:

They are rated 4000 VRMS isolation but for continuous duty only up to =

300V.=20 I would like a transformer that is no larger than these units (about = 1.5=C3=A2=E2=82=AC=C2=9D=20 x 1=C3=A2=E2=82=AC=C2=9D x 0.5=C3=A2=E2=82=AC=C2=9D), and I can supply = the drive and output electronics. I am=20 designing the circuitry now and I plan to design the PC boards in about = a=20 week, so I need at least the basic form factor that a custom transformer =

would take.

I think a split bobbin design might be best, something like the Atech=20 ATS-075A. (See attached files). It has 3750 VRMS hypot but = it=C3=A2=E2=82=AC=E2=84=A2s 600 ohms=20 impedance. I have one and it appears to have steel laminations. That is=20 probably OK for 20 kHz but maybe ferrite is better if I use 40 kHz.

The Triad CSE187L is also the right size, but the split bobbin is of = unequal=20 size, since it is a current transformer with a single turn 30A primary.

I have also made a transformer using a ferrite toroid and about 30 turns = of=20 #26 enamel wire. But I don=C3=A2=E2=82=AC=E2=84=A2t know if the coated = core can be trusted for=20

480 VAC use.

However, perhaps I can just use some very well insulated wire (perhaps=20 Teflon) for the primary and secondary of a toroid (or other type). If = the=20 insulation is rated to withstand 480-600 VAC continuous then that would = be=20 another option.

I have also found some laminations from Thomas Skinner=20

that are =

0.75=C3=A2=E2=82=AC=C2=9Dx1.00=C3=A2=E2=82=AC=C2=9D (EE-24-25)=20 of 4-14 mils thickness, and I found a good selection of bobbins from=20

That should be enough to get you started. I have the PCB laid out for = the=20 Atech ATS-075, but I can easily modify the decal for another type. But I =

would like to have the size be not much larger. Maximum is about =

1=C3=A2=E2=82=AC=C2=9D square=20 and about 0.75=C3=A2=E2=82=AC=C2=9D high. Basically I have 12 VDC = available and I=C3=A2=E2=82=AC=E2=84=A2d like to=20 get about 8 VDC at a minimum of 300 mADC for the gate drive of the = larger=20 SCRs. But if that causes a size problem, I=C3=A2=E2=82=AC=E2=84=A2d be = OK with about 5 VDC at 250=20 mA.

And a major consideration is the 480 VAC mains. I plan to use the 1.25A =

600V=20 SMT fuses for the gate and cathode wires, as discussed in S.E.D. = recently.

Littelfuse 04611.25ER, Newark 34C6800

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

I don't know what more you need. And I was using this as a test to see = if=20 you could offer help for other things in the future. Oh, well. There are =

others who may be able to help with other issues. Good luck on your=20 renovations and main project. And please don't continue this ongoing=20 argument with Larkin, or at least snip all the previous worthless = chatter.=20 What was it you said about his wife that got him all riled up, anyway?

Paul=20