• Have you tried out dark mode?! Scroll to the bottom of any page to find a sun or moon icon to turn dark mode on or off!

diy solar

diy solar

Building the sickest ® VAWT ever. Brilliant minds unite please!!

so the science has finally begun. and again I am not doing it all alone. This time I stand on the shoulders of..

wait for it :)

yes you have guessed it :)

ChatGPT....
The latest and greatest of their models.

Context is now we have a bit of a larger PMA pulley so that the motor does not overheat so quickly and the VFD does not trip out my breaker cabinet so eagerly :)

please let me know if I and/or this parrot which still has no name yet is blundering big time.
-------

User: I have a PMA in an axial configuration with 96 magnets. The PMA spins at 60 rpm. At what frequency does the magnetic field reverse polarity?
Assistant:
To find the frequency at which the magnetic field reverses polarity, you need to determine the number of pole pairs and apply the formula relating mechanical speed (in RPM) to electrical frequency (in Hz).
  1. Number of pole pairs:
    With 96 magnets arranged in alternating polarity around the rotor, the number of pole pairs is half the number of magnets:
    Pole pairs=962=48\text{Pole pairs} = \frac{96}{2} = 48Pole pairs=296=48
  2. Frequency calculation:
    The frequency fff (in Hz) is given by:
    f=(Number of pole pairs)×(RPM)60f = \frac{\text{(Number of pole pairs)} \times \text{(RPM)}}{60}f=60(Number of pole pairs)×(RPM)
    Substituting:
    f=48×6060=48 Hzf = \frac{48 \times 60}{60} = 48 \, \text{Hz}f=6048×60=48Hz
Thus, at 60 RPM, the magnetic field reverses polarity at 48 Hz.

User: Great formula. Can you please turn it around so that we start with 48 Hz and find the RPM?
Assistant:
Starting from the formula:
f=Number of pole pairs×RPM60f = \frac{\text{Number of pole pairs} \times \text{RPM}}{60}f=60Number of pole pairs×RPM
Solve for RPM:
RPM=f×60Number of pole pairs\text{RPM} = \frac{f \times 60}{\text{Number of pole pairs}}RPM=Number of pole pairsf×60
If you know f=48 Hzf = 48 \, \text{Hz}f=48Hz and have 48 pole pairs:
RPM=48×6048=60 RPM\text{RPM} = \frac{48 \times 60}{48} = 60 \, \text{RPM}RPM=4848×60=60RPM

User: I have taken measurements of my PMA and a 3-phase delta wired coil configuration in an open-circuit setting. I forgot to also measure the VDC after full rectification initially, which is why not all rows have the VDC column populated. Can you find any anomalies in this data?
Assistant:
You’ve provided a set of measurements for various frequencies, RPMs, line-to-line VAC, and in some cases VDC after full-bridge rectification. The data appears as follows (using an example table structure):
Frequency (Hz)RPMVAC (line-to-line)VDC (after full rect)
9.64612.05759.688(not measured)
12.0815.111.7814.5
20.0325.037520.02(not measured)
39.349.12539.73(not measured)
42.5853.22543.0355.2
Analysis:
  1. Frequency and RPM relationship:
    With a fixed number of pole pairs, the frequency is directly proportional to RPM. The given frequencies and RPM values match perfectly with the earlier formula. For example:
    • At ~9.646 Hz and assuming 48 pole pairs: RPM≈9.646×6048≈12.06 RPM\text{RPM} \approx \frac{9.646 \times 60}{48} \approx 12.06 \text{ RPM}RPM≈489.646×60≈12.06 RPM The measured RPM is 12.0575, which is spot-on.
    • Checking another point (e.g., 42.58 Hz):
    RPM=42.58×6048≈53.225 RPM\text{RPM} = \frac{42.58 \times 60}{48} \approx 53.225 \text{ RPM}RPM=4842.58×60≈53.225 RPM
    Matches exactly with the measured value.
    So, no anomalies here. The frequency-to-RPM data is perfectly consistent.
  2. Line-to-line VAC scaling with speed (and frequency):
    The line-to-line VAC should scale roughly linearly with rotational speed under no-load conditions. Checking a progression:
    • At 12.08 Hz (15.1 RPM): 11.78 VAC
    • At 20.03 Hz (25.0375 RPM): 20.02 VAC
      The ratio of frequencies (20.03/12.08 ≈ 1.66) matches the ratio of voltages (20.02/11.78 ≈ 1.70), which is close enough to show a near-linear relationship.
    • Similarly, doubling from ~20 Hz to ~40 Hz nearly doubles the VAC from ~20 V to ~40 V. This is again consistent and expected.
    No anomaly in the AC voltage measurements.
  3. VDC after full-bridge rectification:
    The DC voltage after a three-phase full-bridge rectifier at no load can be approximated by:
    VDC≈VL−L,RMS×2−diode dropsV_{DC} \approx V_{L-L,\text{RMS}} \times \sqrt{2} - \text{diode drops}VDC≈VL−L,RMS×2
 
and yet again I am at a loss.

I removed the top magnet disk and ran a few runs.

1733582237859.png


to the best of my abilities to make tests equal this is the situation now

1733581809241.png

I agree that this is just still open circuit but something tells me that the name of the game indeed is frequency of field reversal.

But also somewhere in the back of my mind I hear this nagging sound. You are screwing up big boy. This can't be correct. :(

But we will know more once the loaded test have been done.

anyone please shoot at me. challenge me. hypothesize with me where this is going wrong.
 
Thisisa
TableTestIgnore :)


Header 1Header 2Header 3Header 4Header 5
DataDataDataDataData
DataDataDataDataData
DataDataDataDataPicard

@brandnewb - tables seem to work. In your report, I did not see any [ TABLE ] and [ TR ] (minus spaces) tags - did you edit in BB code directly, or use the table tool in 'graphical' mode?
 
Last edited:
Thisisa
TableTestIgnore :)


Header 1Header 2Header 3Header 4Header 5
DataDataDataDataData
DataDataDataDataData
DataDataDataDataPicard

@brandnewb - tables seem to work. In your report, I did not see any [ TABLE ] and [ TR ] (minus spaces) tags - did you edit in BB code directly, or use the table tool in 'graphical' mode?
I use libre office calc for my spreadsheets.

and then I can select a region and copy it to clipboard memory using Ctrl+c

THe funny thing is that while creating for the first time my post/reply or what ever we call it, to this existing thread then while in edit mode it appears just like the table you are able to produce.

And once I hit the save/post reply button then all things go haywire.

I suspect that you can also see in my change history that my clipboard seems to render to only TD HTML elements. that is missing the TR and the optional THEAD and the required TABLE element.

EDIT: Hahah you already saw ofcourse the above suspects. Ok we will take it from here then.

shall we perhaps create a new test thread to figure this one out?
 
Last edited:
please encounter the all new and improved baseline.

no more vibrations that shake things out of place.

No more fear of losing face.

No more stuff blocking with getting ready to ROCK!!

1733851634257.png


the next few days will be instrumental. I will llay it all down in a way all these books I have bought never could.
 
hahhah how wrong one can be :(
"No more stuff blocking with getting ready to ROCK!!" I said yesterday.

Lord this is sooo frustrating that I can't even continue with gathering the data we are all waiting for from the PMA because the motor started tearing off from the wall that it is attached to when going into the interesting frequencies/RPMs./Amps and wattages

Why is there no book titled?
'Wind for people that think they can beat the odds but do not grasp yet fully what they are up against'

never fear though. I already have a plan to reattach the motor in a different way for it to be more sturdy.

Also never fear for results. This beast should be capable of impressive amps/watts without overheating.
 
with great power comes great responsibility.

I am no longer confident it is responsible to keep pushing as I do in my gym with this DIY setup.

The most recent incident is that the timing belt started to creep out of the PMA pully. Making for a lot of noise, vibrations and in general things one does not want.
It could have been worse because the whole setup still is not balanced yet. If things go poof then (human) legs will break. Or even more worse than that. Mind you we are spinning quite a bit of barely controlled mass around.!!
And that is no joke :(

EDIT: as I realize that this is just dragging on for what seems ages now I will refer back to the maximum wattage I have been able to pull out of this system as to date.

225 watts in a delta / parallel dual disk configuration.

And I just can smell there is more in there to be harvested. The hard part is how to do it.

EDIT: and as always it is as simple as printing a better suited pulley :) I will dive into the new model as we speak.
 
Last edited:
hahahah Things are getting dangerous.
And it is not what one would think.

I just overheard a conversation my wife had with her /peers/team.

That sounds she is on the case and will not let anything slip between the gaps. Mind you the employer is a well known high volume player.
Now I already knew that but this latest example just solidifies my fear.

From all sides I am being threatened by things that can go horribly wrong.
 
It's all about beating that 300 watt barrier.
Yup, I'm sticking with 300 Watts, which is about what you could expect from your Chinese HAWT that you bought all those years ago. I had hoped that you would crush that number, not just match it.

I find it very interesting that education comes at a pace proportional to the desire to learn. You have learned a lot while working on this project. If we had told you to head off to University and get a degree in Power Engineering, you would have told us no way. It turns out that it isn't unreasonable at this point for you to do just that. All you really need to do to get your diploma is quit using AI and start working out the formulas for yourself.
 
Yup, I'm sticking with 300 Watts, which is about what you could expect from your Chinese HAWT that you bought all those years ago. I had hoped that you would crush that number, not just match it.

I find it very interesting that education comes at a pace proportional to the desire to learn. You have learned a lot while working on this project. If we had told you to head off to University and get a degree in Power Engineering, you would have told us no way. It turns out that it isn't unreasonable at this point for you to do just that. All you really need to do to get your diploma is quit using AI and start working out the formulas for yourself.
I just knew it in my bones. that the moment I started slinging this AI parrot around that people would feel less inclined to participate.

I knew it and thank you for confronting me with the confirmation of my gut feeling.

It is important to learn about human behaviour so I can warn my child to not fall into the same trap.
 
I had hoped that you would crush that number, not just match it.
Hey hey hey!! hold on here for a second.

I am no way near done crushing the cheap HAWT that made me start this mission in the least bit.

All I am saying with a motor that is tearing out of its housing is that there is WAYYYYY more power to be harvested.

But how to collect it is all I am wondering.
 
let us not forget that the coils and magnets do not show any signs of heat build up. this tells me there is much more potential in there.

It is just that my mechanical engineering skills are poor to say it less harsh.
 
I just realized that I already crushed my cheap 6KW (cough cough) import turbine output.

Not only by a little. But by a LOT

Nah this is going to be great.

We got Juice!!
 
@justgary can you please tell me what made your statement? or in other words why the pessimism?

if you can motivate, make me realize.

Then this whole mission will be more fine tuned.
Sorry, was I pessimistic? I think you also had hoped to crush the other turbine.

And I do think you have learned a lot, but it seems like you are still hesitant to work out an equation by hand.
 
it is more nuanced than that brother :) Anyway the new pulley is almost ready and then we will try again.

And I will restate that I am already crushing the old turbine. By a lot. Let alone when things are better fine tuned.

Regarding the formulas I will admit defeat without even trying. I can do basic arithmetic and that is about it :(

I think the most important part is is that I value and respect your input. Even though I do not always can generate the most motivation out of them because it confronts me with realism.

Let this old man dream for a little while longer :)
 

diy solar

diy solar
Back
Top