55W PMA: almost no output??

[Ontheronix]

Hi all!

Long time since I posted here

I have bought one of these low rpm PMA's on Ebay, I guess they are recycled from old tape drives or something similar.
https://nl.aliexpress.com/item/90W-DIY-DC-120V-power-motor-generators-for-wind-turbines-Wind-turbine-generators/32361002271.html

Specs:
1000 rpm 50 V   
2000 rpm 100 V   
2500 rpm 120V

Max power is 55W according the Ebay auction..


The problem is that I'm not getting much output from it. Maybe 0.5W? I attached a 12V 4W car lighting bulb and I can see the wire light up, but there is no real light emitted. Voltage measured when driving the bulb is 2,6V It also turns heavy when the bulb is connected and I try to rotate the axle by hand.

Open voltage is 32-33V (max rpm for my cordless grill I guess). And short-circuited I produces 100mA.

My cordless drill is not in the best shape but it can get to 200rpm for sure, wich should be able to outpout 4-5W, if my calculations are correct:
2500rpm = 55W
250rpm = 5,5W

I bypassed the electronics inside, so the wires are directly connected to the brushes. I also switched the coils (rotor) with another coil from the same generator. It didn't make a difference in the output. So I'm completely in the dark here, does anyone have an idea?

[dnix71]

5 watts looks about right for a tiny little fixed magnet motor like that. It doesn't even have end bearings, just bushings. I don't see how you could safely spin it 2500 rpm.

[Ontheronix]

It has two bearings inside, it seems quite sturdy to me?

Do you mean 5W at 200rpm?

[dnix71]

I see the hidden bearings now looking more closely at the pictures. That's a tiny motor. It reminds me of the 5 watt generators that were sold to rub against a bicycle tire to power lights.

This on is 6v and 3 watts an is about the same size as your Chinese made motor.

https://www.amazon.com/Factor-3-Inch-Bicycle-Generator-Light/dp/B003H3Z8CW

[hiker]

Modle build..?

[JW]

The bearings in a small DC motor, it doesn't  have thrust bearing. 

[dnix71]

https://www.amazon.com/ask/questions/asin/B00YG5Z9NI/2/ref=ask_dp_iaw_ql_hza?isAnswered=true#question-TxAY6AIII2VZ2V  Read the reviews and questions here. Someone asked for specs in Dec. 2015 and is still waiting.

Someone asked what the rated output wattage was and the seller refused to answer. Others suggest the best use for this is a science project. The most useful answers seem to be that this would keep a 12v battery topped off.

What exactly do you want to do with this toy motor?

[Ontheronix]

It has ball bearings indeed  I think it's indeed some kind of low rpm motor, maybe from an old tapedrive?
Not much information on the internet, but what can I expect from recycled stuff.. 

My goal is to have more output than a bicycle generator (commonly used here in Europe), so something 5-10W would be nice.  It's enough if it can light a bulb or some powerleds.

[Bruce S]

Welcome back.

From looking at the two output wires, and the notes of it being a 120Vac 90W unit, that the wires inside are probably pretty small.
Knowing that when you spin it with your drill at about 200RPM , that you only get about 32Vac . I'm betting that you are not going to get anywhere near that 5 watt mark until you sustain a much higher speed.
Even then the current isn't going the be very high due the internal wires being (I'm Guessing 30ga).

IF you build up a small mill with say 3-bladed no more than 36 inches , you should see "some" output, but you're going to need to get it into a good high wind speed. Can't guarantee anything
GHURD, Wooferhound, vawtman and a few others (I THINK) built little ones too.
GHURD's miniECM post https://www.fieldlines.com/index.php/topic,138380.msg918562.html#msg918562 

The post of GHURD's post should help guide you.

Cheers
Bruce S

[dnix71]

Impedance matching seems to be an issue with this. One buyer got 16.2 watts out at 13.5v with some help. Otherwise at 120v he only got 2.4 watts.

How many amps will this generator put out? My daughter is doing a science project and she is making a wind turbine that will charge her iPhone.

If you want to use it to produce 120 I only got .02 amps out of it. I ran it though a bridge rectifier and a voltage regulator and got it up to 13.5 volts at 1.2 amps. I hope this helps
By Brian W. on October 13, 2016

[Ontheronix]

Thanks, that's useful information! The wires of the coils are very tin indeed. I guess no more then 30ga indeed, you can see some pictures here:
https://nl.aliexpress.com/item/2PCS-65W-120V-DC-power-DC-motor-generators-for-wind-turbines-3000rpm-150V-free-shipping/1896659917.html

What kind of voltage regulator can I use? I tried replying on the Amazon post, but you have to do a credit card purchase to be able to post. I'm gonna try with one of those chinese DC-DC transformators.
Edit: I tried the transformer but the LEDs didn't light up. Will try again tomorrow when I have my multimeter back.

Why would someone run DC through a rectifier?

[Ontheronix]

I measured the resistance from one side of the commutator to the other side (=180° apart) and it measured 247 ohm. Looks a lot if I see Ghurd measures 14ohms per fase at his small turbine?

[JW]

You cant use a transformer without feeding it AC current. Out of the transformer is what you would hook a bridge rectifier.

[Ontheronix]

It's not a classic AC-AC transformator, but a DC-DC buck/boost converter. For example: https://www.youtube.com/watch?v=nvHwWcwqAzA
Meanwhile I'm reading up on impedance matching (matching the impedance of the load to that fo the generator) but it is hard of you're only a hobbyist in eelctronics :p Is it something like finding the maximum power point in PV or alternators in general?

[JW]

Interesting..

[Ontheronix]

I measured the voltage at the input of the DC-DC converter, with no load connected: 0V! Hence also no output.
I also tried with a similar generator that I had lying around at my parents home, and it has the same problem: almost no output, maybe 0,5W maximum!

Edit: Interesting read on impedance and load matching: https://www.fieldlines.com/index.php/topic,146527.msg1006693.html#msg1006693
Maybe I can build a string of LED's to up the possible power output. Assuming the mA will stay the same.

Edit2: "What this means in "practical terms" is all F&P stators (any PMA) want to put out a constant amount of current with a variable voltage proportional to RPM in order to maximize their output potential." https://www.fieldlines.com/index.php/topic,149388.msg1042978.html#msg1042978

[Bruce S]

Let's see if we can help you sort these out.

Sorry if this sounds redundant, but at first you had some voltages , and now you don't.
When that happens to me, I take a step back to the very basic beginnings.

So , when you have the time, remove everything from the generator, put your meter on DC setting, attach it to you drill and give it a spin.
Make sure it's still got something coming out and let us know.

I know the link you provided says DC but let's make sure it's actually outputting DC and what that is.

Most of the time unloaded voltage has little meaning, but right now making sure there "something" coming out is a good start.

Next tell us how you attached the output of the gen to the convertor ( a picture would be nice ), if you have two meters, attach one to the generator side of the convertor the other to the output and let us know what you get. IF you only have 1 meter, attach it to the generator side of the convertor and make sure the convertor ins't dragging the voltage down to 0 (zero).

That DC-DC convertor may need a minimum voltage to begin working. The guy in the video is using it as a Buck convertor , this may also be the problem.

Bruce S

[Ontheronix]

Hi Bruce, the gen still can produce unloaded volts up to 30-33V (I was also afraid something broke inside ). + and - should be connected correctly to the DC-DC converter, as I doublechecked with the multimeter. So I guess the converter is just dragging the voltage down, maybe because it has 2 big caps.
(It's a boost converter, I now know I need a buck converter).

The reviewer at Amazon extracts 13,5V at 1,2A, which means the impedance (?) of the load is U/I=R -> 13,5/1,2 = 11Ohms. I don't know about the impedance of LEDs but maybe I can put some bicycle bulbs in series to reach a similar impedance?
This is loosely based (=what I can grasp) on the mathematics of Hugh Piggot: https://electronics.stackexchange.com/questions/177513/which-of-the-following-motors-best-suit-as-a-wind-generator/177584#177584

[Bruce S]

good to know the gen is still working .
Do not put any LEDs on that output, the voltage is way too high unloaded. LEDs normally max out at 3.2Vdc and only need 15mA to light up.

Also; Your are correct, on that convertor, the output probably needs to get up to about 5Vdc and enough current to charge up the input so it can regulate to what you want it to do.

Knowing that the car light bulb lit up then there's something working .
Now that we know that; what voltage do you want to work with? and use it for?

IF this is to learn about this stuff and a hobby, then you picked a good learning one 

[Ontheronix]

It's indeed a learning/science project. It's a steep learning curve, but I am learning  The goal is to demonstrate that a mini windturbine can be made for nearly 0$. I made blades from an old plastic waste bin, total diameter approximately 1m (3,5'), but siting will be bad so I'll be happy if I can fully light the 4W bulb I guess 

I was talking about 3W powerleds  LEDs are current driven so voltage doesn't matter as long as the power does not exceed the 3W, if my understanding is correct.


While the battery of my cordless drill is charging, I measured the ohms of the lights:

4W 12V bulbs:
5,7 Ohm
20 Ohm (seems a bit off)

10W 12 bulbs:
1,7 Ohm
1,7 Ohm

3W 3,5V leds:
N/A (voltage of multimeter is probably to low)

[Bruce S]

Okay;
Getting to the meat of things.

If you get noting when trying to measure across a LED remember it is nothing more than a diode at heart.
Try turning the swapping the leads around. IF my cheapo HF meter will light a LED .
While they are truly current devices at some point the voltage has to go somewhere, and an in-rush of high voltage without a way to dissipate the resulting will pop that LED literally in the blink of an eye (BTDT).

Also;
Knowing what the ratings are and what you're seeing with your meter tells me your using a digital meter. Once inexpensive meters get down below a few hundred Ohms their reading can be "off".  Without doing some corrective calcs.
Here's a link to an online calculator that I've used http://www.ohmslawcalculator.com/ohms-law-calculator.
Best to go with listed ratings to keep things working, to start off with.

What are the wind readings in the spot you plan to put this?
 

[Ontheronix]

I have a professional multimeter: https://www.velleman.eu/products/view/?id=387040&country=us&lang=en so all should be good.
My leds all light when I put it on diode measurement

Wind readings are non-existent, the turbine is at ground level between obstacles so the speed will be very low on average. It's meant as a 'proof of concept', maybe to build a bigger one later. So actually it's not really a concern

Meanwhile I have connected some more powerleds:
With 8 powerleds in series, the gen produced 25mA @ 23,24V, producing 0,58W @ max. 700rpm (cordless drill)
With 4 powerleds in series, the gen produced 75mA @ 11,7V, producing 0,88W @ max. 700rpm (cordless drill)
With one powerled, the gen produced 110mA. I didn't measure the voltage but with the voltage curve of the led in mind, it must be around 2,5V.

I also drove it as a motor. I don't have 120V here, but at 20V it consumes 23mA.

Maybe the guy at Amazon was able to drive it towards 120V/100mA and then converted it.

I think the above numbers show that the generator is having difficulties generating more than a few watts, which contradicts the findings of the Amazon reviewer that got it to produce 16W. This means that an impedance mismatch is the only culprit left. So how can I match the impedance? 

Edit: I tested with another generator of the same size and type, and the same problem occurs: only 20mA at 23V with 8 LED's.
(The LED's are Cree XPE's)

[dnix71]

I went back to the aliexpress pictures. The person who listed this as a wind generator is a criminal, period. Look at the rotor. I know exactly what this is. Anyone who would sell a mini rotary tool motor as a wind generator should get a boot so far up his bum he could taste leather.

I have a Dremel that used a spring to connect and cushion the output to the collet. The spring broke so I replaced it with a piece of clear vinyl tubing. The vinyl doesn't last long but it preserves the motor and forces me to work within the limits of the tool. This motor was made for a Dremel or similar tool. Mine pulls 90 watts.

His motor is direct output with no coupling. Simpler and cheaper to make but if you lock the rotor it will burn up quick.

The hole is to lock the ouput shaft when changing bits. Even the bearings are the same. I know of no simple way to run this backwards. It runs from chopped a/c.

[Ontheronix]

Wow interesting discovery! It looks very similar indeed! But why is it not suitable as a generator? It has 2 ferrite magnets in it, and as far as I understand, DC brushed PMG's like the Ameteks work in the same way? Like this:



(I ran it at 20V DC, and the axle rotates fine? But maybe it also works on chopped AC?)

Edit: I gathered some interesting information on load matching through capacitance:
Since the output of the rectifier is a dc voltage, the maximum power transfer will happen when the load and source impedances are matched. This means that the resistance that gives the highest output power will also represent the internal resistance of the generator and rectifier.

It is known that the maximum power transfer occurs when the load resistance is equal to the magnitude of the source impedance. From this we can say that the internal resistance of the generator is 3.7Ω. This resistance includes all components of the system, not just the stator windings, but the rectifier circuitry as well.

The internal impedance of the system will also change with the rotor speed, so this test will need to be implemented when the internal impedance is desired at any rotational speed
Src: http://digitalcommons.calpoly.edu/cgi/viewcontent.cgi?article=1212&context=eesp

Edit2: Complementary info: http://forum.solar-electric.com/discussion/14464/i-keep-reading-about-mppt-for-wind-generators-heres-an-alternate-idea

[Ungrounded Lightning Rod]

don't know about the impedance of LEDs but maybe I can put some bicycle bulbs in series to reach a similar impedance?

If you're doing it as a load, i.e. the idea is to light them up to a particular level, that works.

if you're trying to use them for testing the output, you have a problem:  The resistance of an incandescent lamp varies by a factor or several between being cold and being fully illuminated.  So if you're using one as a load for measurements you need to measure both current and voltage, rather than inferring current from voltage and the lamp's resistance.

In principle you could infer current using a resistance-versus-current, resistance versus voltage, or current versus voltage curve for the lamp, but they also vary substantially with age, as the metal evaporates from the filament.  It also varies with ambient temperature.

The resistance-varies-with-voltage effect was used in electronics in the days before varistors.  Some early tube-type radios had a "tube" that was really a lamp (ran about a dull red) that they used for power supply regulation.  (I think it was to allow it to be used in various places with different supply voltages, such as windmill/battery farm installations.)   Back in the '60s I assembled a Heathkit variable sinewave audio oscillator.  It used a lamp as an automatically-variable pure resistance to regulate the oscillator's signal level by attenuating more as it became higher, so the oscillator would produce a good sinewave at a particular level, rather than build the amplitude until the amplifier element of the oscillator went nonlinear, distorting the waveform.

[Ontheronix]

I'm sorry if it wasn't entirely clear, but I measured voltage and amps in the first to measurements. I did not measure them at the same time though. But I did measure them at the same rpm. Thanks for the advice on the light bulbs, I will not use them anymore to measure output 
Meanwhile I ordered some flux leds. With 10 2,4V leds in series I measure 70mA, totaling 1,68W which is new record. I still have 20 or something flux leds lying around, so further upping the voltage will probably up the output succesfully. I'll test to confirm.

Edit: with 20 2,4V leds in series @100mA, the output is 4,8W!  RPM slightly above 1000 (let's say 1250). I'll test if I can parallel another string of 20 leds tomorrow, but I have my doubts because above 70mA, the amps rose slower.