Window Motor Doubles Voltage At The Power Supply [Overunity ?]

[DadHav]

Hello Gyula, Cat. Thanks for showing interest in my video. You're certainly welcome to have a look at my other videos as well. I think Gyulas option #2 is probably close to what's going on. I'm not knowledgeable enough to hit the nail on the head but I can take a SWAG at it. (Sophisticated Wild Ass Guess) I think It might go like this: The output from the bridge is a few volts higher than the input voltage; For some reason there is an impedance match with this power supply that I can't seem to match with a battery or anything else; The special match allows the higher voltage from the bridge to overcharge the filter capacitors in the power supply to a higher voltage than the supply is outputting; It may be possible that part of the return is out of phase from the input pulse of the motor as well; Since the capacitor is now higher, the motor tries to use the increased voltage and does; The motor speeds up and produces yet a higher charge to the capacitor; The motor might be efficient enough to let the cycle increase rather than draw the capacitor down to the power supply voltage. OK, Just a theory. I don't think the scope will show anything across the output of the supply because of the heavy filtering. AND, the current also goes up as the process goes into acceleration. at least the power supply meter indicates this when switched to amps rather than volts. So, I don't think this is over unity but more like a special effect related to an unusual impedance anomaly. What do you think. Old Dad is full of prunes right?
But wouldn't it be cool if I could get this to work with a battery and have the extra voltage trying to charge the battery rather than the capacitor of the power supply?
PS Just a note on the efficiency of the motor: This one will run at about 200 micro amps. If you have time, take a look at the capacitor run tests. I think it's one or some of these.
http://www.youtube.com/user/DadHav#p/u/0/xBpir5azwTo
http://www.youtube.com/user/DadHav#p/u/1/FqE_mpNPIZg
http://www.youtube.com/user/DadHav#p/u/2/T5HrqoLCNCI

DadHav/John H

[poynt99]

I think the simplest and most likely the correct explanation is that the output from your bridge when connected to the supply, is fooling the supply's output current sensing and/or voltage regulation circuitry. It is either seeing a drop in current or the voltage sense is being fooled by spikes such that the supply is cranking up the output voltage to make the "correction". I am aware that your supply is only supposed to be limited to 20V output, but it will actually be capable of more than that in order to compensate for its own output impedance and to regulate the output voltage properly with the current spec.

You did say that the supply's output current goes up in this mode of operation, so this all makes sense.

It would be great if your window motor really was adding that much power to the input, but I highly doubt it is based on what I've said above.

.99

[powercat]

Hi John
Thank you for responding so quickly and welcome to the forum, I think you will enjoy it here as there is a wealth of knowledge and many experienced members.

cat

[DadHav]

Thank you for the welcome Cat. .99 Your suggestion as to what's going on may very well be correct. Just a few notes though: I have a friend that has nearly the same thing happening with a simple toy train transformer with no capability of current or voltage sensing. Also I'm not using the sensing circuits on the Lambda supply. I would really be surprised if there was 200% reserve power just to adjust for voltage or current losses as it would pertain to normal usage. I don't know. The more I think about it the more I think I would be disappointed in my power supply if it could change its mind about what I have it set to put out for voltage and all of a sudden put out twice as much as it's supposed to have the capability of. Oh, there's another thing I haven't mentioned: 42 volts wasn't the limit. I could decrease the trim resistance further and get more voltage, but I don't have heat sinks on the transistors and they where starting to heat up.  Hey maybe I could get an opinion from an engineer at Lambda or something. Another interesting test would be to run a a parallel load to the motor and see what it takes to draw the voltage back down to 20 volts. Maybe thats what someone was suggesting already.
Thanks guys. I appreciate the input.
DadHav/John H

[tinu]

@ DadHav/John H

From my experience, a window motor shows increased efficiency when energy (back emf) is properly recovered. I have no doubts here. Actually, most if not all pulse motors will do so. Hence, a certain/reasonable increase in speed at constant voltage (constant power) is explainable. In practice, because the input power decreases when employing the recovery circuit, a slight increase of voltage can be measured, which is also normal since most power supplies shows a I-V dependence. Usually this behavior is not properly explored, understood or maybe just not reported back by some former experimenters I’ve followed; things get complicated because until motor reaches a stable point on its functioning curve, we’re dealing with a transient.

Regardless of the above, imho the voltage can not become double after connecting the recovery or the motor would fly apart given its characteristics and efficiency. Therefore, I suspect the voltmeter on the supply is tricked due to voltage spikes. The ammeter might be tricked the same way as well.
I kindly suggest repeating the experiment using several 9/12V batteries and a mini-variac to properly adjust the voltage and then to calibrate everything (DC vs AC/power supply) using the VDC-RPM curve.
Please report back if you decide to try solving the mystery and I apologies for interfering with other suggestions; I’m not an expert in this area.

Awesome built!

Best regards,
Tinu