Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Art Welter]

125watts from solar, and 250watts from wind "day round" (peak, ave, other?). You suggest a load of 1-2 ohms. A voltage range of  10.5V min for regulator operation, max V of 18V(?).

FWIW 18V is only 325W into 1 ohm load so if power output is capable of 18V that sounds ballpark.

If an external (hard) 10.5V power supply has a steering diode in series with it's (11+v) output, it can supply current only when the solar and wind is not. Once the solar and wind is putting out more than 10.5 amps at 10.5 volts, the power supply diode will stop conducting and the load is completely driven by green energy. 


JR

Yes, 18V is 325W into a 1 ohm load, which is why I am presently using a 2 Ohm load for the 125 watt (peak) solar panels. When the generator kicks in and raises voltage to around 17, the 2 ohm load is dissipating around 144.5 watts. Panels no load go to around 20 V, but with  max sun in to 2 ohms they do about 14.5, when the generator kicks in with screaming wind it can hit 17 V (in addition to the panels), which is it's limit on paper.
An "unobtanium" load that would start at around 100 ohms at 10.5 volts and drop to 1 ohm at 17 volts would be ideal.

The wind generator is really variable, it would only put out 250 watts at 25 MPH, but regulates to 200w at speeds above.
Takes almost 18MPH to do 100 watts. At 10 MPH, it only does about 20 watts.
Only 38kW, about $4 worth of energy per month at 12 MPH average wind speed.

You answered the question I posed in #3, if I get what you are saying, my Micronta 13.8 V 8 amp PSU should be safe to drive the load and not be damaged by the higher voltage from the panel and generator if it goes through a beefy diode.

Would you suggest using a diode on both the + and - of the PSU, or is one on the + sufficient?

Thanks,
Art

[John Roberts {JR}]

You only need one diode with anode toward PS and cathode toward load. If that PS is adjustable you could extract more energy from solar panel by dialing down closer to 10.5V min. You could stack some diodes in series with 13.8V supply, while diodes would get hot from extra watts of dissipation, they would more efficiently extract solar power.

A little rube goldberg, but a temperature switch on the diode heat sink could detect when heater is running from PS only based on heatsink temp and cut it off. Unfortunately these are generally self resetting so it would cycle itself to death. A resettable circuit breaker in series with power supply might work if sized properly. It needs to hang long enough to start system, or you could use a start switch that bypasses the breaker to start. Of course resettable breakers are not rated for lots of cycles either, but this would be low tech solution. The diode(s) would be dissipating several watts max so enough heat to heat up a small heatsink. 

====
You could actually chop the load with a HF switch (like mosfet) to make it appear like a lighter load at low voltage.  At 50% duty cycle 1 ohm looks like 2 ohm, etc. That is the technology behind that speaker protection add on gadget, it scuibs power away from speakers by duty cycling the AC voltage fed to the speaker driver. Of course you would have to roll your own controller. About $10 parts, but way too much design effort for a one off. Of course you could roll a thermostat and time of day controller into this load controller making it slicker than snot on a doorknob.  or not...

JR
 

[Art Welter]

You only need one diode with anode toward PS and cathode toward load. If that PS is adjustable you could extract more energy from solar panel by dialing down closer to 10.5V min. You could stack some diodes in series with 13.8V supply, while diodes would get hot from extra watts of dissipation, they would more efficiently extract solar power.
A little rube goldberg, but a temperature switch on the diode heat sink could detect when heater is running from PS only based on heatsink temp and cut it off.

You could actually chop the load with a HF switch (like mosfet) to make it appear like a lighter load at low voltage.  At 50% duty cycle 1 ohm looks like 2 ohm, etc.  About $10 parts, but way too much design effort for a one off.
JR

The "free" power is only providing a portion of the heating needs, so if the PSU is doing the heating when the sun and wind aren't, no problem.

If you feel like designing the load chopper, I'll pay you 3 months worth of the energy cost it saves  .

Thanks for the tips, off to Radio Shack for some Schottkys.

[John Roberts {JR}]

The "free" power is only providing a portion of the heating needs, so if the PSU is doing the heating when the sun and wind aren't, no problem.

or the diodes getting hot :-)

If you feel like designing the load chopper, I'll pay you 3 months worth of the energy cost it saves  .

I have designed my own DIY heat gadget but that's different...

Thanks for the tips, off to Radio Shack for some Schottkys.

Schottkeys are metal junction low forward drop diodes. For your application the more drop the better (I'd be tempted to throw a few in series ) something like a power supply diode bridge can be wired as two diodes in series and is easy to heatsink.

JR

[Art Welter]

Schottkeys are metal junction low forward drop diodes. For your application the more drop the better (I'd be tempted to throw a few in series ) something like a power supply diode bridge can be wired as two diodes in series and is easy to heatsink.

Well, RS does not carry Schottkys or high powered diodes, so I'll be using four 3 amps in parallel (all they had, I thought of what you said about series, and went back for more, the cupboard was bare)..

[John Roberts {JR}]

Well, RS does not carry Schottkys or high powered diodes, so I'll be using four 3 amps in parallel (all they had, I thought of what you said about series, and went back for more, the cupboard was bare)..

Not to ruin your fun but diodes in parallel will not share nicely... the hotter diode will have a lower forward voltage so get a larger share of the current, making it hotter, causing the forward voltage to drop more, you can imaging how this ends (melted silicon).

The technical term for making parallel diodes, or whatever share, is to use "degeneration" resistors, something like .1 or ,22  ohm in series with each diode will swamp out the bad sharing behavior caused by temperature coefficient.  A .1 or .22 ohm will also drop some more voltage which is a good thing in your case.

I would still use a !0a diode bridge or something like that, two diode drops in series.   

JR

[Art Welter]

Not to ruin your fun but diodes in parallel will not share nicely... the hotter diode will have a lower forward voltage so get a larger share of the current, making it hotter, causing the forward voltage to drop more, you can imaging how this ends (melted silicon).

So far the diodes in parallel seem OK, hot to the touch but well under 212 F.
The RS PSU also seems OK powering the  2 Ohm load through the diodes, voltage is holding at about 11.8 after dark. No wind, of course- seems the wind has been far less ever since i put up the generator...
The loaded PSU makes a bit of noise in the stereo, but that goes away when the sun shines  .

[John Roberts {JR}]

The old spit test for 100'C is OK, silicon doesn't melt until well above 150' C while some plastic packages degrade before the junction melts.  If all the parallel diodes are roughly the same temp they are matched well enough.

JR

[Art Welter]

The old spit test for 100'C is OK, silicon doesn't melt until well above 150' C while some plastic packages degrade before the junction melts.  If all the parallel diodes are roughly the same temp they are matched well enough.

JR,

Well, the epoxy diodes lasted the night, and all still seem the same temperature.

At the present wind velocities, the $4 diode investment should pay for itself by the end of the winter season .

Thanks for the help, the goal of little expenditure (for a little gain) has been achieved!

Art

[John Roberts {JR}]

JR,

Well, the epoxy diodes lasted the night, and all still seem the same temperature.

At the present wind velocities, the $4 diode investment should pay for itself by the end of the winter season .

Thanks for the help, the goal of little expenditure (for a little gain) has been achieved!

Art

So you don't want to pay me $100Hr to design up a custom gadget for you... ?

You could squeak out a little more green energy by adding those resistors in series with the diodes. As long as it still makes enough voltage to just turn on the regulator you get that extra low output solar and wind. Resistors should be cheap while the gains will be marginal, but real.

Of course you are not billing for your time either.

JR

PS: I just tweaked my DIY smart electric heater. I was using a very old auxiliary electric heater, connected to my smart controller. The old 650W heater came with my house, so it was old 30 years ago.  It still makes heat, I had to replace the on/off switch that crapped out maybe 15 years ago, but my complaint is that the heating wires sing and rattle when the are heating up. Not a loud noise, but loud enough in a quiet bedroom in the middle of the night. I tweaked my heater to duty cycle 20%, 40%, 60%, 80% or full on based on how far below the threshold temp it was. This knocked down the old heater noise by a similar fraction. So I decided to see what modern technology has to offer. The cheapest wallmart aux heater, for $19.95 makes 2x the heat, has a fan blower, and a thermostat. Better yet it has a tilt safety switch and thermal overload fuse, so safer too.  The new heater doesn't  make any wire rattling noise, but the fan motor didn't like the 20% speed. It thumped at like 30Hz, so I added 150 ohms in series with the motor, that quieted it down and slowed it down at max heat... I now have twice the heat if/when I need it, and silent running. 

[ProSoundWeb Community]