Hi folks ...

I am about to start experimenting with an alternative energy project, the details of which are not particularly important save to say that there is an alternator driven by a small water turbine.
The available power will vary quite a bit and I have decided to use a shunt regulator for two reasons …

1. It will not drop any voltage when the available power is low.
2. It will reduce overspend and hence over voltage problems.

I would like opinions in respect of the following configuration options.

A voltage sensor with a small hysteresis which will create a free running stream of pulses at no particular frequency or pulse width into the dump load.

A PWM generator with the duty cycle being varied over a small voltage range, essentially the same range as the hysteresis above.

I appreciate that PID is a better option but don’t want to get that complex as the applications loads will not require a closely controlled voltage.

The target is 24V nominal and I would like to get within +/- 0.5 if at all posible.

As always… any and all comments welcome.
Thanks

 

I'v answered my owne question .....
Since I need to know duty cycle in order to calculate averidge power it is going to have to be PWM whilst I am testing I think ....

If anyone knows differently I would apriciate any comments.

Once I am set up knowing how much power is being shunted will not be important so at that point the originnal question still stands.

Al

 

I think the free running will be enough, just simply crowbar the output into low resistance as long as it is over the voltage limit.

 

"crowbar" now thers a blast from the past ...... Did you used to have a CB?
Or was it us sad folks hijacking some existing terminology?

Maby I could fit a 100W burner and then waste all the power in my MatchBox

Al

 

Nope, guess I am too young for CB.
Now that I read your first post more carefully...
Are you using a car alternator? Made for 24V? AFAIK these are regulated by setting current in the field winding, which might be easier and waste less power than shunt regulation.

 

No its a burpose built permanent magnet unit designed to achieve 24V at 500 RPM.
Not cheep either I might add.

It is capable of 1300W but my input turbine, given it head and flow, will never achieve more than 350W, even in flood conditions.

My power requirement is sub 100W so I will have enough water to do what I need all season but I want to keep the turbine loaded and warm in addition to holding the voltage down.

If I were to use a seriese reg the turbine would run seriously fast when water was available, wasting the water and leaving me with a silly high voltage to deal with.
The power disipated would be greater with a seriese reg because the turbine would be running faster and taking a greater flow ...

If you stall a turbine the increesed back-pressuer reduces the flow and therfore the available instantainious power whilst letting it run overspeed increeses frictional losses.

Keeping the thing apropriatly loaded will therfore maintain a ballance, maximising usable power and minimising water useage.

Obviously controlling water flow to control power is the better way to go but this is a little axillary turbine at the inlet of the headrace, supplying power for wireless monitoring, so active control of water isnt a viable alternative.

Shunt is definatly what I need, I am just attempting to decide exactly how to impliment it

Thanks

Al

 

Do you have the turbine allready installed, so that you could do some tests on site?

As it is an AC machine, I would first rectify it. Then through a diode to some small, high voltage capacitor for the regulated auxiliary supply for the comparator, probably just with a standard 78xx.
The comparator should measure the instanteneous voltage coming from the main rectifier, and turn on a resistor to load the alternator. The resistor should be low enough to make the voltage sag.

You say you will be operating the alternator at lower RPM, are you sure you can get enough voltage to get the 24V in such conditions?

 

Instaled no, but it will be hooked to some 4" over one of our lacal weirs as a testbed ... testing somwhere easy is a must given that it will be up a welsh mountain when it done.
Your circuit description is prety much where I have gone, the resistor is a hugh WW capable of disipating several hundred watts and is just under one Ohm.

I have gone with a 78xx reg for now but that might be a problem I need to address befoer I fit the production circuit.
78xx regs will only handle 37v according to the spec so if the alternator was already overspeed when the circuit is connected it could be an issue.

The Alt is here http://www.energistar.com/ click PMG's and then power curve for 24v unit.
It dosnt actually give an open circuit voltage but it is probably hundreds at high RPM, an interesting design challange dont you think.

The circuits / modules I have now, which can be configured for PWM or free runing pulsed load will be fine for testing but I am sure that connecting them to a overspeed alternator would kill them long before they got the system under controle.

If you have any thoughts on that I would be interested ... When I am testing I will establish exactly how high a voltage I have to deal with should I need to capture an uncontrolled turbine.
Its a problem you just dont get with a 'Normal' system because you would be controlling the water flow but I am sure if I dont design for the scinario it will come back to bite me.

Al