controlling the dumping

[Jason Wilkinson]

Hi to all. a friend and i were discussing the dumping of load from the battery  and he came up with this idea , since he don't like the idea of cyphoning of  current when the batteries are full. I thought i'd run it by THE BOARD  and hear your comments



hi jason

the aim of this project is to design a dump load circuit that will not discharge the battery when connected to it ,but to take the excess power from the turbine and redirect it to low value resistors (aka dump load).

my proposal is to place a high current diode between the three phase rectifier  and the battery.next connect the dumpload at the anode of the diode .this serves three purposes .

1) to keep the turbine always connected to the battery.
 
2) to allow the voltage to the dump load to at or about that of the battery voltage.

3) create a low resistance path for the dump load

my thinking is that with arrangement it will not interfere with the operation of my mx60 charge controllers as they move from bulk/absorb/float operations
 
the voltage settings for the dump circuit will eg. high end same as absorbe voltage .and low end below the float voltage of the outback mx60.

give me your thoughts on it.

regards
 
ken


Jason


























 

[wooferhound]

A properly designed Dump Load Controller will not discharge a battery, it simply sends the excess power to low ohm resisters. While the dump load is operating the battery is not disconnected from the Turbine/Solar Panels and remains floating at the set dump voltage as long as more power is coming in. It's operation works almost exactly as he is describing that it should work.

Like this . . .
http://www.fieldlines.com/images/scimages/231/HookUp.GIF

[TomW]

Jason;

I think your friend has a misconception about this process and how it should work for wind power anyway.

Might point him to a few posts here on the subject to save him reinventing the process from scratch and avoiding the pitfalls encountered by others. A lot of the "gotchas" have been encountered and dealt with and they will be explained if he does some research here.

Just saying ..

Tom

[ghurd]

"my proposal is to place a high current diode between the three phase rectifier  and the battery.next connect the dumpload at the anode of the diode .this serves three purposes ."

That will not work.  Not even a little tiny bit.
G-

[Jason Wilkinson]

Thanks guys. He's still not convinced.  Hear him  "" well tell me why it can't work"
   
  Jason

[Flux]

He is right, the proposal as described will work if you do it correctly. There is normally no point in having the series diode in normal operation but it does prevent battery discharge if for some reason the dump load fails.

I have used the scheme very successfully with linear dump loads on more than one occasion for several reasons which are irrelevant here.

I suspect this is not straightforward with pwm dump and that is why you are getting negative responses but at first glance I can't see why it won't work even in this case.

This is not the same as those crazy schemes that disconnect the turbine from the battery and try dumping directly from the turbine, I run a mile from such ideas.

With pwm the effect is in a way that of dynamic braking of the turbine during the current pulses and this is the bit that I am not happy about but as long as the pwm frequency is not too low I suspect the alternator will average the current in just the same way as the battery. You may have to take a few precautions with spikes from stray inductance and you may need snubbers unless you use a fast or schottky diode for the series unit.

Whether it is really worth worrying about something that is nothing more than a fault condition I don't really know, the number of ruined batteries from dump load failure is something I have no data on.

For what it is worth I have also used a relay for this same job and a circuit enables the dump load at something like 13.4v. At normal control voltage the scheme is absolutely conventional. If the dump load fails the battery will quickly drop low enough to open the relay and it will not be flattened. You may end up with a race with the relay going on and off and it may not be a good idea for an unattended system but for something that is observed you would have plenty of time to intervene. Even in a remote scheme you may be better to scrap a relay than kill a battery. The series diode is simpler but has issues of current handling , heat and loss for big systems.

Flux

[zvizdic]

Whether it is really worth worrying about something that is nothing more than a fault condition I don't really know, the number of ruined batteries from dump load failure is something I have no data on.

Flux

I tested GHurd 48v controller for 2 months with paralleled Mosfets and mosfet  burned draining a batteries to 26v .

Lucky I used old batteries  now with  new ones I like to have some protection and might try diode or relay(relay likely).

If dumping power relay is not a big drain but if a diode would work it be silent.

I will tray and post a findings

[rossw]

If dumping power relay is not a big drain but if a diode would work it be silent.

Seriously, if you're going to use a relay, why not use a double-pole relay (or a 3-phase contactor) and dump on the AC side of the bridge?
This is undeniably the most robust solution, doesn't require an additional hefty blocking diode, creates less stress on your main bridge, can't discharge your batteries even in a failure condition, etc, etc.

Has the (arguable) benefit of not putting all your eggs in one basket with the dumpload too, you make it in 3 parts - one for each phase. Even if one fails, you still have SOME load on your turbine, even if it is only 2 phases.

[zvizdic]

Relay is not to dump power it  is  just a safety switch if Mosfet fall .
But if is possible to dump power at AC side, then diode  should work .

For what it is worth I have also used a relay for this same job and a circuit enables the dump load at something like 13.4v. At normal control voltage the scheme is absolutely conventional. If the dump load fails the battery will quickly drop low enough to open the relay and it will not be flattened. You may end up with a race with the relay going on and off and it may not be a good idea for an unattended system but for something that is observed you would have plenty of time to intervene. Even in a remote scheme you may be better to scrap a relay than kill a battery. The series diode is simpler but has issues of current handling , heat and loss for big systems.

Flux

[Jason Wilkinson]

Thanks guys Ken and i have;nt spoken since .but he'll hit the roof after reading "flux"
 Jason

[zvizdic]

Works good so far .
I only have low wind with some gusts and it switches nice and slow better then before.
Going to stick with it for a wile .

[zvizdic]

Here is video.
http://www.youtube.com/watch?v=ddehq57syiU

[ghurd]

Here is video.
http://www.youtube.com/watch?v=ddehq57syiU

Looks good!
G-

[oztules]

Jason, I think Ken is on the right track here.almost anyway.
My own turbines do not have dump loads, they are connected to an electric car, when the car feels it is fully charged, it shuts the turbine off..... but...

The AWP here uses almost the same kind of thing Ken alludes to.

Instead it uses effectively two three phase rectifiers, one to the batteries as normal, and one to a pwm dump load. ( they share one half of the main bridge as an economical solution to two rectifier banks).

It is simple, it works, and has found to be bullet proof. It cannot discharge the batteries, and makes the bank look bigger if the voltage goes too high..

Not sure what the fuss is about. Using a diode in series as Ken has explained, adds another half volt loss, dual rectifier system does not..... and seems more elegant, as all the DC power does not have to go through only 1 diode.



.................oztules