Charge controller and dump load

[Capt Slog]

I found an old convection heater in a skip, rated at 3kW 240v.  I've been looking for one of these to use as my dump load thinking it might have the bits I needed and when I got it apart I wasn't disappointed.

The uprights/spacers are ceramic. In the above picture I've stripped out most of the heating elements, just one remains, I thought I'd better take a photo before I went any further.

Ghurd sent me one of his charge controller kits, and gave me a lot of guidance.  With his help I worked out what resistances I would need for my dump elements.

I did this by connecting one end of a wire coil to a lab power supply set at 14.5v and then finding how much wire gave me 3A current.  This turned out to be around 72 turns of the existing element, about 5 inches.  I then unwound a little bit of coil and counted another 72 turns. Using the two ends together as one terminal and the centre tap as the other, I now had two parallel resistors capable of 6A.

I mounted three of these doubles in a cut down piece of the heater chasis.

And added a 12v CPU fan to give a bit of extra air flow. At one end of the dump, all the 6 ends come to one terminal, at the other end you can see it's wired as three pairs.

Here the whole thing is mounted on a board.

 

You can see Ghurds charge contoller PCB at the bottom left. This little unit came with all the bits plus a few more, and even the heat shrink needed for the JFETs (even though I used some different stuff as I didn't have Ghurd's supplied stuff with me where I built it).  It went together really well.

I know it's over-kill, but I put three JFETS in this each contolling 6A.  This is partly through lack of knowledge and partly through lack of trust in components; I'd much rather have something running cool than hot.  I'm trying to outwit Murphy here; it's just my luck that I'll get the highest winds ever experienced at my site on the day that I'm not there + the power is not being used + my furling is stuck!

This was only completed yesterday, so I haven't put it on the turbine yet.  It also needs some diodes put over the resistors to deal with any stray inductance made by the coils.  The unit is calibrated to begin dumping at about 14.3 volts.

Thanks to Ghurd for the kit and his help.

[claude]

Nice job Capt Slog,

It seems like Ghurd's tiny dump controller is going to be a success. Seeing it next to the big toys you got there it's like David keeping Goliat under control. :-)

No doubt your setup is built to last. What kind of batteries are you using?

Claude

[Capt Slog]

Hi Claude,

It will be battery (singular) to start with, I told you it was overkill!  It's just an ordinary leisure battery, deep cycle 85 Ah I think, but it will get some partners soon.

[Bruce S]

Hello Capt;

  Very nice!! careful tho, those little david's will get to be a habit!! I'm 5 up and begging G- to send at least 3 more. They are perfect for the HF controllers that have, lately been problematic.

The HF I don't think you can get in the UK, they are pretty good for the price and all the stuff you get with them. The controller is the weak link. Not a real solar controller, but does well for the price. There is a screaming speaker in the unit that once the incoming power has recharged the batts and the incoming is high enough, the little speaker starts an awfull noise. I have begun to replace that with this neat little dump controller and ooh it is perfect for it!!

Your unit will be great to read about when the winds start to get it heating!!

Keep up the great work!!

Cheers

Bruce S

 

[ghurd]

That turned out pretty slick!

I like how the fan is pulling in air over the MOSFET heat sinks.

How hot do the coils get?  Like bright red, glowing, not too hot, warm?

G-

[Capt Slog]

I wouldn't like to touch them, but they don't glow, just get hot and smell like it.

I don't have a PSU powerful enough to test the unit as a whole.  When I tried, the circuit made a buzzing sound, the LED flickered and the dials on the psu stuck at 15v and 1A.  Trying with individual MOSFET/resistor channels worked ok, at just under 15v the dump dumped and the needle went off the (5A) scale. So I'm pretty certain that the problem is the PSU.

[Bruce S]

Capt;

  Have a small question? How tight do you have the hystersis ? with those wires and the possible inductance I'm a bit worried at the dump timing.

I see you have the 10uF cap but am still a little worried about those wires.

Thoughts?

Bruce S

[behoof]

Capt Slog,

Really nice!! Great job and a good read too. Thanks for sharing your project and its assembly in pictures. As they say a picture is truly worth a ton of words.

behoof

[Capt Slog]

The hysteresis is 0.1 volts with the components I put in, and Ghurd tells me that it runs at 50-100Hz.

Are you refering to the wires in the heater?  If so, I still have to add some diodes across each resistance coil to handle any stray inductance, I'd taken the photo before I realised I'd left them out.  If this isn't what you mean, I'm sorry but I don't have sufficient knowledge of the electronics to give you an answer, perhaps someone else will have to step in!

Ghurd also informs me that I should have said MOSFETS not JFETS, (refering to the three heatsinked components below the fan), and that I sould have separate wires powering the circuit to those which supply the dump.

Thanks for you comment.

[Bruce S]

Capt;

   This is a things of beauty, and it pains me to worry so much about it.

I had a closer look ( increased the picture 150%!) .

IS that block under the 'FETs aluminium? instead of what I thought it was, (wood)?

If it's Al then you have them all connected together, which makes it ONE BIG 18Amp unit instead of 3 seperate 6Amp units.

Very cool way to drain off the heat, but if you can seperate those 3 by using seperate blocks of Al. Then check the hystersis :-) you might be a little suprised :-))

Don't worry too much about mis-naming the components , I can't speel worth a hoot!! Your attentions to detail and ability to assemble these items speak volumes.

AND as has been said before much much more is put into doin than talkin :-)

IF by chance that is a block of wood and I still can't see :-D then after you have the diodes. try slowing down the hystersis a tad.

Cheers

Bruce S

 

[Capt Slog]

Hi Bruce,

Well spotted!

I realised what I had done after I posted yesterday, I mounted all the mosfets and their heatsinks onto a thick strip of brass (with tapped holes for the mosfet screws) and then mounted this to the wood. The heatsink tab is also the centre pin.  I think when I first looked at the circuit I got the idea that these were a common connection anyway, WRONG!, as you correctly say, I've connected all the drains of the mosfets together.  They won't like that at all, and I'm going to put it right, (it's not in use yet, I haven't had time to work on it).

As for the slowing down the hysteresis, I'll need advice on how for that.  

Thanks for your kind comments and your interest, it's nice to know that someone is looking out for us beginners.

[Bruce S]

Hey Capt;

  No worries about looking out for one another. Between you and Claude. My solder and hack togethers are pretty shameful.

Once you have the FET seperated look at the hystersis , it should slow itself down.

I emailed G- to put up a small chart, but before you mess with the circuit slip those up and work with the wire layout. The coiled stuff will affect the readings.

I would like to thank you for doing the "just how much will these DUMP" test for us tho. Been wanting to do that, but too much fun setting this up around the homestead. I now ahve enough neighbours that I've helped wit these that I can get 'em to weld some new PV racks up were they'll get 3x the sun.

Keep up the Great work!!

Bruce S