10' machine update

[Janne]

Hello readers!

It's been a few weeks after i got the machine raised up. Earlier post of the machine can be found here: http://www.fieldlines.com/story/2008/1/13/112931/026

Yesterday I lowered it for the first time with the help of a friend. I untangled one of the guys, that got tangled during the initial erection of the machine. Also turned out that some nuts needed re-tightening, especially those holding the rotor together, I guess the wood has crushed some. The blades were also covered in ice and snow, i cleaned them up.. for my surprise the generator was not frozen up.

Anyways, today we had some good winds and i got to make some measurements. What i saw was an average output of 15amps, wind was averaging between 5-6m/s(about 11-14mph). I also witnessed it furling for the first time, when it started to furl it was putting out about 40 amps. I also saw 55 amps for a few times.

So all in all, the furling speed seems right for me, no need to push it more than that, since winds this good are not common here. Pendant cable hasn't twisted up badly in the time it has been up, so i think it is enough to untwist it mayby 3 or 4 times in the year.

What's next is to install the dump regulator i made last week. schematic and a picture can be found here:

http://www.otherpower.com/images/scimages/8104/dumpregschematic.png

http://www.otherpower.com/images/scimages/8104/dumpreg.jpg

i also uploaded a few videos to youtube on the 1st clip you can even see some furling action thought the most violent gusts eluded the machine while the camera was running.

http://youtube.com/watch?v=Kr7Nc2ZzPyM

http://youtube.com/watch?v=t2uWBYQjJRA

[ghurd]

Looks Great!

I have a few issues with the controller, as it is shown.

About the mosfets. Do they get 5V at the gate?  Are there only 3 mosfets for 40 to 55A?  They supply a single load?  Are there gate resistors for each individual mosfet?

I worry too much.

G-

[Janne]

Hi ghurd, and thanks for your concern. I haven't measured the voltage at the gate, but they started conducting when i tested it out. They all work to switch the single load on / off. No individual resistors. They are rated at 50 amps each, so i'm thinking 3 of them would be ok right?

I really am not very experienced using Fet's , so it might very well be that this thing still needs some developing.

[ghurd]

I honestly expect the fets will blow up, leaving the windmill unloaded, and then it might blow up.

I believe the circuit you made is repairable, but it would be a lot of work or money.

Is the dump load a single resistor?

"dump piv a stus" means nothing to me.

G-

[Janne]

Hello again!

the dump load is a single resistor.. made of 6 strands of thin steel wiring. Are you thinking, that the fets would blow up because only one of them would end up conducting at a time.. until the one conducting would blow up? If so, i think the situation could be improved by separating the wires into buches of 2 and wiring each bunch to their own fet. I can't think of any other reason why they wouldn't last, since i think a fet rated at 50 amps should handle 20 amps perfectly ok?

Anyways, because of these concerns i'm going to put the whole dump system to a more throught out test before putting out onto it's own. And thanks for your advice so far.

[martin1]

It seems that you only have 5 volt on the Gate. I'm not sure such a low gate voltage will handle 20 amps. The Rds(on) will be high and a lot of heat will dissipate in the fet. Check the datasheet.

http://www.datasheetcatalog.com/datasheets_pdf/I/R/F/Z/IRFZ44.shtml

[joestue]

Yup.

Rds (on) is too high at only 5 volts.

replace that 5 volt regulator with a 12 volt one and you're good.

[Janne]

Hi guys!

I think i'll just swap the 3.3k pull up resistor to the line coming from the battery, that should solve the problem of too low gate voltage.

Thanks again for pointing me in the right direction. Janne

[ghurd]

I believe they will still only have 5V at the gate.

It would be best to get at least 12V.

I am not sure if changing the 7805 to 7812 will work unless the 7812 is a low drop out version, or the set point voltage is over 14.2V.

I would separate the 6 wires, and use 6 fets, each on its own heat sink.

G-

[joestue]

Please don't use 6 Fets unless you have an infinite supply of them.

My understanding was that 78XX regulators passed current if the input voltage was above the dropout voltage. (We don't care about regulation here, but now you would need a different voltage reference)

A simple solution would be a separate 12 volt zenar for the gate supply.

[ghurd]

Three more fets? It's only $5 worth of parts to avoid a LOT of potential problems with balance.

For under $20 total, I would have sent him one of my kits and 6 fets.  Even to there.

Most 78XX pass current below XX input, but the Vf will make the already marginal gate voltage even lower.

A couple TO-92 fets and 3 more power fets should fix both problems.

Or logic level power fets.

G-

[Janne]

Hello again!

I just connected the pull up resistor (R7)directly to the line that comes from the battery, now i'm measuring 14.5 volts at the gate when the cirtuit starts dumping. I also mounted the 3 fets on a rather large heatsink... i tested it for a little while and currents through all fets didn't vary wery much. The heatsink however got rather warm, but not hot though. It might work ok as it is... i'll be doing some more testing when i have better time and then decide if it needs improving.

Thanks for all the advice so far. More to follow.

[ghurd]

The fet current variation happens VERY fast.

One turns on first.  One turns off last.  During these periods, one fet will carry all 60A.

It would be safer to use 3 large heat sinks.  On a single heat sink, the fet drains are connected through the heat sink.

Make each fet feed 2 wires.

The variation of current and timing would no longer be an issue, because the loads are seperated and balanced for each fet.

G-

[CmeBREW]

Congrats on your nice 10 footer Janne.  Looks really good. Thanks for the videos.

40-50amps is like a dream to me!