Wind Turbine Upgrade - 7.5 HP Conversion

[SparWeb]

Ever since last year's new tower build, I've enjoyed power from Spirit of Zubbly.
https://www.fieldlines.com/index.php/topic,150448.0.html

That wind turbine was built in phases, where the first phase was the conversion of a 3HP motor into a generator, then carving of 10' diameter blades from laminated cedar wood, and finally a 70' truss-tower.  It worked great, so I should be satisfied.

But nothing is perfect, especially not me.  So here I go again.

This is Toshi, a 7.5 HP motor that I converted into a generator a very long time ago (2008!).  It's been sitting on a shelf ever since, waiting for its day in the sun.  Time to do that.



Basic stats:
Toshiba 7.5 Horsepower nameplate rating
Totally enclosed fan-cooled
Original ratings at 230V parallel / 460V series
Reconfigurable wiring to star and Delta
4-pole winding, 1740 RPM rated speed
125 pounds, cast iron case.

Here's the final assembly process as the rotor was gently lowered into the stator.  I say gently but I mean "as much as possible".  With 24 big magnets it went gently until the forces ramped up, stretched the suspension rope and slammed into place.  Never coming out again!



I could recap the conversion process but frankly it was so long ago I don't remember much!  I posted about the process on Fieldlines at the time but I can't find it now (before 2010 searches get sketchy).  I started a page on my site but I never finished it...  http://sparweb.ca/toshiba/
I guess you can say this motor-conversion hasn't got the attention it deserves until now.

Probably the most important thing about Toshi is that I thoroughly tested it using a lathe to drive the shaft against a variety of electrical loads.  I ran it up to the peak power of the lathe, pulling 2kW out at times.  Those tests will be extremely valuable because the data will let me corroborate what I'm measuring when it's up on the tower.



This summer I have been gradually building parts to mount it into Spirit of Zubbly.  I won't take credit for much originality here: I have a mount system that worked well for the Baldor 3HP conversion which I'm replicating for the Toshi conversion.  Everything is a bit bigger, but otherwise the same.  I'll post pictures so that you can see how it works.


I've actually been working on it for a while.  Not posting about it, because I'd rather be in the shop than at my computer these days, please forgive me.  So this is a bit of catch-up on my part.

There are several important topics on my mind as I plan to put Toshi into Spirit of Zubbly.
1) Matching an oversized generator to my my 10' diameter blade set will be challenging.
2) Experimenting with MPPT control
3) Reconfiguring the battery bank to 48VDC
4) Scaling up my tail furling mechanism to see if it works as well when it's 50% larger
5) Datalogging to confirm performance predictions of mismatching and matching
6) Selecting a new diameter of blades to build, based on measurements to match the generator
7) Experimenting with non-battery loads, such as resistance heaters

Not necessarily in that order.

[Bruce S]

SparWeb;
Any possibility of put up another tower?

Bruce S

[SparWeb]

No new towers in the future.  The 7 things in that list will keep me busy for a while yet!

This is a replacement of the Baldor generator, which has given almost 12 years of service.

Mounting the bigger generator makes everything in contact with it get bigger.
The hub needs to fit the 1.38" shaft



And it needs a new base frame to mount on.



The frame is a tubular rectangle, with sleeves welded into the holes where the bolts go.  One corner is cut away and welded to the stub tube that it yaws around.



The stub tube fits over the tower stub.  It has a cap to stop the tower going through, with a hole for the power cable.  There's also a bearing that goes in side to help it swivel smoothly.



Here you can see parts of the hub, generator, base frame, and a little bit of the tail together.
The point of balance is several inches ahead of the tower stub, but when the tail is extended the point of balance is right at the tower.
The offset of the generator from the centerline of the tower is about 5 inches.  It was 4" for the Baldor.

[SparWeb]

The tail design and geometry also isn't too different.  I didn't want to re-use the same tail because I will want a bigger prop in the future.
So this one had to be bigger but not much heavier.

[kitestrings]

Wow, great progress already.  This should be fun to see take shape.

Say, what is different on the root end of your blades?  Is that just the unfinished portion covered by the hub plate, or is there another material?  It almost looks like coarse sandpaper.

[bigrockcandymountain]

This is coming really nice.  You have a lot done already. 

What is the length of the grip area that the hub sandwiches on your blades? This is an area that i tend to stress about, almost to the point of lunacy. 

Otherwise, I'm just enjoying the watching.  I'm pretty sure it's going to be a very successful machine.  I bet the magnets to convert it today would cost 4x what they were when you built it, so it's good you did it when you did.

Do you have an inverter lined up for 48v?

[SparWeb]

Hi.
I got in my time machine and...

Actually you're seeing several months worth of work in my pictures.  I haven't posted because it's been hot...  not kidding, the shop has been nicer than the house this summer.  Computer is in the house and the welding torch is in the shop.  Easy choice. 

KS,
The blade roots have cork on them.  I added that as a damping material against vibration.  Given the pulsation from my motor-conversion generators, I didn't want any of it turning into noise in the blades.  I thought the cork would peel on and off during subsequent disassemblies, but in fact it seems pretty firmly bonded to the wood/varnish now.  Another odd thing you can see is that some of the black paint of the hub plate peeled off also stuck to the cork.  That reveals that the so-called powder coating finish that I paid for wasn't actually powder-coating.  Or a powder-coating that wasn't worth paying for.  However you see it, I have to re-paint the hub plates.

BRCM,
Here's a picture of the blades assembled.  What dimensions are of interest?  If you want to see the bolt pattern, I can share a drawing detail.
The grips are about 3.0" thick, 7.25" wide, and they are flat from the center point to about 6" radius.  They are cut out to allow the shaft and hub fitting to pass through.

Swapping from the Baldor to the Toshiba, I went from a 1.13" shaft to a 1.38" shaft, and the hub around it got bigger in proportion.  In the hub photo above you are seeing the combined fit-up because I had to cut more off the root of the blades to fit the bigger hub through.

This is an area that i tend to stress about

Pun of the day. 

No, my balance of system is not ready for this yet.  I have to do it, but focused on building while the weather is nice.  I don't absolutely need a Midnite Classic for this to work, but it would be nice and fun to experiment.  I do need an inverter, and watching Kijiji regularly.

[SparWeb]

Also on the subject of inverters, I'm going to bump up a post I made before:
https://www.fieldlines.com/index.php/topic,150644.0.html
Hoping someone has experience with the Outback Radians.
Specs look good, but it doesn't seem to be a go-to choice, and I don't know why.

Since I am buying an inverter, I would prefer one as up to date as possible, to lower any hassles with outside agencies that might have something to say about their power lines.

[bigrockcandymountain]

Yep, those measurements are all I was looking for. Is the hub just a set screw to hold it on the shaft?  1-3/8" is the same as my 5hp here.  It has a nice "not going to hurt it no matter what you bolt on" feeling to it. 

I had an off grid outback inverter 24v to 120v that I picked up a few years ago with a used solar panel deal.  I didn't have a use for it, but i tested it and was very impressed with the build quality.  I ended up selling it. 

So I don't have any experience with the radian, but i would definitely think the quality would be right up there with the best.  Their "mate" controllers are very complicated, but capable of adjusting way more things than the Magnum that i have. I think you would appreciate that.  I'm not positive that the mate goes with the Radian, but I think it is compatible with almost everything they sell.

[SparWeb]

The hub really is just 3 typical keyed shaft hubs, welded together.  Yes, they have set screws.  Not content with just that, I also have a 3/8" bolt that threads on the end of the generator shaft to keep the front plate of the hub on, too.  But that's not all - I've been concerned about the hub plate being able to work out the 3/8" screw so I made up a larger plate for it, and it is screwed down with a pair of 5/16" screws, too.  In total about 6 threaded fasteners have to come loose before the blades can come off of this shaft!

[bigrockcandymountain]

Oh haha I was wondering if 1 set screw would be enough but didn't want to come right out and say it.  Sounds like you've overkilled it in a big way.  That should hold together through almost anything. 

Nice job. It looks like you are finding lots of excuses to use the lathe too.  That's always a good thing.

[SparWeb]

Yup, the lathe's been busy for this project; just the way I like it.  It was a couple of years fussing around, giving up, buying more tooling, etc. before I really had it set up to be effective.  It takes time to set up a machine like a lathe and it's not a myth that the price of the lathe is equal to the price of the accessories and cutting bits.  Now 4 times out of 5 I can just turn out a part I want without being stuck needing another widget.

[SparWeb]

The "balance of the system" sunk into my brain today.  I wrote down everything I need to change (and to buy) and the list is pretty long.
Batteries go from 24 to 48 Volts.  Sounds like just a rearrangement of the terminal bars but actually the cells are in the wrong order to do that so I'm actually going to need to make fat jumper cables.  I wish I could just pull them out of the stack and flip them around, however these batteries are in steel cases and firmly wedged in.  Fortunately the reconnection will put them all into a single series string, so jumpers can't cause nasty problems like they could in two parallel strings.
That leads me to need new busbars, too.

...and until I get a Classic the wind turbine will be hard-wired to the battery stack, therefore another diversion load is needed.  The dump load for my Baldor/24V combination just won't pass muster and I have to build another one.

[bigrockcandymountain]

I'm still in the first 2 years of my lathe and still run in to lots of tooling related snags.  It sure is fun though, and I'm getting some things collected for it. 

So far, the only things I'm really lacking are good boring bars and possibly some carbide insert tooling.  I've been enjoying learning about grinding hss, but especially for threading, some inserts might be nice.

I made first chips with the metal shaper the other day, so i can now do flat work.  It seems like the learning curve on the shaper is way steeper on the shaper, so it will take some time to get very useful. 

Balance of system just means "no fun and lots of time and money to be spent".  It is the difference between a well done system and a crappy one though, so lots of motivation to get it right.

I wish you lived closer.  I have a hydraulic terminal crimper that would work good for you to build some nice jumpers.  I only use it about once a year to do new battery cables on old machines.  It's crazy how much better they are than the bolt on hardware store battery cable ends. 

[ruddycrazy]

Being a machinist all my life I do have a good selection of tooling and when my dad died I ended up with close to 9 ton of his stuff where about 2 ton was machinery and the rest tooling and a library of books. Now living off the grid coming up with a solution to get the best out of my machines I put a VFD on each and programmed each one to suit the motor, now with my German toolroom lathe I set the hertz up for 100Hz just like my Bridgeport mill so I can double the speed out of the motors. Going the VFD route is less stress on an inverter as the startup current is the same as the running current as pre programmed into the VFD's.

Now I do make a lot of my own tooling and when I secure a carbide tip I just machine up a tool holder for it then after a real good clean including being as anal as pouring 90% neutral moonshine as the last cleaning step and letting it evaporate off by using a gentle heat then the tip is silver soldered using flux and just enough in the right place to do the job which results in no cleanup of the joint needed. My next tool will be a 20mm radius tool where I scored a pre ground chip for $5

spar I don't want this to sound wrong but looking at that pic where you welded everything together there does look to be a couple of speccy welds one vertical and one where it look like the weld either just hit the other part or missed the joint,So a revisit should be on the cards for those welds to ensure they are sound.

Now with the motor have you done a rewind or just using the existing windings as the old adage of just putting magnets in and leaving the windings intact is the output will be closely matched with the name plate current.

what I did with mine was a one wire test and tested the voltage output at 240 RPM which was my desired cutin speed and off memory it was about 30 turns of 3 inhand 1mm wire and as said before trying to drive my motor conversion on my big lathe was when cutin hit even in 5:1 backgear the VFD said noway I can drive this and went into protection mode.

Cheers Bryan

[SparWeb]

Bryan,

My own Dad would part with a bunch of some of his great equipment but it's 3000 km away.  So I collect my own.

I agree that my welding is not particularly good.  Many things I could improve and certain things I can't... no need to apologize.
I can't see up close as well as I used to, so I'm often in the situation where I have the bead started but I can't get my head far enough away to focus my eyes.

While I love my MIG welding machine, I've learned that some jobs should have been done with TIG.  Particularly that bolt head because the heat needed to soak into the disk before the wire started in, but the MIG starts both at the same time.  The low penetration is obvious.  I also learned this month that the last 100 psi in the ArCo tank don't come out the regulator as fast as it should, so I had some really rotten welds before I got a new tank.  The last time I changed that tank was 10 years ago - at my rate a bottle lasts a long time!

BRCM,
The magic of discussion forums like this one...  Some very helpful folks there:
https://canadianhobbymetalworkers.com/whats-new/posts/3312656/

I was given a box full of carbide toolbits for free by a local guy retiring.

Someday I'll bite, and buy one of those cheap little boring-bar sets that you can see on Amazon.

[SparWeb]

Now with the motor have you done a rewind or just using the existing windings as the old adage of just putting magnets in and leaving the windings intact is the output will be closely matched with the name plate current.

It's the stock wiring.  I've never actually done a rewind.
I'm optimistic for the results.  The limiting factor will be the size of the blades, which seem to be a bit too small for this generator.

[ruddycrazy]

SparWeb the only hard part about doing a rewind is the sore fingers for 2 weeks after the job is done ( insert pain emogj) Oztules mentioned to me to hammer some nails on a board to the same pattern of the winding so one can lay each wire in already pre bent and I can say doing 3 inhand for the first rewind was a fun task but I did get there.

I also pre wound the single wire turn for seeing the voltage at cutin and thats how I came to about 30 turns for 24 volts, now when I did insert the magnet rotor into the motor I sat the motor case inside a "G size" warman pump bearing housing that weighs close to 500Kg's. I used a chain block on the lifting jib on my tractor and when the magnets bit the front wheels of the tractor came off the ground so yes if I ever decide to take the magnet rotor out it will be interesting to see if the huge cast iron housing does in fact lift off the ground.

Cheers Bryan

[bigrockcandymountain]

When I put my rotor in, I used a bottle jack up through the middle and let the jack down as the rotor slid in.  I wrapped the outside with 6 mil poly vapour barrier to prevent scratching.  I can't remember what I used for the frame but something tractor related too I think.  I have less magnet material than you guys (1/4") but it went well.  The bottle jack takes all the stretch etc out of the equation and is very controllable. 

[SparWeb]

Maybe I'm losing my nerve to change up to 48V...

As I got out my wrenches to start removing the bus bars from the battery, I started to have some doubts.
I checked on the results I got from my lathe test of this generator.  It would work if I left the battery at 24V.  Not spectacular but just as good as it was with the smaller generator.

The turbine is just about ready to raise today, but that's delayed by all the fussing about with the battery system. And then all the consequences of rendering components like my inverter useless.  Other things like the charge controllers and data logger in need of reconfiguration.

There's a maxim among aircraft designers:  "Fly first, optimize later".  Today I'm thinking about just leaving the batteries as they are, and running this turbine at a sub-optimal, but safe, speed for the winter.  That would also give me time to choose the right inverter, plan the installation properly.  Rather than rush the thing before the snow flies which is what I think I'm setting myself up to do now...

The graph is a bit cluttered, but you can find the blue lines for the 24V battery, using a Series-Delta connection of the generator.  The test isn't the whole story, because a 10-ft propeller will be stalled or nearly-stalled most of the time on this generator.  But a stalled rotor that's still producing power is both safe and self-regulating.

[DamonHD]

Yep, optimise later.

Or my current maxim:

Do it, then do it right, then do it better.

Rgds

Damon

[SparWeb]

Ready to lift. 
Staying with 24V for now leaves me with the ability to log data and run tools from the inverter.
1001 things I don't have to do.
I have all day tomorrow (Sunday) to check everything for the 3rd (or 4th) time before actually hauling it up.

[kitestrings]

Charge!  Oh and, charge  .

[SparWeb]

It's up now.  The lift went well.  I had a moment of panic half-way up, when I thought the battery running the winch went flat (it was a little weak to start).  Then I checked the connections and I had somehow bumped a cable off the terminal post a little.  This is why I keep wrenches in my back pocket during tower lifts. 

[bigrockcandymountain]

Awesome.  I can't wait to see results. That big toshiba has been sitting on the shelf long enough. 

[SparWeb]

Proof!

Today is also a fulfillment of Piggott's Law;  "On the day you raise a wind turbine, there will be no wind until at least the next day."

[kitestrings]

Looks great!

We've had a really nice run of weather here - maybe similar there - and I used it to pull maintenance on ours (Friday & Saturday).  So yes, now you must wait  , but you are well ahead winter conditions and it's always good to be here.  Anxious to hear how it runs.  Best, ~ks

[SparWeb]

Anxious to hear how it runs.
Me too.  Still no wind to start it all day.  But yes, that does mean beautiful weather for enjoying the outdoors.

[SparWeb]

Finally a breeze.

And my current sensor is clearly backwards!

[kitestrings]

Hoorah, it's producing power that's the main thing, and a relief I'm sure, though no one here doubted you 

[DamonHD]

It's YOU making all those winds!

Rgds

Damon

[SparWeb]

Nah, I'm just full of hot air.


Something funny is going on in the datalogger's RPM though.  Haven't figured it out yet but the speeds don't correlate.
It's probably something I did because in preparation for a switch to 48V I replaced a bunch of resistors to condition the input to 5VDC, then when I changed my mind I had to switch them BACK.  A lot of soldering on the same little pads and I could have screwed a number of things up.  Something to check on this weekend.

[bigrockcandymountain]

I'm itching for updates on your new machine.  It has been extremely windy here the last few days, so hopefully you had the same to get some testing done.