Another wind 'event' with my 17' turbine, 50+mph unexpected burst of wind encountered

[Dan Lenox]

All,

Over the last month or so I had been waiting for a calm warm day to take down the tail and replace the wood with some lighter FRP material. That day never seem to come.

I had been closely monitoring the forecasted wind and when I saw 18+mph winds that I typically would shut off the turbine. Over the past couple of months I have visually confirmed that the turbine is fully furled in a 28 mph winds. However when wind hits approx 35mph have noticed that the turbine gets a big increase in power output of 4-5kw.

Last Monday evening I awoke at about 1:20am heavy winds were slamming against our house, only 14mph winds were forecasted, so I deciced to check on the turbine.

First looking at the output from my software I was miffed, 0 amps output and 0 rpms... So went into the back room where inverter was, and acknowledged no output from the turbine, went outside and it sounded like a helicopter was landing in our field.

Went back inside and threw the contactor to short out the stator, the turbine came to a halt and I decided to keep it off. Looked back at my logs and had determined that for the last 10 minutes the turbine was free-wheeling!! For the previous 5 minutes the logs showed that the output went from 800watts and peaked at about 4.8kw. At first my thoughts were that the rectifier blew, and went back to bed.

Next morning replaced the rectifier with a backup unit and started up the turbine. While I watched my monitoring software I saw the rpm's steadily climb and then quickly the rpm's dropped off to nothing again, while the ampmeter and software showed that 0 amps were being output. Went outside to again hear the helicopter, shorted out the stator and again it stopped, and went to work.

That evening decided to make sure that all electrical connections were good, opened up the DC distribution box and all appeared to be normal. With short daytime and working long work hours decided to mull it over. I believe that sampled turbine voltage that gets fed into the DataQ A/D controller exceeded it's operating limits and that is why the software had shown 0 rpm.

I also checked the anemometer that measures wind speed at the top of the tower and at 1:15pm it recorded 48mph. This anemometer does some 'averaging' so I suspect 50-55mph winds were present.

At this time I have not put a meter to the system to confirm my suspicions, but at this time believe that I burnt out one phase of the stator, luckily there appears to be two phases left and that is why I am able to stop the turbine by shorting the stator (I still have the mechanical brake as my backup).

I probably will wait till the weekend to give thurough examination as I leave for work in the dark and arrive back home in the dark so can not visually examine the stator that is still at the top of the tower.

In doing a lot of thinking about things I believe that the offset of the turbine from the yaw bearing is insufficient, and that in high winds the turbine seeks back into the windward direction, effectively un-furling itself.

My plan is to (once I confirm my suspicions) is take down the turbine (hopefully some warm/calm day) in the next month or so and increase the offset of the generator by an inch or so, rebuild the stator and reduce the weight of the tail, and possibly increase the air gap between the rotors to help reduce the stall condition a bit.

At the current time I believe that the offset of the turbine is about 8-1/2" center-to-center from the yaw bearing.

One thing that I am very happy about is that mechanically the turbine and blades have survived such an event!

Still learning!
Dan Lenox

[Rob Beckers]

Hi Dan,

Sorry to hear about the turbine trouble! Though I suppose congratulations are in order for a very sturdy turbine; Throwing the stop switch when it is running full speed is often a recipe for a burned-up stator, and not loosing the blades while running unloaded is also a testimony to its strength.

It's easy to measure if a bridge rectifier is blown or working. Just measure with an Ohm meter if the diodes are conducting where they should, and block in the other direction. How do you actually measure output power? If I remember your setup correctly you directly charge batteries, with a diversion controller to avoid overcharging those batteries. Correct? Do you have a shunt in the DC line of the turbine to measure current? Then use battery voltage to determine power?

Turbines shouldn't speed up when they are furled (unless, maybe, when they are running unloaded but yours can't do that if it directly charges batteries). Sounds like something is wrong with the furling mechanism and it's not actually furling as much as it should.

-RoB-

[Dan Lenox]

Rob,

I have what Xantrex refers to as a hybrid inverter, grid tied, battery backup, it's like being off grid except if I produce more power than I consume the extra goes out to the grid. I like to consider the grid my *big* battery backup..

I am using the software that I wrote, it listens to the output directly from the Xantrex XW inverter itself to measure voltage and amps. At the time of the 'incedent' last recorded output was about 4.8kw, and it was for a very short period of time. But I also have an ampmeter directly tied to the turbine output for visual confirmation.

Actually since it directly charges batteries, it 'should' not run unloaded - unless the rectifier blows, I still have to test the 2 units that I have as well as the wiring/components before I know exactly what happened.

The stator may actually be ok, I have yet to take a meter to the system to find out the source of the problem. Since the turbine stops when I short out the stator, it implies that it 'should' be intact. Won't know till a couple of days when I can trouble shoot and visually see the stator using some binoculars during day-time hours.

I think that my generator needs to be offset about 1" more from the yaw bearing as well as reducing weight on the tail by replacing the wood vane with lighter FRP.

I'm very happy with it mechanically, an analogy would be putting your car in park and pressing the gas pedal, maxing out the engine for 10minutes. It would be iffy that the engine did not destruct.

I thiink one thing that may have helped the stator is that recently I moved the contactor another 130' further away from the turbine (about 330' total), that much more transmission line to help disperse the tremendous amount of power when the stator is shorted...

Dan

[Rob Beckers]

If I understand your setup correctly, what you are measuring from the XW is not actually wind turbine output, but how much the sell-back engine is pushing onto the grid. That is related to the wind turbine production, but not in a one-to-one relationship. The sell-back engine is simply voltage triggered; when the battery voltage exceeds XX Volts it starts loading the battery bank and pushing power onto the grid, trying to keep that voltage at the set level.

While you obviously cannot sell back more power than the turbine produces, there's nothing to guarantee that turbine production at a specific instance in time matches sell-back power at that same instance. In fact, that would be quite a coincidence. If you want to measure turbine production (for example to relate it to wind speed, so you can create a power curve), you'll have to measure turbine current (either AC or DC after the rectifier, though DC will be much easier since there's no power-factor and 3-phase stuff to deal with) and use current and voltage to calculate power.

What I mean with this, is that if the inverter is pushing back 4.8 kW that does not mean the turbine is producing 4.8 kW, nor did in fact produce 4.8 kW at any time (maybe it did, maybe it didn't, you have no way of knowing without proper power measurement). It may have produced 2 kW for 2 seconds or so, and the sell-back engine didn't pick that off the batteries until a few seconds later, by going to 4.8 kW for short time.

I do agree with you that measuring sell back energy (power times time) at the XW is the proper way to measure overall energy production of the wind turbine, since it takes all the conversion losses into account. It is not the way to measure turbine power though.

-RoB-

[Paul Bailey]

Hi Dan : Are you exceeding the current capability of the 3 phase bridge to lose 0ne of the legs (diode) of that bridge so soon??. Or was it a stator problen first?? I think you can run 2 in paralell/tandem so if you lost a Leg its business as usual on the bridge and you still get the same DC output.... Paul:notrust:

[Dan Lenox]

Rob,

No I'm measuring much more than sell-watts!

The Xantrex XW *spews* out tons of information 2x per second, everything from grid line voltage, incoming/outgoing watts, power coming from battery, status, etc.

I am typically looking at the battery output that the Xantrex inverter draws from, so I get a very accurate reading of turbine output since that is the only DC source that is pumping energy into the battery bank.

Paul - no I'm using a DF150BA80 3-phase rectifier block that has 150amp capacity.

Anyways curiosity got the best of me, and I have to admit that temporarily I was pretty bummed out...

After getting home tonight I found the culprit - a blown #4 wire that connects between my recitifier and my load resistor. It was a litte hard to see as the resistor is mounted just above eye-level, but checking around with the meter found it!

With this turbine somewhat reguarly producing > 3kw in hind-sight the #4 is undersized. As luck would have it I had already ordered up some 2/0 cable and will be shortly be replacing cables in a couple of days.

While turbine is shut down in a couple of days I will also take the meter around and double check EVERYTHING before turning it back on.

Whew - this might have been a close one.

Luck counts!

Dan

[Rob Beckers]

I am typically looking at the battery output that the Xantrex inverter draws from, so I get a very accurate reading of turbine output since that is the only DC source that is pumping energy into the battery bank.
Dan

Well, that was what I was trying to point out: While the turbine may be the only DC source, that does not mean that inverter input (DC draw) is the same as turbine output. Yes, when measured over a long time period (hours or so) the two should match as far as kWh's are concerned. However, turbine output power in Watt, is not the same as DC power into the inverter in Watt. To verify turbine power you would normally not take battery losses into account, another reason for not measuring power draw from the batteries, but input from the turbine into the batteries.

-RoB-

[Dan Lenox]

Rob,

Yea I hear what you are saying and understand.

I have a digital ampmeter directly connected to the turbine output, and often I have compared this value with what the inverter says is coming out of the battery and most times it is within about .5 amps of each other.

As the turbine is speeding up and increasing the power into the batteries the inverter has a slight time lag behind it, with the inverter reading being slightly lower. Then as the turbine is slowing down the inverter is pulling slightly more power out of the batteries than is being put in - so the end result is pretty close.

This is one reason that I put the ampmeter into place to help validate the numbers that I am seeing. It will be off for the instantaneous reading, but over a short period of time it is very close.

Dan

[Dan Lenox]

All,

Pertaining to the recent 'event' that I had.

In diagnosing the wiring problem that I had, I noticed that no remnants of the wire existed inside the soldered connector - it would appear that the (guessing 100+ amp gust) was so intense that it *melted* the solder.

Once the solder melted the wire exploded out of it's connector!

Now I understand why the recommendation is to CRIMP connectors.

This weekend I will be replacing as much as I can with 2/0 stranded cable with crimped connectors.

Hope others learn from this lesson - don't underestimate the power of electricity!

Dan

[Joe Blake]

Once the solder melted the wire exploded out of it's connector!

Now I understand why the recommendation is to CRIMP connectors.



But on the other hand, couldn't it be argued that the solder had acted as a fuse and possibly saved greater damage from occurring further down the line, so to speak?

Joe

[Dan Lenox]

Joe,

Maybe - but it was probably the weakest link as well. And fuses inline with a turbine are bad news...

Dan

[Brian McGowan]

Dan,
Solder is not good for making connections like that. Good physical contact is best. Crimp is good but for wire that size you will need hydraulic crimping equipment. I have actually done this and it is no fun.
I started using these for my heavy connections.
http://www.homedepot.com/webapp/wcs/stores/servlet/ProductDisplay?storeId=10051&langId=-1&catalogId=10053&productId=100183915&N=10000003+90068+501516

At least there is good secure physical contact and you don't need to get out the giant crimper and if you change your mind about how to run the wire or move stuff around, it is easier to change than putting a new crimp connector.

Wow you blew up a #4 wire? Got any pictures? Hope it's in a fire proof location. You are getting good power!

Brian

[Dan Lenox]

Brian,

At last recording from software turbine was putting out 4.8kw of power, as I log an entry every 5 seconds it could have gone over 5kw, and once before I did see 5.5kw! At the time of the 'event' I'd bet that it was putting out over 100amps @ 54volts.

I'm not sure that I blew up the wire per-se, what I suspect is that

- the solder melted
- steam developed in the connector
- the steam assisted in reducing contact area
- reduced contact area between wire/connector blew up
- pushing the wire out from connector

Dan

[Brian McGowan]

Still WOW. I have a hard time figuring where steam would come from as solder melts at higher than the boiling point of water. I consider it possible that once the solder melted and flowed away it just arc welded the end of the wire until the gap was too big to jump or it just vaporized the solder or if the wire was tensioned away from the connector it would have just pulled away from the connector after the solder melted. 100A@54V is arcwelder territory. Either way, it might have looked pretty spectacular while it was happening. #4 is rated at 60A for power transmission and 135A for chassis wiring so changing to #2/0 is a good idea and should cover it for all conditions generated by that turbine.

Brian

[Dan Lenox]

Brian,

Quite possibly there was a small air gap inside of the connector, with the melting solder heating it up, it would expand the air, resulting in pushing the solder out from the inside of the connector.

Once the solder melted I would think that the actual contact area reduced probably by 1/2 allowing the amperage overload to take place, and making it blow. So it wasn't really like blowing up an intact #4 wire, I doubt that it had the amperage to do that, but it would be much easier with the smaller contact area.

Dan

[Dan Lenox]

Today I replaced all of the #4 wiring throughout the DC distribution box with #2/0 stranded cable!

I do not want a repeat of the incident that occurred this week, and with the beefed up cable - it should not happen again.

All of the cable connections were crimped using a small A-frame hydraulic press. I made up a jig that the connectors/wire fitted into so that when I crimped the connector it did not deform too much. I also change the wiring slightly having the battery bank connect directly to the main 250amp breaker, at the other end of the breaker was the inverter and + bus bar connection.

In addition, the wiring from the DC distribution box to the inverter and *all* the battery connections throughout are #2/0!

No more 'melt-downs' for me!

BTW: started up the turbine, it worked like a champ with no visual signs of what 'could have happened' with having it free-wheeling in a 20mph wind for 10 minutes. I have to say that I am *VERY* happy with how the turbine held together in this (hopefully) worst-case scenario - mechanically and electrically!

Dan

[Joseph OLall]

I'm curious...what are the symptoms of a "burnt out" PMG that has been in high winds?
:suspicious:

[Dan Lenox]

the most obvious clue is a 'runaway wind turbine'.

ultimately something will break in this condition and stop the turbine from turning.

dan