Operating temperature of wind generators

[Mel Tyree]

Hi all,
I have started a new thread because I need ideas on where best to mount a temperature sensor to meaure the temperature of stator coils while a small turbine is generating power. Specifically I want to measure temperature of a 10 kW Bergey Excel-s, but information about the physical configuration of any small turbine might help at this stage, but specific information about a Bergey is desired if anyone has had a close look at one.
I have a 2-channel radio transmitting data logger connected two LM335 temperature sensors (in a T0-92 package about 3 x 4 mm) on a smaller diameter cable. I want to insert this into the stator coil assembly. Is there likely to be enough room between the coils and the moving magnets to accomodate 3 or 4 mm clearance? I have the parts breakout and list for a BWC Excel but no dimensions are given. Does anyone know if the stator coils are pressed right up against the rear bearing ring (parts 44 & 43 in the attached file)? Perhaps I could just epoxy the senor to the surface of the bearing ring??? Does the bearing ring move?
WHY am I doing this?
Well I have learned that most generators can work okay at a maximum temperature of up to 80 or to 120 deg C depending on the coil insulation and magnet quality. Both insulation and magnets can be degraded if temperature gets too high. I want to know how near the 80 deg C limit the Excel-s is and if there is room to drive it harder, i.e., draw out more power.
The Excel-S manual is 'pending' on the Bergey site so I have extracted pages from the Excel-R which should be the same machine but with external components for battery chanrging. However the file is still too big. So if you think you can help please download the file from http://www.bergey.com/Technical/Excel.R.OM.2004.pdf

--Mel

[Brian McGowan]

Mel,
Let me first state what I think I am looking at.
I believe the rear bearing assembly (part 43) is also the front of the nacell. The stator assembly (part 44) is bolted to the rear bearing assembly so it sticks out in front and looks like it's wound toroidally? The magnet can (part 59) has magnets mounted to the inside of it and fits over and rotates around the stator to apply radial flux.
I will assume your temperature sensor is the TO-92 package and thus no bigger than 4.7mm wide by 3.68mm thick by drawing doesn't specify how tall.

If the above is not correct then disregard the following.

I don't think there should be enough gap to put the sensor between the stator and the magnets. I wouldn't even try to put the sensor wire there. In the drawing it looks like the stator coils might stick out beyond the "spool" they are mounted to but I can't tell if it's enough to mount the sensor to the rear face of the spool under where the coils over hang in direct contact with the bottoms of the coils also. There is probably some tiny gap between the coils and the rear bearing plate to sneak your sensor wire through even if you have to route it through the gap where the radii of 2 adjoining coils meet.

If not there try the same place in the front of the spool and route the wire through the spool and out the back as before.

If not that then the last best thing is mount to the inside of the spool under the coils and route the wire as before.

As a last resort I would personally get some extra LM-335s and sand the package down to a size that will fit somewhere you want it to. The silicone is very small and you could probably remove a fair amount of the case thickness without getting to it and possibly .5mm of width on each side without upsetting the leads too much for a total of 1mm.
That's all I have for now. Hope it helps.
Brian

[Mel Tyree]

Brian,
I talked to a guy who installs Bergey Excel-s turbines. He says the rear bearing ring (#43) rotates with the magnet can (#59). They both rotate on the fixed shaft shown in the diagram (#38) with the aid of the front and rear bearings (#49 and #42). Hence the only fixed part is the shaft thru which the 3-phase wires are routed. So the only way to get a wire in there is thru that shaft assuming there is enough space left between the 3 power wires. The generator would have to be taken apart to do what I want. And the low voltage signal lines would be parallel with noisey power lines.

It seems my only alternative is to use an IR temperature sensor I have that can measure the surface temperature of the moving rear bearing ring. It should get hotest near the outer edges. I have a couple of these IR sensors in my lab. They work fairly well but don't know how they will hold up to the harsh environment inside the turbine. See:
http://www.omega.com/ppt/pptsc.asp?ref=OS136
if you are curious about the sensor.

Thanks for thinking about my problem!
--Mel

[Brian McGowan]

Mel,
On closer inspection of the drawing I can see that now. Good that you have someone with hands on this unit. Dissasembly will be required no matter what you choose. If you had put the LM-335 under the coils as I said it would also be exposed to that noise. The noise should be a fairly low frequency sine wave not something nasty like static. I'm not sure how much of a problem it would be. Shielded wire should help and if need be, caps can be added between the shield and both conductors at the transmitter end to reduce this even further. There may already be caps in the transmitter for this purpose. I would have put them there. I still think if you could fish a small balanced audio wire (twisted pair and a foil shield with drain or braided shield) through that shaft you would be able to mount the sensor to the inside/bottom of that spool and I don't think the noise would be a problem. The transmitter could maybe then be mounted somewhere inside the nacell.
I don't think the IR sensor will get a good read on the back of that bearing plate or more correctly, I don't think the temperature of the bearing plate will be an accurate representation of the actual temperature of the coil. That sensor is 3.5" long and I don't know how big that spool is or I would also suggest mounting that inside the spool pointing at the underside of the coil. I have had no problems with omega sensors. This one seems to be built for harsh environments so I wouldn't worry about that.
Glad to think about your problem. Sorry I coudn't be more help.
Brian

[Steven Fahey]

I've tried this (though not with the equipment you are using).

http://www.otherpower.com/images/scimages/4378/Temperature_Sensor.jpg

A relatively simple automotive sensor from princess auto. Up in the air, the needle never once raised off the peg. The motor-conversion in question would generate up to 15 amps through windings which were rated for 10 amps on the data plate.

I drilled a hole directly through the side of the motor, and tapped it 1/4 NPT to match the sensor plug. There is electronic thermal bonding grease where its end contacts the stator laminations.

Placement of the sensor may have affected the results. A block heater directly on the end of the sensor made it work when I was testing it on the bench. I should have tested the sensor in-situ by running DC current through the wire to heat up the whole stator.