PM Generator Voltage 2

[newenergy]

We are designing a Permanent Magent Generator and will be using an older conventional generator as a starting point. We intend to replace the conventional rotor assembly with a PM rotor. We have determined the magnet type, size, rotor/stator air gap and rotational speed, but are uncertain how to determine the voltage of the electricity that will be generated in the stator windings. Can anyone suggest how the stator voltage can be calculated?

 

[waross]

Give it up. I will not be a commercial success. From your questions, you don't know enough about the basics to understand why your idea will fail. You have not worked with generators and you are wasting your time and our time, but mostly your time.
If you had any idea of the basics or could read a basic textbook you would not have to ask that question. You do not seem to be aware of the dynamics of variable field strength to compensate for the internal voltage drop of the generator which varies with the load.
You will get your generator working and the presence of an output voltage will prove to you that you are successful. When you then try your generator under load you will be back asking why the voltage drops so drastically.
If you have heard about permanent magnet generators in large sizes you have misunderstood what you read and jumped to some erroneous conclusions. Look up some brochures on large generators with the PMG option and find out what a PMG is.
Sorry to rain on your parade but save your money.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rmw]

Other than that Bill, do you think it will work?

rmw

 

[waross]

Hi rmw.
Yes it will work. There was a company building and marketing permanent magnet generators years ago. I haven't seen one for years. The size as I remember was about 2 KVA, not 1000KVA.
The design was a pancake design, very thin and a much greater diameter than similar rated small gen sets. The working winding was very low resistance to minimize voltage drop under load. A diameter two or more times that of similar rated sets resulted in less turns being needed to develope the same voltage as the smaller sets. Again less resistance and with fewer turns, less reactance. Not a principle you can apply to a larger set.
And, I have had occasion to run a couple of sets on a fixed field. One was a 35 KVA set with about 3 KVA to 6 KVA load.
The voltage regulator failed and the owner could not afford a new voltage regulator. I rigged up a small variac and a bridge rectifier to supply the field. At this point, the behavior was basically the same as would be expected from a permanent magnet field. The output was usable, barely, and got the camp through for a couple of months. However, based on the performance, the owner bought a new voltage regulator when he could afford it and had it installed.
Let's see: 6 KVA on a 35 KVA set is about 17% load. That was barely usable and was repaired.
Other times, I have taken readings with a fixed field that showed the same voltage drop under load.
If these folk go ahead and waste their money, and solve the voltage issues, there next post will probably be:
"We converted a generator to a permanent magnet field. It works great but we have one problem. whenever we get over about 15% or 20% load the voltage drops about 15%. What can we do?"
They may be able to supply base excitation with permanent magnets and over wind it to control the excitation and the voltage under load. I doubt that there will be enough space for the extra hardware.
I don't see the point of it all. Take a good generator (even if it needs some repairs and spend a lot of money on an idea when they do not have the basic knowledge to know what the results may be. It doesn't take much energy to excite the field.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[newenergy]

Thanks for your critical insight. The generator that we are using as our starting point is a conventional generator from a 1 MW wind turbine. It is our intention to replace the conventional rotor with a rotor design using permanent magnets in much the same manner as GE, Enercon and Siemans have done (and proven) with the new wind turbin PM generators. All three of these companies successfully demonstrated up to 2.5 MW outputs in these applications and are currently building them.

 

[davidbeach]

PM synchronous generators in wind applications are allowed to run wild (over some range of speed/frequency) and the output fed to a rectifier/inverter combination. That does solve some of the load/voltage problems. But at the cost of the necessary power electronics.

 

[waross]

It's funny how much difference a little bit of missing information will make.
An induction generator on a wind mill has several characteristics that are very different than a conventional generator.
David summed it up well;

PM synchronous generators in wind applications are allowed to run wild

You can get an estimate of the voltage by comparing the strength of the permanent magnets with the strength of the original field. Then do a ratio between synchronous speed and actual speed. This will give a good estimate of the no load voltage. The problem is determining the strength of the original field (at no load).
The (no load) voltage will be proportional to the field strength and the speed of rotation. The original rated voltage will be the base.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[newenergy]

Thanks for the inputs. Sorry about the lack of clarity in my original quesiton.
We have measured the field strength of the original field in a no-load condition and found 230 mT with a 5 mm air gap. The PM rotor that we are intending to use will have 260 mT with a 8 mm air gap. As we can manually adjust the air gap (when the unit is stationary) down to 5 mm we will use the 8 mm gap as a starting point to determine the amount of electricity generated. Adjustment of the air gap will be made until the optimal output is achieved (sweet spot). This will require a bit of trial and error.
The synchronous and actual speeds will be the same as we are using a closed loop system to keep the drive at a constant 15 RPM to give us a 50 Hz output frequency.
Based on waross' recommendation, we should be seeing a stator voltage very similiar to the original generator with a conventional rotor system. That is, approximately 460 volts.
The generated electricity will be frequency stable but, as previously discussed, voltage variable. It will therefore be fed through a voltage regulator (still to be defined) and then through a transformer to ensure the electricity meets spec. Any ideas on a an off-the-shelf voltage regulating system that could handle this application?

 

[waross]

The length of the magnetic circuit is critical. Doubling the length of the circuit will cut the field strength in half. But, an inch of air is equal to about 10,000 inches of iron. You will probably discover this when testing. If you run at a fixed speed, you restrain the ability of the generator to self adjust it's voltage.
If you study induction motors, and look at the regeneration effect when they are running above base speed, you will see how an induction generator works. Same thing. Often interchangeable.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

I ran out of time. More info:
Voltage regulators, these generally work by controlling the input power to the exciter. Amn external voltage regulator to handle your full output will be large and expensive. It will probably push your project over budget in both initial cost and in any energy savings you hoped to gain by going to a PM field.
You haven't said what your configuration is or why you want to go PM. Are you on a grid? Is this stand alone?
As David beach said, there is a lot of electronics involved in the power delivery of a current wind turbine with a PM field.
By the way, in regards to your adjustable air gap, if you can vary the gap on the fly you can use the air gap to control the voltage, and all the air gap doesn't have to be between the rotor and the stator. It may be difficult to get fast enough mechanical action to control this way.
The saturation of the magnetic circuit will most likely limit the open circuit voltage to about 12% to 15% above rated voltage at base speed. BUT, your new rotor is part of the magnetic circuit and may not be equal to the old rotor.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[newenergy]

Thanks for the tips. Been digging into some textbooks.
Currently, this application will work primarily on the grid. Later we intend to use this technology for off-grid applications.
Our company has developed a new technology to drive generators. This proprietary technology has already been sucessfully proven using conventional generators.
Our next logical step has been to simplify and lighten the rotor assembly. We are doing this by replacing the rotor from a conventional 1 MW generator and replacing it with a PM rotor consisisting of individual magnetic assemblies. The mechanism that we have developed to vary the air gap can only be adjusted when the generator is standing still. For this reason we cannot adjust the air gap to regulate the voltage. The drive system however maintains a very accurate rotor speed, which ensures a stable 50 Hz output. Rather than using a rectifier/inverter system that wind turbines use, we will use an external voltage regulator and are looking for a qualified supplier now.

 

[waross]

I suspect that a full current voltage regulator will be large enough and heavy enough and expensive enough to overwhelm any savings in weight or expense that your new design provides.
A suggestion, subject to the comments of others;
How about a hybrid rotor? Can you add booster coils to the PM rotor?
The PM would supply basic excitation and supply full load voltage or close to it at no load. As the load increases and the voltage drops, the booster coils would increase the excitation to compensate for the voltage drop.
If you leave in the brushless exciter, this may be controlled by a conventional Automatic Voltage Regulator.
A basic AVR may be less than $1000, and be built on a small printed circuit card. It will have the advantage of the ability to accept a quadrature input for load balance when operated in parallel with other generators. (If it does not have quadrature capability the feature may be added externally with a properly connected resistor. (A quadrature CT will be needed in either case.)
I expect that an external full current voltage regulator for a 1000 kW generator may be about the same size and weight as three 100 KVA transformers and more expensive.
If you find different please let us know.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[newenergy]

Waross - sorry for the late response. Booster coils are an excellent idea;however the rotor design does not allow this option to be easily installed. Based on the outcome of our tests with this prototype PM generator, we will be able to determine if booster coils are needed. Good point.

We are investigating using a voltage regulator used for 1 MW wind turbines. These are proven units that are rugged and, on the used (but servicdeable market) should fall into our budget.

 

[waross]

Hi newenergy;
Can you post a link to any information on the voltage regulators you are looking at? Thanks in advance.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[newenergy]

We've been reviewing the design specs for the first unit and have decided to incorporate the ability to vary the rotor speed (frequency) for the initial trials. This will enable us to play with the speeds in order to find the point at which the generator operates most effectively. In view of this it would make sense to either use the conventional rectifier/inverter power converter format to facilitate voltage and frequency regulation, rather than strictly a voltage regulator as previously indicated. We are discussing various packages with a number of suppliers at present.

On the other hand, we have taken a look at a four quadrant matrix frequency converter incorporating frequency and voltage regulation (for example the Concycle system offered by Woodward SEG website:

http://www.seg-pq.com/uk/download/assets-pq/pdf-concycle-gb

). This system appears to offer lower losses and greater flexbility over the conventional route. Has anyone had experience with this type of converter and, if so, any inputs?

 

[TurbineGen]

Newenergy, there might be an esier solution to your generator issues, but it will require some serious "out of the box" design.

In a standard PM generator the Volts/hz ratio is fixed by the amount of magnet/stator interaction. If you can change this dynamically, you can effectively change the characteristics of the machine.

How do you do this? If you can dynamically change your airgap with the machine running, you can effectively change your excitation. One way to accomplish this is to move your rotor slightly out of the stator field. This will lower the volts/hz ratio. As you need more voltage, move the stator over the rotor and the voltage will increase.

This is hard to do on a standard inrunner. I'm suggesting an outrunner style (where the PM rotor is outside and the stator rests in the middle). An example of an outrunner is a CD ROM motor. I have much experience with these types. It is much easier to do on an outrunner IMHO.

Your fellow motor/generator builder,

-TurbineGen

------------------------------------------------------------------------
If it is broken, fix it. If it isn't broken, I'll soon fix that.