Motor-Generator Setup, Generation

Chris, I just re-read a post you made a few months ago about generator mode demonstration:

"Personally, we don't work on or demonstrate generating mode very often. Just when we turn off PWMs to our inverter while the motor is spinning :)"

Made me think: perhaps I should do something about the PWMs? Maybe it explains why the generator stops working when I disconnect it from the DC power supply? For now I am using ControlSuite HVACI Sensorless, almost unmodified, in torque mode).

It may be a software issue? (Sorry, still very new to all this, just used ControlSuite for the fist time this week).

Thanks a lot!

Adela,

What are you doing today with your inverter control?  Yes, you need to continue to control the voltage source inverter if you want to maximize the power transfer from the generator back onto the bus. 

Chris,

So on the generator side, I just run it in torque mode (controlSUITE based), doing nothing else in particular when I disconnect the generator DC output from the bus (so I just let the program run).

Was I supposed to do something specific with the inverter control? I am not sure what you  mean.

Thanks,

Adela

Adela,

Ok, so you are controlling the inverter, that's good.

What exactly is your question then?  Perhaps I need to read your first post in more detail.

Don't understand that why "1.  ... the frequency becomes lower. That should be expected when there is generation."

Due to the remained speed you have, with increase of generator torq, you should have higher freq, isn't it? Because you're moving T vs. w curve towards right.

Yu Zou,

In the watch window I change IqRef from 0 to -0.2, so I am in the quadrant mode where speed is positive but torque is negative, so there should be generation (the slip is negative, hence a lower frequency observed (measured from one of the motor phases) compared with when IqRef = 0).

If this is not making sense, definitely let me know please, I have yet to fully explore all the routes.

Got it, Adela. I believe you are right. The T vs. w curve is essentially moving towards LEFT to obtain more torq.

I just reviewed the curve below again and found the correct operation zone (red circle) we're talking about.

Thanks a lot, Yu Zou.

Technically I should obtain generation, and I have been trying to clearly prove it by disconnecting the generator DC output from the link using a switch and dumping the power generated into a load. However, when I do that, it just kills off the generator and I still can not understand why. 

I am not sure if it is because:

1- there are "safeguards" within the ControlSuite code that prevent the generator from maintaining generation fine when it is disconnected from the DC power,

2- Or if it is the nature of my load (big 25 ohm planar resistor), maybe I should use PWM control to only load 25% of the time or something, as maybe 100% load is what is killing the generator.

Maybe it is my lack of understanding of induction generators that is an issue, I must be doing something wrong here.

Any help would be greatly appreciated, thank you so much,

Adela

Without dc link support on your generator side, your generator is to feed a "stand alone" load and your inverter works as a rectifier.

I'm not sure the capability of drv board to be used as a rectifier. BUT, obviously, in your system, you don't have any control of output dc. In other words, you are trying to operate your generator in torq. mode, but this is based on "inverter fed by a known dc bus", which you dont have at all.

I don't think such system would work...

Also, the "motor-generator system controlled by two drv board" from TI Instaspin, essentially, is a dyno system which emphasizes the control of the motor(NOT generator) under certain load. It seems to me, not for your purpose. 

Yu Zou, Chris,

You are correct, I am using the inverter power module as an rectifier, pretty much.

Below is a diagram of what my setup.

My end goal is to gain knowledge of how to control induction generator and dump power into a load.

The product we are working on developing is a much larger induction generator... so of course if I cannot clearly prove "generation into a load" for those small motors TI is selling (1/4 HP), I am in trouble.

If you believe my approach is not correct, what would be your suggestion to demonstrate generation with load?

Thank you so much, I truly, sincerely appreciate.

Adela

I was going to say the same thing....I don''t think the controlSUITE project will work in four quadrants.

for you generator you can run InstaSPIN-FOC proj_lab05a Torque mode....or if you need to use a sensor you could modify one of the proj_lab12 with InstaSPIN-MOTION to use an encoder feedback.

The InstaSPIN projects use Vbus compensation and can operate in generator mode no problem.

By using single rectifier after your generator, I dont think it gonna work. It seems your project is pretty much a power generation control and if this is the case, then at the generator side:

According to my experience, you need another inverter right after the rectifier dc link. Also, applying control on both rectifier and inverter, the rectifier control should maintain a CONSTANT dc link voltage while the inverter right-after should control the output power to the load. 

Besides, I suggest you firstly to feed strong power grid as the load, before moving to the stand-alone load. This is generally a easier way.

I'm sorry that probably, my solution makes you feel like stumbled. But, there are two different use for the same motor-generator setup, as below, and you are in the 2nd case.

1. Dyno system for motor driving load contorl;

2. Power generation system for generator control;

Yu Zou,

1- Indeed I am trying to implement a power generation system for generator control.

How do you implement the rectifier control so that there is a constant DC link voltage?

 Could you also please give me some hints about the inverter right after to control the output power to the load?

2- I had no idea that ControlSUITE does not support 4-quadrant operation, that means that I need to get back to using Motorware code for the generator.

I had used Motorware code initially but  switched to ControlSUITE for the following reasons:

- Motorware Lab 5a (torque mode) did not have a consistent behavior, it would occasionally change speeds, stall, etc...
 Those changes happened when I changed the prime mover's speed (motor side), or value of IqRef on the generator side.

- I wanted to change the duty cycle of PWM7B so that the resistive load is not pulling 100% all the time. It is apparently faster to implement in ControlSUITE.  PWM7B is going to control a gate driver (IR21834) which controls a power mosfet. The duty cycle would change according to the DC bus voltage. But maybe you have another suggestion?

Thank you, Adela

Chris,

So I will switch back to running lab12 sensored speed mode on the generator side and lab5a in torque mode on the generator side to demonstrate generation.

You said that "The InstaSPIN projects use Vbus compensation and can operate in generator mode no problem." Do you have any other suggestions in terms of hardware or software modifications to my setup so that the generator can indeed have its DC output disconnected from the DC power supply, aside from the suggestions from Yu Zou's? Does what I am trying to do with the controlling the load with PWM7B make sense or should I use a second inverter right after like he suggested?

Thank you,

Adela