Paul and Sabrina's Cheap 3 Phase Inverter (AC Controller) with Field Oriented Control

[freebeard]

For me, all the fun is in the act of creation. It's not in having everyone tell me how stupid something is, and try to change everything so everyone is happy.

Alrighty, then. We're counting on you. I'm encouraged by thingstodo's input, and he got the Office Space reference. Time to watch it again.

[MPaulHolmes]

I'm sorry I wasn't very kind. I actually have what I like to call "rump roast syndrome", but the rest of the world calls it Aspergers (I made that joke up myself). One of my flaws is an almost crippling fear of failure, and fear that I will be discovered to be a fraud. So, miraculously, I got a job where I work from home, and my social issues don't cause too much trouble. Haha.

[e*clipse]

whooo boy - and I look to you as a model of how to handle this stuff!

What a team! Maybe we can work together on this stuff 'cause we're not really together - LOL!

Well, suffice to say you're in good company. And a least coming from me, I have a pretty decent idea how you feel about this stuff. (I'm looking for a *hug* icon - but that might be inappropriate.)

- E*clipse

[MPaulHolmes]

Edit: haha. Thanks e*clipse!

One thing that I know is needed for the ACIM case is the rotor time constant of the motor. Fortunately, it seems to always be in the range of 0.035 to 0.1. Closer to 0.1 for small motors, and toward 0.035 for large ones. At least the Tumanako devel list used 0.035 or something close to that for their large ACIM. I would be interested in hearimg what sort of characteristics they measure, and the values. Worst case, I think we can manually tune the pi loop. Did you know that you can make the AC motor behave just like a series wound dc motor just by setting Id to Iq?? That sets the field current to the same as the torque causing current.

[thingstodo]

Originally Posted by MPaulHolmes

Did you know that you can make the AC motor behave just like a series wound dc motor just by setting Id to Iq?? That sets the field current to the same as the torque causing current

It would be *AWESOME* to have the starting torque of a DC motor available to an AC motor. I guess I have some reading to do .. and to see if the parameters that the industrial controllers measure will correlate to what the PIC libraries want to see.

One thing that I know is needed for the ACIM case is the rotor time constant of the motor. Fortunately, it seems to always be in the range of 0.035 to 0.1. Closer to 0.1 for small motors, and toward 0.035 for large ones. At least the Tumanako devel list used 0.035 or something close to that for their large ACIM. I would be interested in hearimg what sort of characteristics they measure, and the values. Worst case, I think we can manually tune the pi loop.

I'll see if I can remote into work and read some of the parameters. I'll post what I find.

[thingstodo]

Originally Posted by MPaulHolmes

One thing that I know is needed for the ACIM case is the rotor time constant of the motor. Fortunately, it seems to always be in the range of 0.035 to 0.1. Closer to 0.1 for small motors, and toward 0.035 for large ones. At least the Tumanako devel list used 0.035 or something close to that for their large ACIM. I would be interested in hearimg what sort of characteristics they measure, and the values. Worst case, I think we can manually tune the pi loop.

Here's a 575 VAC, 250 HP, 228 amp motor running on a belt conveyor

Motor nameplate info
Rated voltage 575 VAC
Rated Current 228 amps
Rated Horsepower 250 HP
Poles 4
Rated rpm 1775 rpm

P 82 Motor excitation current 57.7 amps
P 90 Motor constant R1 32.1 ohms
P 91 Motor Constant R2 10.73 ohms
P 92 Motor Constant L1 15.22 mH
P 93 Motor Constant L2 15.22 mH
P 94 Motor Constant X 5.63 ohms
P859 Torque Current 220.2 amps

Attached is a few pages from the mitsuibishi controller manual explaining the parameters. I don't see a rotor time constant ...

The file is too large to attach. I put it in dropbox - hope it works for everyone. Let me know one way or the other!

https://dl.dropboxusercontent.com/u/...ual_Tuning.pdf

For the record, this is a Mitsubishi A760 Variable Frequency Drive (VFD or controller).

[MPaulHolmes]

Here's a helpful description of the rotor time constant from Edward Cheeseman:

Rotor time constant is L/R rotor resistance divided rotor inductance, in the simplest form. The units are seconds.

We used 0.04s, and did calculate this from a 11kW industrial motor I had the nameplate details from. (not the motor we used, but close)

Rotor Resistance
Basically, from the efficiency data at rated power, very approximately the rotor resistance is equivalent to 1/3 the total losses. (assume 1/3 stator, 1/3 core losses, bearing friction and windage etc)
Rated current was 19.8A, Full load efficiency 92.5% (a 400V50Hz motor)
11kW/.925 - 11 = 891W total.
Per phase (/3) =297W
In rotor (/3) = 99W.
Normalised current (current if rotor had same number of turns as stator) P=I*I*R
R=99/19.8/19.8=.25ohms

Rotor Inductance
Using above data plus Power factor (full load) .85
Find the complex motor impedance at rated load 230V(per phase)/19.8A(at .85pf)=Zmotor
Zmotor is Rmotor+2*pi()*f(50Hz)*Lmotor
Lmotor = 0.02H,
Basically assume motor inductance is 50/50 rotor/stator,
Lrotor = 0.01H

Rotor Time Constant
0.01/0.25=0.04

We have had success with values 0.04-0.048, depending (we think) on how warm the motor was. At any time/temperature, any number in that range would produce usable torque. Rotor Time Constant ideally needs to be dynamic - the rotor resistance changes with temperature, the inductance with current (saturation), and some IEEE papers also say with slip (although I haven't read up on that yet)
Resistance changes ~25%, Inductance a lot more. Both serve to reduce the constant.

So with a little data you can get a ball park figure!

[Astro]

Does the pre-charge output switch off after the main contactor output switches on? Or does it remain high?
I am thinking of just using a solid state relay for the pre-charge as they are cheap and it will be lowish current and for a short duration.
Something like this Solid State Relay SSR 25DD DC DC 5 250V DC 25A 3 32VDC | eBay

[MPaulHolmes]

I switch it off after like 0.1sec after the main contactor closes. But it's 1 line of code, so I can do whatever you want.

[Astro]

Switching off is good for me.
No point in holding the pre-charge relay on/energised if it is no longer needed.
Thanks