3 phase syncrho's

[dr pepper]

This might not be a good place to post this but here goes.
On my quest for driving aircraft instrumements directly I found that on some it isnt practical to remove the synchro motor (and replacing it with a stepper or whatever), so driving one would be the only other solution.

Generating 3 phase controlled 120v sine waves for the stator and a phase related 120v sine wave for the rotor isnt that easy with a micro.
So I had an idea, run a synchro backwards, generate 3 fixed 120 degree phases, use a power op amp or audio amp to drive a standard mains transformer to generate the 3 phase 12ov and stick this on the stator windings, then apply just one modulated sine wave (generated in similar fashion) to the rotor, shifting the phase from 0 - 360 degrees with relation to phase 1 of the stator ought to give me the 0 to 360 degree rotation of the synchro shaft.
Generating 3 phases at 400hz with a dac is no hard task, and generating a fourth phase shifted output according to the required display variable shouldnt be that difficult.

Anyone with knowledge of this sort of thing?

 

[ronsimpson]

I need some background.
You have aircraft instruments that live on 3 phase 400hz. There are motors (inside the instrument) that use this 3/4 phase power.
And you don't have the 3 phase power now.
Is this right?

 

[kubeek]

It is a little more complicated than that. See https://en.wikipedia.org/wiki/Synchro for reference.

 

[Reloadron]

Most aircraft systems including instrumentation rely on 400Hz power for obvious reasons. That is pretty much a given.

Position indicating instrumentation relies on a synchro transmitter and receiver. For example a motor drives flaps and a synchro transmitter sends the flap position information to a synchro receiver driving a indicator of flap position. This WIKI link explains it pretty well.

Now if you just want to generate 3 phase 400 Hz 120 Volt power you can rotate a synchro. However, there are a few things you need to overcome. First, if I apply 120 VAC 60 Hz to R1 and R2 (Rotor 1&2 of my synchro) the output on S1, S2 and S3 will be a frequency dependent on the rotor rotational speed modulated with 60Hz. If you want 400 Hz you need a rotational speed on the rotor that will give you 400Hz and then you need to demodulate the 60 Hz to get your 400 Hz. The direction of shaft rotation matters not. I worked with some very old systems that used this method for generating very low frequency AC. The synchro was rotated using a geared variable speed Bodine motor. Now if you want 400 Hz power you are looking at maybe 400 x 60 = 24,000 RPM of shaft speed so good luck on that.

You don't mention your power requirement?

Ron

 

[kubeek]

Erm, I think dr pepper is trying to drive the synchro indicator from a microcontroller, not generate power or anything like that.

 

[Diver300]

Can you just put DC on the rotor and stator? To move the instrument you would have to alter the direction of the stator field by varying the mix of DC currents in the stator windings. The original synchros used 400 Hz because the transmitter would not have generated anything if they stopped moving, but you don't need a transmitter.

 

[kubeek]

Good idea Diver, just take care not to overload the coils.

 

[dr pepper]

Tried dc, wont work, the armature relies on a changing magnetic field, the whole thing works like a transformer.
Rotating a synchro sort of generates 3 phase power, but not as implied, it would actually be a envelope modulated at 400hz.
These things do not lend themselves well to being controlled by a micrcontroller.

The supply to a synchro is single phase 400hz, the output from a 'encoder' to the 'servo' is 3 phase, I was proposing to run one the other way generate constant 3 phase and apply that apply to the output, and control the angle of the synchro by shifting the phase of what would have been the power input, that way i only need to concern myself with modulating 1 waveform, instead of 3.

The flight surfaces in plenty of aircraft is controlled in a similar manner, either 2 phase or 3 phase motors, sometimes more than one synchronized, or in the case of ailerons inverse synchronized.

 

[Diver300]

Are you trying to get an instrument to move in response to an electrical signal only, or are you using the encoder part as well? The encoder part won't work on DC, because it won't generate any power unless it is moving. However the driven end, which I understand is the instrument, should respond to DC.

That is how stepper motors work. They usually have a permanent magnet in the core, and putting DC on a winding will make them move to align the core with that winding. One step is where that winding is turned off as the other winding is turned on.

If you have an instrument that works something like this:-

then if you energise the rotor with DC, and energise the top terminal with DC, the rotor will turn to vertical and stay there. To hold the position in the diagram, the left winding needs to be energised in one direction and the right winding needs to be energised in the other direction.

Any angle can be achieved if the three windings currents can be separately adjusted over a range of positive and negative currents.

 

[dr pepper]

Yes that makes sense and is why I tried dc.
2 of the reasons I gave up on that was the fact that the rotor coil was pulling 200ma at 12v, I'dve thought it'd only pull a ma or so at its 120v rating telling me the core is probably saturating, and at that the core didnt respond to stator energization, that might have been beacuse I had 12v on the stator instead of 120v, assuming the ratio core to stator is 1:1.

I have some other instruments that have just 3 wires, I think they may be perm magnet synchro's, they are decca navigation indicators so probably work on 2 or 3 slightly shifted phases as that is pretty much how decca worked.

 

[KeepItSimpleStupid]

Depending on exactly what your trying to accomplish, there are these: http://www.wholesalesteppermotors.com/ but it's like 180 deg .out of phase from what your asking.

 

[Diver300]

Yes that makes sense and is why I tried dc.
2 of the reasons I gave up on that was the fact that the rotor coil was pulling 200ma at 12v, I'dve thought it'd only pull a ma or so at its 120v rating telling me the core is probably saturating, and at that the core didnt respond to stator energization, that might have been beacuse I had 12v on the stator instead of 120v, assuming the ratio core to stator is 1:1.

I would have gone for the same current with DC as the unit would take at its rated voltage in AC. That should lead to the same heating and the same magnetic field as with AC

 

[Reloadron]

Well as you know DC won't work. The coil impedance needs considered. If we open a brush type synchro and observe the brushes they are fine wire, unlike traditional brushes running on slip rings. The coil will heat up and the current will well exceed what the synchro is rated for at 60Hz or 400Hz. Think about what happens to the coil impedance of R1 to R2. This is what you saw with the high current applying DC to the rotor terminals.

If you want to try an experiment hack the drive motor out of an old CD/DVD drive. There is a tray motor, a head positioning motor and the drive motor. Look at the drive motor (3 phase) and spin the motor while looking at the three terminals on a scope. The motor will also likely have terminals for the position sensors (HALL EFFECT SENSORS), ignore them.

Ron

 

[alec_t]

Could you get away with using three rectangular 400Hz waveforms 120 degrees apart? Easier to generate than sine waves.

 

[Diver300]

If you are not wanting the complication of a sine wave, you could try using what is called a "modified sine wave" that a lot of mains inverters used. It is like a square wave, but with "off" periods between driving positive and driving negative.

 

[dr pepper]

I have allready converted some aircraft instruments with stepper motors, vehicle instrument steppers, I posted that on another link, this time I'd like to drive synchro's direct.

Yes I'd considered square wave, and what cheap inverters call modded sine wave, I assumed that would cause issues with accuracy and jitters, but maybe not at 400hz, that would have to be tried, having said that I'd need 4 high voltage bridge drivers and an inverter, 4 audio amps and mains transformers would give me 4 clean sine waves, that is however a lot of stuff for a little synchro.

Ron yes 200ma at 12vwas as far as I wanted to go with the little brush wires to the rotor, if I went to 120v then I'm sure I'd just see a little blue flash, the inductance and max flux density the core will stand denotes how long a voltage can be present, 50 hz would probably fry them too.

Something I'd considered too was putting a neo magnet in place of the rotor coil, turning magnets is a messy job though.

 

[kubeek]

Putting 2.4W through a coil seems a little excessive.
Note that the impedance at 400hz will likely be much much higher than the DC resistance, so I wouldnt throw the 120Vac 400hz operation out of the window.

 

[dr pepper]

I agree, 200ma at 12v was of course dc, the coil is meant for 120 ac.

 

[dr pepper]

OK I'm reviving an old thread.
I sussed it, here goes:

And a 6 phase one:

The 110 ac synchro is operating on 12v, which is 24v after the bridge.
3 phases is being generated by the 'duino, or 6 for the differential 6 phase one.
I was going to calculate the phase positions, but ended up using a lookup table, mucg faster, I can make the synchro whizz around.
This one works like an old car speedo using eddy currents, the synchro drives at magnet at a few hundred rpm's:

 

[Les Jones]

I like the way you have done this by keeping the three phase constant and changing the phase relationship of the single phase. When I started reading the first post in this thread I was thinking about modulating the amplitude of the three phases. Your method is much better. You should put the design in the articles section of the forum.

Les.