The simplest way to get this pair of devices working in a non-400hz 3-phase environment seems to be to program a pic to generate three-phase and amplify the output. If I have three pins per phase and run each through a resistor of some calculated value I can make a psuedo-sinewave. Is this likely to have any issues because it's not a true sinewave? This is a hobby-level application, so good enough will be good enough....
Hello Rene,
Some uC don't have enough timer links for three-phase. I think most of the MSP430 can do that and then PWM clearly looks like the solution here. No idea about PICs but it would be worth to scour the app notes for any three-phase tasks.
The issue with not-so-perfect sine waves is often audible noise and sometimes faster wear due to resonances that travel to the bearings.
Regards, Joerg
What speaks against PWM ? It is simpler to amplify to any level rather lossless.
Rene
-- Ing.Buero R.Tschaggelar - http://www.ibrtses.com & commercial newsgroups - http://www.talkto.net
Just look at some of the PICs designed for motor control. The have three phases for PWM.
Your post seems to be a continuation of another thread. As I understand your problem, you need to provide drive for a synchro or selsyn for some kind of motion control application. There is a common misconseption that a selsyn/synchro is a 3-phase device. For the usual motion control application, this is not true. Time-wise, the voltages are either in-phase or 180 degrees out of phase with respect to a reference. As the synchro shaft rotates at a constant rotational velocity, the AMPLITUDES (forgive my shouting) vary sinusoidaly and are
120 degrees apart with respect to the MECHANICAL (I did it again) position. To generate these voltages, you only need to generate a single phase (electrically) sine (or pseudo-sine) wave of an appropriate frequency, i.e., 400Hz. You then vary the amplitudes of the other two voltages so that they vary sinusoidally 120 degrees apart with respect to the desired mechanical position of the shaft. ~ There is one application that does require drive voltages that are 120 degrees apart electrically. If you want to generate a sine wave whose electrical phase varies sinusoidally with respect to a reference, you can drive the 3 stator windings with sine waves that are electrically 120 degrees apart. Now, as the shaft is rotated, the rotor voltage electrical phase will vary linearly from 0-360 degrees as the shaft position varies from 0-360 degrees. ~ I hope this helps. Regards, Jon
Yes, that's right. And one more thing - I didn't see the OP either - as regards the quality of the sine, fuhgeddabowdit, I've seen some truly hideous signals driving aircraft synchros just fine.
Bob
Um, yes, but...
First off, it looks like you know what yer talkin' about.
(I don't.)
I have a synchro sender for a fuel tank and an indicator that matches it. I wish to provide them with the electrical necessaries to get the indicator to show what the sender sees.
They have each three wires (but a clever fellow like you would know that).
What do I need to make this happen? I thought it was three phases at 400 Hz, but...
Oh good.
Would square be too too hideous?
Ah....Been 40 years or so. A selsyn / syncro sender/receiver has 5 leads: 2 for the AC drive (nominally 120VAC 400Hz), and the three from one goes to the 3 of the other. Technically, they are transformers, and the movement of the shaft of one "picks off" various phases of the exciting field; those voltages drive the other and set up the same field. The rotor of the second then moves to minimum energy (via magnetic attraction and repulsion), which is the same position as the first. If connected improperly, the "receiver" will turn in the opposite direction. There are a few rare 60Hz versions; look at the label / plate for info; small ones are always 400Hz. They can be driven with 60Hz, but at vastly reduced voltage (being
400Hz transformers, they have a much lower inductance at 60Hz). Naturally, the torque will also be a lot lower.
A square wave will not do too well, and one could zap the insulation if one uses a 120V 400Hz square wave to drive the 2-lead drive winding.
(snip)
(now the problem/question becomes MUCH clearer ...)
UIAVMM (and it has happened) you are looking at these devices the wrong way.
A Selsyn has a "single phase" rotor and a "three phase" stator. The intended application for remote (shaft) position indication used excitation of the rotor to induce voltages in each stator leg which represented the vector components seen at the three 120deg displaced positions.
The stator windings of one Selsyn were connected to those of the second unit, and when the rotor of the sender was excited the receiving unit would adopt the same rotational position if its rotor was excited in phase with the sender.
So your requirement is to excite both rotors with a common SINGLE PHASE alternating drive. Ideally a sinusoid, but square waves work as well. Do note though that you are driving an inductive load.
As always, I am ready to be corrected, as I haven't played with them for 30 years.
This particular pair of units want 12v (so the labels say).
But square is too bad, oh well...
If i remember correctly, (been about 40 years), 120V units rated for
400Hz run nicely on 24V 60Hz. Never heard of 12V units, rather hard to believe; typically selsyns / synchros are run from the line and are made for military applications - hence 120VAC 600Hz.
IIRC 12 and 26 volt rated synchros were fairly common a couple decades ago when I was using them. I see no reason why a square wave would not work for a simple indicator application, perhaps with the voltage slightly reduced for the same RMS voltage or with the sharp edges removed with a low pass filter. Square wave excitation of 12 volt reslovers is quite common and works well.
It wouldn't be square for long driving into a nearly pure inductive load :(
Again, I didn't see your original question so I may be talking through my hat, but one technique I have used in the past is to run the AC reference voltage into a four quadrant multiplying DAC and modulate it with a cosine wave lookup table in stored in ROM. This way with one stored wave you can generate the phase shifted 'legs' by simply indexing into the table.
One other thing. Most synchros will work just fine in an 'S2 grounded' configuration where you ground one of the legs -S2- and only need to drive the other two.
Bob
...
...
OK, should we all vote on whether it's AMPLITUDES or PHASES? >:->
Cheers! Rich
-- #!/usr/bin/bash echo \'Cheers!\' ; echo \'Rich\' echo \' ------\' #echo `fortune all` OFORTUNES=`ls /usr/share/games/fortunes/*-o` echo \\"`fortune $OFORTUNES`\\"
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.