Need info on phased motor instruments.

[Guest LeoL]

I'm currently working on my output needs (displays, gauges, etc.) and I started looking into various options. I got lucky and managed to get my hands on a few aircraft gauges.Some of these seem to be controlled by phase motors. Does anyone have specific electronic info on how to drive these instruments from an analog or digital output interface?I opened up the radio compass and it has 4 wires. The label inside says its a "Bendix". The rotor is 1 phase @ 22V and the stator is 3 phase @ 9.1V. There's another number "100" with a sine wave which I believe it to mean 100Hz.I've never worked with phased motors and I have only a very basic idea on how they work in theory through magnetic vectors, but my knowledge ends there.Any ideas? If I can't figure this out, I might just try to fit in a stepper motor, but this might not be so easy.-Leo

[Guest MikePowell]

Good morning Leo,Sounds like what you've got is a selsyn based remote reporting instrument. (Except that I would expect five wires rather than four. Two for the rotor and three for the stator.) "Selsyn" is actually a trade name and there are several manufacturers with different names, but selsyn is now the name most frequently used.A selsyn is actualy a rotary transformer with pretentions of becoming a motor. It is very useful as a sensor of angular position. When used as a motor it has very little torque and so is useful in this mode only for very small mechanical loads like instrument pointers. In these simple applications selsyns are often operated in pairs. One is used as a transmitter, sensing the angular position of of something of interest like the float position in a fuel cell. The other is electrically connected to the first, and its shaft turns to follow the first. This receiver selsyn is generally in the cockpit making a pointer move.The wiring is straightforward. Each rotor receives AC power. The stator phases of the paired selsyns are simply connected Phase A to Phase A', B to B', C to C'.When the instrument is a remote reporting ADI with a large mechanical load, the selsyn must be paired with a motor to provide the necessary torque. The selsyn is used to report the actual ADI sphere position. An error amplifier compares actual with commanded positions and supplies motor power to position the sphere.There are a couple of links to more information on my site /instruments/use real ones (or something like that) down at the bottom.Driving these instruments is generally regarded as a complex project. Personally, I don't see why. Of course, I've never tried, so may shortly be looking pretty foolish. Still, I'd be willing to help if you decide to pursue it.Mikewww.mikesflightdeck.comInfo for simpit builders.

[Guest TOMCAT]

Hello Leo and Mike,personally I wouldn't take the rocky road using the selsyns. I don't want to disencourage you but you should consider some facts:One big problem looming ahead is the merge between an analog world, driven by AC and angular information hidden in voltages that represent the sine and cosine values of the rotational angle and a digital world which resolves any kind of information into a bit-stepped DC voltage level.Interfacing these worlds apart means a heck of work - and unfortunately a lot of money as well. For example let's transform an A/D-converter's DC-output into a selsyn signal:First you have to figure out the appropriate angle values for a sine and cosine signal - OK it's just the pocket calculator. But then you'll have to design some kind of circuit which can accomplish converting the DC-signals into two AC-signals. Aww, and I forgot to mention that these ACs have to be almost perfect sine waves.Even if you manage to do so this signal is still not usable for a selsyn. You need some kind of special transformer - it's called a Scott transfomer - to convert that two ACs into a signal you can actually feed into a selsyn driven instrument. And of course these Scott transformers are mainly used in avionics and thus bear an extremely ugly price tag. Yikes!And I really know what I'm talking about: At the moment I'm finishing my training as an avionics technician - and whenever one of these very costly "Digital-to-Synthetic-Synchro" or "Synchro-to-Digital" boxes with a malfunction comes along it's returned to the manufacturer. Even the function schematics look horrible and I definitely wouldn't be very keen on building one these on my own.I think replacing the synchro by a digital stepper motor is far more sensible. And there are also some solutions which use conventional cheap R/C-modeling servos for driving the instruments.Just my 2 (Euro)ctsRegards Thomas

[Guest LeoL]

Thanks for the short lesson Mike. You cleared up some misconceptions I had about how it all works.You're correct, the stator does in fact have 3 wires but it shares one with the rotor. What comes out of the housing is only 4 wires.So if I understand you correctly, by applying the appropriate signal to the rotor I will be able to move the needle. The stator is simply used as a feedback signal for position sensing? Or is it the other way around?As a first step I simply want to get the needle moving and then worry about position sensing...that is if the two can be operated separately, which I hope, or this can get very complicated very quickly.I'll have to read and reread those documents your page links to and see if I can decipher it all.Thomas: thanks for the word of caution. I hear ya!!! It certainly looks daunting at the moment, but until I fully understand the mechanics (electronics) of how it all works, I'm not ready to give up on it just yet.Rigging steppers would certainly simplify my life, but as Mike succinctly put it on his web site, I'm not a "watch maker" and don't intend to become one. ;-)I'm much more agile at electronics than with mechanical stuff. That's the reason I'd rather use the instruments as is and deal with the electronic headaches rather than go crazy fitting a square peg into a round hole wrt stepper motor and the gauge.Yeah, I bet you'll be telling me "I told you so", not too long from now. I'm still on the fence right now, but thats all right. I don

[Guest MikePowell]

Hi Thomas,Thanks for your comments. Almost certainly it would be a real challenge! I think there are several viable approaches. Certainly, one is simply to brute force emulate the terminal voltage and current conditions needed for a given rotor position. Another is to operate the synchro/selsyn with 3 phase on the stator and single but variable phase on the rotor. This arrangement is simple to implement with PWM and a cheap micro processor, but I don't know what impact it would have on torque and stator heating. This would not in any way be suitable for flight rated operation of the instrument, but should be okay for a simulator.Mikewww.mikesflightdeck.com

[Guest MikePowell]

Another reference, Leo:http://www.naii.com/pub_files/Chapter4_edited4BAe.pdfMikewww.mikesflightdeck.com

[Guest LeoL]

Thanks allot Mike. I'm beginning to see the light...Started doing some testing. Quick test reveals that by applying a voltage separately to each of the three stator wires moves the needle to different positions. Good good good...I was worried the windings were fused.The funny thing with this radio compass autosyn is the fact that the rotor shares one wire with the stator when tested with an ohm meter. Odd...Next up, programming a PIC to generate a fixed rate 3 phase output and wiring this up to 3 8-bit DACs (it will have to do for now). I'll finally put those bloody things (DACs) to use...I knew they'd come in handy someday. ;) This should be enough to make the needle spin at a constant rate...or like a raver on brain boosters, who knows!I'll worry about the rotor position sensor as a next step in addition to adding Scott T's to the DAC outputs.I'll look through this latest doc tonite. Thanks Mike.-Leo

[Guest MikePowell]

I find the common connection a bit curious too. Mikewww.mikesflightdeck.com

[Guest TOMCAT]

Hi Leo and Mike,interesting - it's the first time I hear of someone trying to emulate the synchro stuff in a home cockpit. Leo, you like electronic challenges, don't you? :-lolBut when I take a look at your site it might be a more sensible way for an electronics buff like you rather than "making watches". ;) I'm really curious how that project finally works out...That 4 wire thingie on the instrument also seems a little bit odd to me, I haven't seen a synchro of that kind before too. Well, one never stops learning. ;)Mike, thanks a lot for the links, pretty useful stuff. I have added most of them to my avionics web favourites.Regards Thomas

[Guest LeoL]

> you like electronic challenges, don't you?Ha! Yeah, I'm part of the "Electronics Fight Club". We stand around late at nite throwing components & ICs at each other...just call me crazy! Uh-Oh just broke the first rule of Fight Club! ;-)Anyway - my take on the 4 wires is that since this autosyn runs on a varying DC voltage as opposed to a true AC voltage (+ & -) you can "legally" connect the primary (stator) and secondary (rotor) together in a common ground configuration. I've seen this done for car ignition coils. The primary and secondary share one common connection since the voltage applied is a varying DC voltage on the primary coil. Well, that's my take on it.So this definitely makes this instrument unique and thankfully makes it easier for me since I won't have to worry about generating a +/- voltage source.BTW - I've been misquoting Mike...turns out it's "clock maker" and not "watch maker".-Leo

[Guest MikePowell]

I find myself in clock maker mode these days. Maybe watch maker too! Those gears are small and my fingers rather large. I've assembled the mechanical section of a simulated sensitive altimeter. (Except for the Kollsman window; I'll probably use a rotory encoder and 4 digit 7-segment display for that) I'm now working on the mechanical section of a turbine RPM gauge that has a small units pointer in the lower right quadrant to supplement the main, central pointer. I'll drive this with a surplus stepping motor. It will actually be a servo system with a position feedback pot. The relatively high torque at stall makes a stepping motor a viable choice for do it yourself servos. Fewer gears are required for a given load.Thomas, many of the references on my website are to rather old books. Could you recommend more up to date ones that cover aircraft instrument function?And speaking of books, if either of you are interested in analog communication electronics, I highly recommend Experimental Methods in RF Design by Hayward, Campbell and Larkin. It takes a very hands-on approach to investigating design options.Mikewww.mikesflightdeck.comInfo for simpit builders.

[Guest LeoL]

I have a few instruments (RPM gauge as well) that are not autosyn type and will likely be posting a "Need info on servos & steppers" some time soon, so be ready with the info Mike. There's likely no escaping it...I'm doomed to becoming a clock maker (no disrespect to any actual clock makers out there) ;-)Thanks for the book referral. I will certainly look into it at my local bookstore and/or online. Do you the ISBN number perhaps? Unfortunately, my book budget is currently reserved for several AIAA aircraft/aeronautics engineering books, but some of these can wait.BTW - that last link was very helpful. I now understand the use of the Scott T's. Seems they're used for generating the 3rd phase signal from sum of the sin & cosine signals and to filter the resulting steps on the analog output from the DACs.Since I can readily generate the third phase signal with a properly programmed PIC I don't think I'll need these transformers after all. Perhaps a DSP might be better suited for this application so I'll have to see how well it works out using a PIC.-Leo

[Guest MikePowell]

Leo,The ISBN is 0-87259-879-9. It's available from the ARRL www.arrl.org order number 8799. It's also carried by Amazon, which is where I picked it up.BTW, what's on your AIAA wish list? Sim proceedings?I agree with you about the Scott connected transformers. If you can generate all three stator voltages directly, and transformer isolation is not otherwise needed (it shouldn't be), then leave the transformers out.Mikewww.mikesflightdeck.comInfo for simpit builders

[Guest TOMCAT]

Hello Mike,I'm sorry, I can't help you with the books. Our technical references are all classified Lufthansa stuff - the aviation industry has always cultivated a certain safety spleen, but since 9/11 it's become a full fledged paranoia. Anyway I'm not allowed to make even photocopies of this material for my personal use nor do I have access to everything. But I definitely shouldn't think about further redistribution in any form.And the non-classified stuff we're using for training might be even older than yours :-lol - I got a whole load of copies and photocopied material here that was published between the mid sixties and early eighties. Spankin' brand new... :-lolSo myself is also always in the pursuit for good current documentation on aircraft technology and avionics. But due to the aforementioned circumstances it's quite hard to get your hands on such stuff.Regards ThomasBTW: I agree about the Scott T's - heard a rumor they might not be found in some latest avionics devices... ;)

[Guest MikePowell]

Thomas,Security for company specific material makes sense. I was hoping you might have some recommendations on public domain material. Avionics.com has two relatively new titles. One is by Helfrick and I expect it is good. Unfortunately, I cannot find it in a bookstore locally. I have to order it sight unseen and take my chances.Years ago I lived where there were several universities and a huge public library. I had access to university libraries and bookstores. I was in Heaven. Now I have to drive quite a bit to get to a technical bookstore. Best regardsMikewww.mikesflightdeck.comInfo for simpit builders.