Electric Motor Guru Question

snip

Wiring it with the shunt open is the only way that you could wire it for no current in the shunt. So you need to specify what you mean when you say "no current in the shunt".

Your newsgroup agent is mangling the links (mine is just showing what got posted to the group), photobucket can't find anything. Figuring out links (tinyURL?) would help a lot, I suspect.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html

or

>>>

snip>> Wiring it with the shunt open is the only way that you could wire it forno current in the shunt. So you need to specify what you mean when you say"no current in the shunt".> Or you can connect one end of the shunt to a resistor and the other endof the shunt to the other end of the resistor, then connect one end to B+.Yes it was a mistake, but it worked best this way.> Your newsgroup agent is mangling the links (mine is just showing what gotposted to the group), photobucket can't find anything. Figuring out links(tinyURL?) would help a lot, I suspect.> Thanks, I hope I fixed them. MikeK> -->> Tim Wescott> Wescott Design Services>

>

Do you need to implement control loops in software?> "Applied Control Theory for Embedded Systems" was written for you.> See details at

All links fixed )-: I need some help explaining a motor I have been using for over a year. It is a shunt field motor. Please stick with me as I develop the situation. If you know of better place to ask me question please let me know. (forums?) I have a shunt field motor that I have been using with a series field controller. I have been using this on an electric gokart and it has worked fine for over a year. Shortly after finishing the wiring and testing (it ran great btw) I realized I made an error when wiring the shunt. I wired it so no current flowed in the shunt, The wiring put a 20 ohm resistor across the shunt. Here's a simplified schematic of the wiring. The 20 ohm resistor is for current limiting, as I'm overvolting the motor.

I wired it properly and the motor ran slower, so I ran the shunt open and this cause my controller to become defective ( bad day). After I got the controller fixed I wired it as original (no shunt current) and we have drove it for a year. Here's a link showing all three wiring diagrams.

So, how does this even work with the shunt wired to the 20 ohm resistor and not current through the shunt? I expect some answer about current induced by rotor flux, but we'll see.

Next problem, the motor smoked the other day, I pulled it apart and found the field overheated. The field has 4 poles and is a bit odd. Two poles use #16 wire and two use #24 wire. The field measures 41 ohms and has not changed in the last year. I can't tell if the #16 and #25 are series or parallel, however, I think it is a good guess they are in series because I think the #16 wire fields would have much less than 41 ohms. The fields do show opposite magnetic fields as I go around the stator. (Tested with small magnet on a stick). Here's a picture of the stator with two different wire sizes. Note three different color wires in #16 wound field.

I understand the motor could have compound or interpole windings, but I don't think either of these are true, based on my measurements. Here's a diagram of the motor connection. Only three wires out labeled A2, F1 and F2.

Here's the original motor wiring plate, I don't think it is relevant because it is clear this motor has been rewound. So, Why the different size wire in the field?

The #24 wire is overheated at one end of the motor, any idea why it would have overheated, since I don't battery current through the shunt?

I will be happy to do any more measurements requested if I can.

Any help to enlighten me is appreciated. Thanks, MikeK

>>>

snip>> Wiring it with the shunt open is the only way that you could wire it forno current in the shunt. So you need to specify what you mean when you say"no current in the shunt".> Or you can connect one end of the shunt to a resistor and the other endof the shunt to the other end of the resistor, then connect one end to B+.Yes it was a mistake, but it worked best this way.> Your newsgroup agent is mangling the links (mine is just showing what gotposted to the group), photobucket can't find anything. Figuring out links(tinyURL?) would help a lot, I suspect.> Thanks, I hope I fixed them. MikeK> -->> Tim Wescott> Wescott Design Services>

> Do you need to implement control loops in software?> "Applied Control Theory for Embedded Systems" was written for you.> See details at http://www.wescottdesign .com/actfes/actfes.html

I'm not sure why it worked in case A. I think you really had it connected such that the shunt winding was across the field winding (i.e. the shunt connected on the opposite side of the motor than you thought). You could test this by wiring it as in "A" with an ammeter in series with the resistor -- I expect you'll see current flowing, consistent with the average voltage out of the PWM.

In case C you turned it from a motor into an inductor, then you put DC into it. This was a lot like driving your PWM controller straight into a dead short, and it became dead, shortly.

In case B you were putting _lots_ of current through the field winding. This makes a motor very tourqey, and slow. The reason is because the constant of proportionality for torque/amp is proportional to the field current, and it is (if you get your units right*, and have a lossless motor) exactly the same as the constant of proportionality for back-emp/speed.

So, you made that motor constant big which made the back-EMF big, which means that the motor doesn't have to go very fast to cancel out the terminal voltage, which means that the motor doesn't go fast.

I'd experiment with putting a resistor in series with your field winding (what you're calling the shunt), either before the PWM controller (drawing B with a resistor in there), or after (drawing A with the resistor moved to the other side of the motor, if drawing A really is correct). Then I'd experiment with the resistor value. If your field winding resistance is 40 ohms, I'd start with a 40 or 80 ohm resistor, and go from there. The bee's knees would be to put a small controllable supply on the field, with a "gearshift" knob on the dash.

The bigger you make the resistor the more speed you'll get out of the motor, but the less torque for the input current. It's a tradeoff.

• If you express your motor constant as Newton-meters/Amp, then with a bit of light** dimensional analysis you'll find that the motor constant in Volts/(radians/second) is identical. This isn't just nifty dimensional analysis tricks: it's the first law of thermodynamics at work; saying that if you put electrical energy into the thing you have to get exactly the same mechanical energy out, or visa-versa.

** Hah!

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html

What made you decide that it's a shunt, not series? Is the motor case connected to anything?

Grant.

>>>

I new that would be the first response :-) I have checked this several times, several ways, and then a year later still questioning it, I checked it again, by following the wires, and using an ohm meter.

I'm not PWM the shunt current, The plan was to just supply a continuous dc flow through the 20 ohm dropping resistor. (I'm running a 28volt motor on

48 volts.)

I think you are saying with no shunt current the rotor sees no back emf, I have several arguements why I don't think that is the case, the easiest is the controller is current limited, second the fault indicated was on overvoltage and it cleared as it should the first time, it didn't clear the second time. The factory said it was an overvoltage fault when the got it.

_lots_ is a little subjective, it was about 800ma. A 41 ohm shunt resistance plus 20 ohm resistor is 61 ohms. 48 volts/ 61 ohms = 800ma

I have the bee's knees, I'm thinking I have a 50 ohm reostat on the dash. I never seemed to have a noticeable effect, but I have the odd hook up. Here's a video showing my dash, the knob in the upper left of the dash is on the reostat.

Ya, at the moment we have a less powerful series shunt motor, it does not go as fast but it has more torque, my son likes it better because it is easier to do spins! (donuts)

Just a guess. What you have is actually a compound motor*. It has both a shunt field and a series field. By mis-wiring the shunt field, it acted as a series motor.

Two things can happen: 1) A series field motor, unloaded can fail by overspeed. 2) Even when loaded, the lack of the shunt field will result in lower back e.m.f and, as a result, higher armature and series field currents. Which could lead to overheating.

*There are other strange combinations of fields that you might have as well which, when mis-wired, will behave somewhat like series motors.

--
Paul Hovnanian     mailto:Paul@Hovnanian.com
------------------------------------------------------------------
Ask not for whom the  tolls.

Because the shunt has a #24 wire with 41 ohms of resistance, if I put that in series with the .2ohm (or less) rotor windings it would not move. The motor will draw as much as 250 amps of currrent or more, I pin my

250 amp current meter on hard acceleration. MikeK

That would certainly explain why there's two different field windings, wouldn't it?

(I hadn't thought about that. All the motors I've ever controlled have been permanent magnet DC motors, so I haven't had my nose rubbed in all the externally-excited motor possibilities).

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html

I have been through that thought process and cannot find any logical way to make that work. This page shows 3 possible combinations of compound motor wiring, see figure 12-17.

If the motor was like A or B, I could see that with a couple of ohm meter tests. I have checked it is not wired like that. If it was like C I could ohm that out also. I have checked it is not wired like that either. It is wired just like this url shows.

This is not possible because the shunt has a very high resistance,

41 ohms. The motor will draw 100s of amps.

Please look at this drawing, the motor only has three wires out, labeled F2, F1 and A2, logicaly F2 is connected to A1 internally. Oh darn I see in parentheses I have A2 it should be A1 For clearity, A1, A2 is armature and F1, F2 is field.

Thanks for your thoughts, I have been through this thing trying to find some connection other than what is in the above drawing and what you see is all I can find. MikeK

Fair enough :) I was wondering if there was some compounding connection because of the different wires sizes you reported.

Or a different way to connect the field. Leaving it open may have voltage spiked the insulation, let the magic smoke out.

Grant.

I agree the differing wire sizes is odd and, internally every thing is bundled, sleaved and varnished, making it impossible to trace connections. It has only three wires exiting the motor.

Ya, only one end seems over heated, it is the end nearest the brushes. Where can I purchase small quantities of magic smoke, I don't need a lot, I stopped before much came out. :-) MikeK

The armature may draw that, but that current doesn't have to pass through the 41 ohm field.

You know this how? By using an ohmmeter on the external connections? That may not reveal a second series field. In a previous post, you stated that you took the motor apart and found two diffecet field windings on 4 poles. So, your drawing below doesn't seem to reflect the motors innards.

How many brushes does this motor have?

--
Paul Hovnanian     mailto:Paul@Hovnanian.com
------------------------------------------------------------------
If God is perfect then why did He create discontinuous functions?

Hi guys, and thanks for the help. I did some more measurements and I'm much closer to the answer. It looks like I do have a compound motor, couldn't tell with my ohm meter, I had to measure 0.01 and 0.001 ohms. I used a constant current source to measure between leads to get the new info. I have a new drawing with a description of estimated turns and length of each turns to find the resistance of each coil. The numbers don't work out, off by a factor of 15. However by energizing the leads I find a magnetic field on one set of coils but not the other. So now I'm certain of a separate series field. I have a drawing with the two possible compound wiring scenarios. I don't know a way to figure out which scenario I have.

Thanks, MikeK PS. I still need to go back and see how this fits with my improper wiring that worked so well.

This would certainly explain why the "wrong" connection of the 20 ohm resistor worked, but not why the controller blew with the "open coil" version -- unless that was just happenstance.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html

Yes, I have gained much info on the motor in the last few days. It was given to me, after someone disassembled a military lift. To him a 28 volt dc motor was junk, it was just what I was looking for at the time. As far as blowing the controller, I'll never know. So know I have figured out that I have a Short Shunt Compound Motor. Here's the method I used,

At the end you see I have one more question about the motor. Is it a cumulative compound or a differential compound motor? Here's a diagram showing that is just the difference in polarity of the wiring. I cannot see the connections in the stator so I've tested the magnetic fields with current flowing. I'm pretty confused by the data, I will post it later and see if anyone can decipher it. Thanks, MikeK

Now I want to know if my motor a cumulative compound or a differential compound motor? Here's a picture showing the polarity of the magnetic fields of the motor when I apply current.

The fields are reversed leading me to think the motor is wired as a differential compound motor, and not a cumulative compound motor. But I have no confidence in that opinion. Anyone have any ideas? Any other tests I could do for more info? This link shows the polarity difference between the two styles. Thanks, MikeK

it sounds like you have a compound motor ? In which case, you can use the motor with out the series winding and directly connect to the armature, this of course implies that you connect the field in parallel to the same A1 and A2 wires where the motor will operate the same direction AC/DC, or, you can connect the series field S+, S- in series with A+ and A-, with the field connected to a + and - line you'll get a motor that has torque enhancement. If you alternate the S+ and S- series field with the armature field, you get Speed enhancement but a softer bottom end torque.. This is because the series field is with in the shunt field and that is counteracting with the Field's strength.. This type of motor is good for auto field weakening/strengthening effects with out the drive electronics doing this to the shunt field to give the enhancements to the motor you're looking for.

In DC shunt motors, weakening the field yields more speed but less torque where as, strengthening it will yield the torque and lessen the speed..

In today's DC shunt drive electronics comes auto field weakening so the need for compound motors are not so prevalent any more, they still can be employed to enhance it even more, also, problems exist in reversing the direction in a compound motor. In order to maintain torque enhancement in compound motors when changing directions, you must reverse the S+ and S- leads that are joined to the A+ and A1 leads, other wise, you'll be doing the opposite to the shunt field... Older technology reversed these types of motors by connected these leads to a reversing contactor, rated for the load along with the armature leads to maintain the effects on the motor they are looking for, speed or torque in either direction.

In today's world where new drive electronics are connected to such motors, most of the time the series fields are not used.. They are capped and tapped off...

In your case, if you were not concerned about reversing the motor, you can connect the wires this way..

S+..S- to A+..A-

The shunt field should be connected like this.

F+ to S+ and F- to A-

and you would have a single + and - power drive source that also connects to + to the (F+,S+) and - to the (F-,A-)

This will give you a clock rotation looking into the face/shaft drive of the motor with torque enhancement.

This will give you a constant torque with the series field varying the field to adjust torque demand verses speed demand naturally..

Most people do not wire a DC shunt motor this way, most practices place the shunt field at a fixed DC source.. This will give you lots of torque at the bottom end and soften as the speed increase.. The series field can help adjust for this..

have a good day..

your motor is differential acting only if for example, the Series field is weakening the shunt field.

If you do not have the series at the same polarity as the field, you are in differential mode (reducing field strength as current increases through the series field)

And of course, less overall field means less torque and more speed but does serve for some nice tension mode starts for some applications.