3ph motor winding 2

[retailer]

I have a milling machine that has the gearbox and motor share the same casting, it was originally 415v 3ph 50hz 2 speed, this coupled with a 4 speed gearbox gave 8 speeds. At the time as I only had a single phase supply and also as VFD'd were out of my reach pricewise I had a local winder rewind for single phase. It was not 100% successful, low speed lacked any power and was unuseable, however I managed with only the 4 speeds. Now that VFD's are very affordable I would like to rewind the motor back to 3ph 240v and use the VFD get that second speed.

I have lost contact with the original motor winder but have the following info from a different forum posted by a member that has an identical machine and has had his rewound to operate on 240v

Original nameplate specs:
3 phase 50Hz 415V speeds: 2810/1390 Amp:1.6/1.3 kW:0.66/0.48
'Dahlander', 'pole changing' or 'consequent pole' motor
General winding detail:
Slots 24
Coils 12
Pitch 1-6-8
Coil arrangement D, Y-Y (delta, double star)
Original coils for 3 phase 415V: 213 turns of 0.5mm diameter wire.
New coils for 3 phase 240V: 123 turns of 0.67mm diameter wire.

As the motor will be running from a VFD I feel I don't need the 2 speed option so a straight 4 pole delta connection is all that is needed for a speed of around 1400rpm. I have contacted local rewind shops and they are quoting figures around the $700 mark which is out of my budget but I often wind Tube output transformers and have the wire, insulation, varnish etc and feel it is with in my capabilities.

After reading up about 3ph motors on the internet I have drawn up a proposed winding layout and need to know if it is correct, I need it work first time as I don't want to risk damaging a new VFD. I have posted two images, one shows all the windings and interconnect between them and how they are placed in the stator slots, the other pic shows only one phase for clarity and the interconnects between the windings , the same interconnects between the windings is copied for the other two phases. Essentially each phase has 2 groups of 2 coils each in series I am expecting this to be 4 poles 240v 3 ph with a speed of approx 1400rpm. Will this work ?

Pics are here:-

http://s20.postimg.org/mhqbnokst/my_ph_motor_winding.jpg

http://s20.postimg.org/s76k7zqz1/my_ph_motor_winding_single_ph_shown.jpg

 

[mikekilroy]

u lost contact with the mfgr, but does that everyone has?

HOw about their name? maybe someone here has contact with them?

 

[mikekilroy]

I meant milling machine mfgr and spindle mfgr themselves; they may still be around and know exactly what you need or can do it.

 

[retailer]

The machine in question is a mid 1950's manufacture Arboga, I have emailed Arboga and they have referred me back to the motor winder that changed the motor from 3ph to 1ph for me, I have lost conact with him, the workshop he operated from has been turned into a row of townhouses, it has after all been over 20yrs since the rewind was done. The motor was originally a Dahlander or consequest pole type, this type of motor is a single voltage motor only, in this case 415v, I do not have access to a 415v supply so hence the need to rewind for 240v 3ph so a 240v 1ph to 3ph inverter/VFD can be used to power the motor.

I have included links to two diagrams of what I believe the winding arrangement to be, however I'm not 100% sure and would like someone with some experience in this field to either confirm ot possibly point me in the right direction, it would be much appreciated.

mikekilroy:- BTW I have read one of your FAQ's RE: the inductance meter, I hadn't thought of this before but will certainly test the windings once I'm done, I do own and use an inductance meter as an aid for winding tube type output transformers.

 

[electricpete]

I have some familiarity with winding large 3-phase motors.
I'm not at all familiar with Dahlander and single layer windings.

fwiw, your proposal looks reasonable to me.
I gather we can just consider the high speed as Delta winding and compare to that.
You're keeping a delta configuration and reducing supply voltage by sqrt(3).
Number of turns goes down by sqrt(3) to keep volts per turn the same.
Area of turns goes up by sqrt(sqrt(3)) to increase wire crosss section by sqrt(3)

Full pitch winding seems unusual to me from my experience with 2-layer windings with fractional pitch to limit harmonics. For single layer winding, I guess there is no alternative.


=====================================
(2B)+(2B)' ?

 

[electricpete]

Just from curiosity, where did the nomenclature S1, S2, F1, F2 come from?
(I'm used to seeing that on DC motors, along with A1, A2)

=====================================
(2B)+(2B)' ?

 

[electricpete]

Two corrections/clarifications in bold

I gather we can just consider the high speed of the old Dahlander winding as Delta winding and compare to that.
You're keeping a delta configuration and reducing supply voltage by sqrt(3).
Number of turns goes down by sqrt(3) to keep volts per turn the same.
DiameterArea of turns goes up by sqrt(sqrt(3)) to increase wire crosss section by sqrt(3)

Full pitch winding seems unusual to me from my experience with 2-layer windings with fractional pitch to limit harmonics. For single layer winding, I guess there is no alternative.

=====================================
(2B)+(2B)' ?

 

[electricpete]

Forget the comment about full pitch. I was talking about coil pitch, relating to what would be coil pitch if your concentric wound was a lap wound with coil sides in same position (it would be full pitch lap). I'm sure it is correct for single layer winding.

=====================================
(2B)+(2B)' ?

 

[rhatcher]

Pete,
S1, F1, S2, and F2 are the start and finish of the first and second coils of each group. This is used to determine polarity. With this in mind, we can see that Retailer has designed a concentric, 4 pole, dahlander (consequent pole) connection.

Retailer,

There are several things to consider here. First, the wire you have may not be suitable for VFD use. Typically, magnet wire with a 1600V insulation rating is used. However, it is in your favor that you choose 240V as the voltage spikes created by the VFD will be proportionally lower than those for 415V operation, meaning that normal, 1000V rated wire may prove suitable. I will advise that you use phase insulation between the end turns of the coils of different phases. Also, you should pick the lowest switching frequency that the VFD allows. Although this will result in more noise from the motor, it will also result in reduced voltage spikes.

Second, your original motor rating was based on a two wye (two star, YY) connection for the low speed. Your winding design is for a one circuit connection. If we assume that you intend to use a wye connection, then you will need to reduce the number of turns by a factor of two to compensate for the change in the number of circuits. Likewise, you will increase the wire size by a factor of two, or use two of the smaller wires in parallel. This results in 61 turns per coil (213/2) with two 0.67mm wires in parallel.

Third, your connection looks good except that on the three phase connection you mixed up F1 and F2 for some of the coil groups. Except for that simple mistake, you did a good job for someone who is self taught from reading the internet. Anyway, to keep things simple, I will use the labels as shown in this drawing to make the WYE connection. Red-S1, Blue-S1, and Green-S1 will be the line leads U, V, and W. Red-F2, Blue-F1, and Green-F1 will the the neutral connections (tied together). To keep things simple, you can tie the neutral wires together internally. There is no need to install leads on them since this is a permanent wye connection.

Finally, you may want to consider installing thermostats on the winding end turns and connecting them to the VFD. This is a low cost addition that will offer protection against overheating.

I hope this helps you. Good Luck!

 

[Motorwinder]

Without running the fluxes to make sure the turns are right, the connection looks good.

 

[zlatkodo]

You can get calculation either for one or two speed (Dahlander) winding for S1 duty cycle, if you have the correct input data.
This form should be populated.
Before that, pay attention to these explanations.

 

[retailer]

I believe the original motor was connected in Delta mode (4poles) for low speed and double Y (2 poles) for high speed, my plan was to use the same setup and run the motor in Delta mode for low speed and omit the leads that would enable it to be connected as a double Y motor, then use the VFD to get the high speed. In other words duplicate the original winding layout but for a lower voltage (240v instead of 415v) . I may be wrong here, my experience with electronics is limited to a good knowledge of tube amps and tube amp transformers.

It was hard to visualise a rotating magnet field and it wasn't until I drew a series of plan views of the complete winding layout and marked the changing north and south poles at 60deg intervals for each phase that I saw how the poles rotate around the stator


electricpete:
I arrived at the turns and wire size by a different method.

My reasoning for choosing 123 turns of .67mm wire is as follows:-
The turns per volt should remain the same (less volts means less turns) turns per volt remain at 0.513 (this is considerably less than the average tube amp transformer which can be as high 7 or 8 turns per volt).
Less volts means more current (for the same motor power) so thicker wire is needed to maintain the same current density of approx 7A per sqmm.

rhatcher:
thank you for pointing out the error I made and for the correct connections.


I plan to use the same enamel covered wire that I use for transformers, some of which operate at 550v DC, I'm assuming it will do the job, I usually test the transformers with an insulation tester at 1kv.

zlatkodo:
thank you for the link I will need to strip the motor down to gather all of the data, it is at present working as a single phase motor, the VFD has been ordered and should arrive anyday. I am a bit hesitant to strip the motor down, but the desire for a full range of speeds is strong so it will be done

I really appreciate all of the help and replies, I know it can take valuable time time out your day. It may take a few weeks but I'll post the results.

 

[ScottyUK]

I expect that the 2-speed stator will be generously sized to accommodate the two-speed windings, in comparison to a single speed machine of equivalent rating. With plenty space for windings it should be fairly straightforward to rewind it as a single-speed machine for high speed, and use the VFD to provide the speed variation. Of course I can understand why you might want to keep things 'standard' too, especially if this is a good example of an old machine - I have a lovely old Holbrook lathe which dates from the 1940s and which is largely original, although the old motor has now gone the journey.

 

[rhatcher]

Retailer,

I misunderstood your first post to describe the motor as delta for the 2-pole high speed and 2-wye for the 4-pole low speed (constant horsepower connection). It now makes sense that it would be 2-wye for high speed and delta for low speed (constant torque connection).

This being the case, the connection and turns that I gave are wrong. They were based on the original winding being 2-wye for the 4-pole low speed.

If you do want a 4 pole connection, you will need to keep the turns at 123, the wire size at 0.67mm, and connect the winding as a delta.

However....Keep in mind that with a 4 pole winding you will not have the same performance as the original 2 speed winding. This is because the original winding was a constant torque connection, meaning that high speed (3000rpm) winding produced the same torque as the low speed (1500rpm) winding and, therefore, the motor produced twice the horsepower at high speed as low speed.

When operated on a VFD with a 4 pole winding, the motor will produce constant torque up to the base speed, 1500rpm/50hz, and constant horsepower (reducing torque) up to about 2250rpm/75hz.

I recomment that you consider winding the motor as a two pole. When operated on the VFD, you will have constant torque at both the base speed of 3000rpm/50hz and at the reduced speed of 1500rpm/25hz with the high speed horsepower being twice the low speed horsepower (just like the original).

If you are interested in this option, you will need to change your connection so that the second pole of each phase has alternate polarity from the first pole. If you use a 2-wye connection then the turns and wire size can remain the same. If you use a 1-wye connection, you will need to change the turns and wire size as we discussed before to compensate for the change from 2-wye to 1-wye (61 turns, 2 - 0.67mm wires in parallel).

 

[petronila]

Hello Retailer,

Maybe you can try first the Steinmetz connection. You will loose 25% of power, but you could avoid motor rewinding and redesign.

See Attachment


Note : Triángulo is Delta and Estrella is Star

Regards

Carlos

 

[retailer]

rhatcher:-
Thanks for the explanation regarding the differences in torque between the different connection setups, it is obviously better to connect the motor as a double Y for high speed and use the VFD to get the low speed, I did see this mentioned on another forum but there was no explanation as to why it was better. This what I now plan to use.

If I understand you correctly I will have 2 groups of 2 coils for each phase, the coils in each group will be connected in series but the groups will be connected in reverse series that is the finish of the first group will be connected to the finish of the next group, this will give 2 wires for each phase plus a centre tap connection as shown in diagram 1 that I have attached. I will then connect the ends of each of the phase coils into a delta connection as in diagram 2. A long flying lead is then attached to each of L1 L2 and L3 and also to each of the centre taps, this will give 6 leads in total to be bought out to a terminal block where the final connection will be made as either delta or double Y. To run as a double Y I then connect L1 L2 and L3 together and the the 3 phases from the VFD then go to the each of the CT leads, I can then change the direction of rotation at the VFD if I need to. The final connection will look like diagram 3.

http://s20.postimg.org/md9zu3dcd/Diagram_1.jpg

http://s18.postimg.org/ormusbw15/Diagram_2.jpg

http://s12.postimg.org/gi8hwx4od/Diagram_3.jpg

 

[Motorwinder]

Try this connection.

 

[Motorwinder]

opps, here's a double Y connection for a 2 pole

 

[electricpete]

For fixed speed motor, we know you get more horsepower from your materials for 2-pole compared to 4-pole. But if you are going to take these two motors (a 2-pole and 4-pole) and put each of them on a vfd and drive them at 1400rpm, doesn't that 2-pole advantage disappear? Maybe it would make more sense to stick with 4I'd think you might want to stick with 4-pole for the shorter endwindings?

Another thing comes to mind in trying to recreate a winding by comparison to original is that it might not be a good target to recreate original high-speed winding volts per turn if the coil pitch changes significantly. I'm thinking the original winding probably had an oddball coil pitch not much more than 50% of the high-speed pole pitch. If you rewind it for coil pitch = 100% of high-speed pole pitch, then you've got almost twice the area and twice the flux for a given turn, I'd think you'd want to almost double volts per turn of original coil in high-speed configuration, right?


=====================================
(2B)+(2B)' ?

 

[electricpete]

My previous comments address the coil pitch for analogous lap winding.
Concentric brings something new to think about.
Do we normally put less turns in the smaller inner coils of concentric than we do in the larger outer coils of a group?

=====================================
(2B)+(2B)' ?