Star to Delta Conversion of Newman (Elf) type, 3 phase motor installed on a Meddings M4B pillar drill

[Ro80RobParticipant @ro80rob]

My first real post so here goes.

I hope it’s useful and it is rather lengthy, but it may answer some members questions. Apologies for any typos/errors since after posting editing by the OP seems not to be available.

Firstly, this task involves actually making electrical changes to a 3 phase mains motor, in converting it from star to delta configuration. This enables the motor to work with a nice cheap and cheerful Variable Frequency Drive (VFD) maybe from your favourite online auction site or perhaps that rainforest place (I’ve bought from both online).

Consequential advice.

If you are not confident what you are doing in relation to this exercise, don’t let enthusiasm get the better of you, ask someone who does. I can take no responsibility for your own work. This posting just shows what I did and how I did it. I am a qualified and time served electronic engineer and I was confident I could make the physical changes required to the motor safely. I have already converted a Brook-Gryphon motor for another pillar drill and to be honest that was easy enough for me to do. You should think hard on your abilities and competency before embarking on any similar mods to an older 3 phase motor that didn’t leave the factory with the ability to change from star to delta easily.

That said, a little pre-amble……

I bought myself a Boxford-Union pillar drill (£150, found locally) during COVID for a little project, and I converted that to use a single to 3 phase VFD as I wanted to use the speed control and single phase mains. I find that invariably pillar drills don’t spin slow enough for what I wanted to drill (and tap) and the VFD was the way to go. The drill came with a 4 pole, 3 phase, 1/2 HP Brook-Gryphon motor, which I had to open up and find the star point, separate and bring those winding ends out on new wiring and configure them into delta mode for my VFD. The image below shows how it finished up….

I used a ‘simple series’ VFD (0.75 kW XSY-AT1) as shown. It’s cheap, easy enough to setup and works well. There are higher power versions of this device, and I’ve used one of those on a lathe I also have (Chinese GH1440A, retaining all the original controls and contactor interlocking with the VFD). I had to change the motor for a new single speed item as the installed motor had combined 2 speed windings that were never going to allow it to run satisfactorily. Anyway, the lathe also works great, but that is for another time.

(See my next post)

[Ro80RobParticipant @ro80rob]

[Ro80RobParticipant @ro80rob]

(Carrying on from previous post)

I went to town on this drill (note the original control switch was missing). The controls on the side of the drill allow me to select forward and reverse as well as varying the motor speed on the ‘knob’. The rocker switch allows me to send the direction control down to a 2 way (forward/reverse) foot pedal arrangement that I can use when I’m using the drill for tapping at low speed, leaving both hands free to manipulate what I’m working (usually aluminium). There is also a separate foot pedal emergency stop. The eagle eyed among you will notice that the motor on the machine is no longer the Brook-Gryphon motor. I got the chance of a new modern 8 pole 1/2 Hp 3 phase motor very cheaply (£45 delivered) that ran much more slowly than the Brook, with better low speed torque, so I grafted that on with a new pagoda pulley. The Brook motor I use as my VFD test load now.

However…..

Pleased as I am with the result, we have a bench mounted Meddings M4B drill in our little workshop at work. I think it’s also almost as old as me (and I’ll be retiring in 2026), but it is a super nice piece of kit with its high/low gear selection. Those in the know may be interested to note that it still has its original phenolic paper (tufnol) high range gear as the drill has had a really easy life, added to the fact that it is mainly used in low range. It really is a Rolls-Royce bit of kit and I’ve done a lot of work (and homer) jobs on it over the years and I have always hankered after one.

Whilst perusing our favourite online auction site (again) a few weeks ago, looking for a birthday present for wifey to ‘buy me’, I came across a similar machine, in good cosmetic nick and with its high gear recently replaced by a Meddings factory sourced delrin replacement. The seller rightly wanted proper money for this nice bit’o’kit (£450), and it was a bit of a drive to get it, but I did the deal and went to collect. Wifey even helped me get it out of the boot when I got back home (in several pieces) as it’s much heavier than the Union.

(See my next post)

[Ro80RobParticipant @ro80rob]

(Carrying on from previous post)

Once again, with 3 phase not being available in my garage/workshop, a VFD conversion was required, meaning surgery inside the motor. Despite having converted the Brooks motor, information is scant online in making conversions and info on making the necessary changes to the drill’s original Newman motor was proving elusive. There is a nice picture of this motor’s windings on this site, but that is all. That owner also wanted to convert the motor to VFD driven single phase operation, but wisely decided that was a step too far for him. Not deterred, I decided that I would look into the prospects for conversion myself, with the acceptance that if it wasn’t possible/straight forward, I would have to bite the bullet and buy a modern motor to fit instead. I do like to have a fallback position but like to keep it cheap.

So. I was finally ready to work on the Newman (sometimes also additionally badged as ‘Elf&rsquo 3 phase star setup motor. I have a good amount of photos, so hopefully you can see how I did it. The motor ‘boiler plate’ is as below. Looks to be a 1964 item.

Firstly, I removed the motor mounting flange from the drill with the motor attached and I mounted it in the ‘jobber’ workmate I have fitted to a set of old wooden steps. These are a much better height than just the workmate as I wanted to stand up to address this work, but a workmate (and seat) will do.

Turning attention to the electrical connections on the other end of the motor, beneath the ‘oval’ plate (two slotted screws) I found the electrical connections and a rather oil soaked piece of paper with info about the motor. The 3 terminals on the tufnol insulated plate are for the 3 phase winding connections and the screw into the motor for the Earth connection.

I released the 3 black wires from the terminals as they need to push through the hole in the tufnol plate as the bottom face of the motor is removed.

Now I undid the 4 through dome headed nuts around the motor casing perippery, that hold the motor front and rear end plates onto the motor central body (onto which the foot mounting plate is attached). Note that the nuts unscrew from the pulley end of the motor. The rear 4 dome headed nuts seen above are held in captive recesses in the rear motor plate, so they will not unscrew. I ended up with 4 long shafted threaded rods with 8 nuts that I set aside somewhere safe.

(See my next post)

[Ro80RobParticipant @ro80rob]

(Carrying on from previous post)

Then using a large flat screwdriver and a rubber mallet, I placed the flat blade against the slightly raised edge of the rear housing where it met the main body and using the screwdriver like a chisel, tapped the rear housing backwards (i.e. away from the pulley end) and off, carefully pushing the wiring through the hole in the tufnol terminal plate as it came off. This wasn’t hard to do. Using the screwdriver like a chisel and gently tapping on each side soon opened up a gap between the body and the rear flange. Then using two large flat blade screwdrivers (one each side), I carefully inserted them (not too deep so as not to damage the windings) into the gap that had opened on up opposite sides and levered the two screwdrivers, so as to ease the back face from the motor, pushing the electrical wires through the holes in the insulated tufnol plate as the back face came away. I noted that the rear bearing remains located in the rear housing, held in by two slotted screws and clamp plates.

I then used a similar technique and removed the front face complete with the motor’s rotor. I was then left with the central motor winding section of the motor.

Note:- I then undid the Allen key retaining screw from the pulley, removed the rather loose fitting Woodruff/drive key and used a 3 legged puller to get the pulley off the shaft, as it wouldn’t quite come all the way off by hand. I noted that the front bearing remained on the motor shaft and I left it that way as the bearings seemed good.

Since I required to work on the electrical terminals end of the motor central housing, I decided to undo the motor from the mounting flange (4 bolts) and turn it around and re-fit it to the mounting flange. This then allowed me to mount the remains of the motor in the workmate in a vertical plane, clamping onto the motor mounting plate. I added some bubble wrap so I could ‘sit’ the motor down and to protect the delicate windings on the opposite end from damage.

(see my next post)

[Ro80RobParticipant @ro80rob]

(Carrying on from previous post)

So far so good, I now had to find the ‘star’ point. The point where the opposite ends of the 3 sets of windings (i.e. U2, V2 & W2) were joined together. It was impossible to see by eye where the joint was. There were many joints covered in hard red sleeving. All of those that could be seen easily seemed to only join 2 wires or windings together. The whole lot had also been give a light coat of varnish after being laced down with some flat almost vegetable (raffia?) looking lacing cord. There was only one thing for it, I had to start cutting the lacing cord and inspecting the many joints for the three wire joint. I thought (logically ?) that the joint I was looking for could well be on the opposite side to the 3 black motor wires seen to the left. As I said, there were joints everywhere, so I started there as below, cutting very carefully with a scalpel, inserting the blade into air gaps in the lacing and cutting upwards. My aim was to always avoid damaging that precious and rather old insulation on the windings. Progress is shown below…

I then teased the joints carefully by hand and plastic spatula away (looking after the insulation as always) so I could inspect them and look for the three wire joint.

No luck there. All joints were just one wire going into each end of the red sleeves, so I carried on cutting more and more of the lacing cord carefully, ending up cutting every piece of the lacing cord and teasing and inspecting the joints until finally I found the triple joint sleeve right by the 3 black wires that connected to the terminal plate. That maybe important knowledge for next time (if there ever is one). I also carefully fished out all the loops of lacing cord through the windings as they were certainly never going to be any use again. As always, I was careful not to damage the windings or insulation.

A view (below) of the star point (soldered joint) with the red insulation sleeve carefully removed via more judicial scalpel and small wire cutter work.

I decided to ‘identify’ the wires at this point as shown. I don’t know if the ordering is important (i.e. could I have arbitrarily labelled the wires, but I decided to label them left to right as they came out of the motor stator, anti-clockwise as I looked at them. I’m sure someone will tell me if they can just be labelled arbitrarily.

(See my next post)

[Ro80RobParticipant @ro80rob]

(Carrying on from previous post)

Anyway, before I cut off the solder joint, I decided to do a quick insulation test with a 500 V DC tester I borrowed from the electricians at work. I connected the +ve lead to the soldered joint and the -ve to the body of the motor and switched it on. I got a reading of 145 MOhms (million Ohms), which I thought was pretty good. I hadn’t damaged any insulation on the windings it would seem.

I then cut off the solder joint (as near to it as possible) and cleaned the insulation from the end of each wire by scraping a short length of it off again with the scalpel. I then did a resistance and inductance test on each winding, again with a borrowed meter.

The resistance and inductance of each winding was almost identical on each of the 3 winding circuits. All good.

I then soldered on some short lengths of wire to each and added a little heat shrink to each over the joints.

I made sure I had sufficient length to terminate these new wires in some solder on rings I had (I guess one could use crimp terminals), so as they would pass through the other hole in the terminal board when the motor was re-assembled. I then started to ‘re-lace’ the windings, well not quite….

As you can see, I used plastic/non-metal cable ties (sometimes called zippy clips). I did have some high temperature (180 deg C) flat lacing cord, but I also got hold of some special 3.5 mm wide 150 deg C zippy clips from RS, which I thought would do nicely and were going to be less fiddly to install as I could tighten each one individually as I fitted them. I couldn’t find any high temperature zippy clips on our favourite auction site, but if you use your search engine, looking for ‘high temperature cable ties’ you will find a high temperature nylon range, good for 125 deg C, which is probably good enough and cheap (~£15 for 100). Also, the ones I had were only 100 mm long and weren’t quite long enough, so I had to join two together each time. If you get the 140 mm length ties, they should do the job with one zippy each. I completed the zippy ‘lacing’, ensuring that I ended up with the wiring in a suitable format to be able to pass through the terminal board in the back of the motor as I re-assembled it onto the motor body.

(See my next post)

[Ro80RobParticipant @ro80rob]

(Carrying on from previous post)

I then terminated the windings with my soldered on ring connectors and using a multi-meter on the resistance range, identified the winding ends (U1-U2, V1-V2 and W1-W2) and labelled them. As I said, I’m not sure if it’s important which winding circuit is labelled as which, but this is how it looked when I’d finished…

Turning attention to the rear housing, I added some nice similar brass BA screws I had in a draw that were identical size to the original ones on the terminal plate. However, these new screws need to be about 3 to 4 mm shorter than the originals, as the cover plate for the connections on the back of the motor is on a sloping face, which requires the slightly shorter newly added screws so as not to short out onto the electrical cover plate when re-fitted. I also added some tape to make a label for the connections. Labelled up corresponding to info found on the internet and shown below.

Resulting in the below. Note the original larger holes at each end to pass the motor winding cables through. Sorry it’s labelled upside down compared with the sketch above, but it’s clear enough. For Delta, U1 needs to connect to W2 etc…..

This terminal plate was then re-fitted into the back of the rear motor flange as below. I checked at this point that the added 3 screws were not too long so as to short out on the terminal cover plate when re-fitted. I also took the opportunity to undo the two screws holding the rear bearing in, tapped out the pressed in bearing cap on the rear face of the motor housing using a 3/8” socket extension, which allowed me to tap inwards the rear bearing for a clean and long life (and importantly not too much) re-grease. I then put it back in the right way round and re-fitted the clamp plates and screws. The bearing went back in easily and then I tapped the bearing cap into the other side of the rear face.

Then I offered the back cover towards the central section of the motor, feeding the wires through the holes appropriately, and tapped it onto the centre section locating ring/edge with the rubber mallet. Result was as below…

…. and connected the wires to the brass terminals according to my labels. I also marked up the inside of the terminal cover plate for clarity with a permanent marker pen…..

(See my next post)

[Ro80RobParticipant @ro80rob]

(Carrying on from previous post)

I now stood the motor on the rear housing (on my bubble wrap) re-fitted the rotor into the motor, making sure I didn’t push the rear cover off as I did it (since it’s not actually properly secured on yet). Gentle taps with the rubber mallet on the end of the shaft were all that was required until the shaft was fully home/seated.

I re-greased the pulley end bearing too, and replaced the wavey spring washer back into the top housing bearing recess, using 3 dabs of grease to keep it in as shown below…..

I then slid the front cover onto the shaft, aligned it appropriately and tapped that onto the motor centre section with the rubber mallet.

I was a little surprised that the motor shaft didn’t turn at this point. I realised I needed to re-fit the 4 motor through bolts and domed nuts and tighten them up in order to pull the whole job fully square. The shaft turned easily then.

Now into the home straight, I needed to link the motor connections accordingly to set the motor to Delta mode. I had a little bit of thin brass sheet, so made myself some jumpers, but I could have just as easily wired them across. You may have noted that I had gotten a little excited and put the jumpers on before fitting the through bolts, but the image does clearly show the captive recesses for the rear through bolt domed nuts, illustrating again how I had to to remove and tighten the pulley end domed nuts to strip and re-build the motor.

I then did another insulation test with the 500V DC tester. Since in Delta mode, all windings are effectively connected together so again only 1 test is required. I got a value again of ~145 MOhms, which was very reassuring. I also checked the resistance between each terminal jumper and they were all almost identical (and importantly not a short circuit), which I was also happy with.

(See my next post)

[Ro80RobParticipant @ro80rob]

(Carrying on from previous post)

I then connected it up with my test VFD and everything worked fine, with the motor running smoothly.

I finally re-fitted the pulley, though I did make a new key for the shaft/pulley as the original genuine one was rather sloppy in the motor shaft keyway and I didn’t want the pulley twisting backwards and forwards when I reversed the motor. I then stuck a couple of inches of tape on the pulley to use my electronic reflecting optical rotational speed sensor (available cheaply online) and checked the unloaded rpm of the pulley with the VFD frequency set to 50 Hz. It came out as ~1400 RPM, and the motor plate says 1425, so pretty much spot on then (I watched a YouTube video where the 3 phase motor was purposely wired up incorrectly and that ran considerably slowly and took a lot of current). I also checked the consumption of the VFD and motor running at 50 Hz unloaded using my electricity supplier provided smart meter readout. It came out at about 75 Watts (powering the VFD and motor), which also seemed OK to me. I’ve run the motor (unloaded) for an extended period and it has behaved faultlessly.

Looking at other motors that already have the ability (from factory) to change from star (~400 V) to delta (~230 V), it seems that the current increase to the motor is broadly a factor of 1.9 to 2 times the star current. Important to consider when thinking about the cable requirements needed and VFD selection, which as a guide should always be a bigger power rating than the motor intended to be driven by it. ‘What’ll do a lot’ll do a little’ as the saying goes.

I hope my description of this conversion is useful to forum members. As I said, I converted a Brooks motor similarly and that too ran perfectly afterwards.

However, do take care if you have a go yourself!

Cheers… Rob.

[Clive FosterParticipant @clivefoster55965]

A most excellent write up Rob

The star point is never where you expect it to be. I used either proper bootmakers / saddlers linen thread or old fashioned brown string (both probably unobtainable these days) to redo the lacing on the three or four I did in the past. Several cats of shellack stiffened the lacing nicely and bonded it to the windings. Hafta say I'm unconvinced by zip ties as lacing substitutes. They ought to be just fine but windings do vibrate so I feel that bonding a thread the old fashioned way ought to be more secure. But maybe its just my luddite tendencies surfacing due to old age!

Before resorting to a sharp blade to initially separate the bell housings from the body it's worth going round a time or two with a dead blow hammer making carefully angled strokes. This often shifts things just enough for a kinder implement to be used to finish off separation. Next time I encounter a stubborn one I shall try making a proper puller using expanding grippers inside the bell housing bolt hose. A bit off creativity using "rubber" tube and small diameter studding ought to result in a suitable gripper.

Handy that your bell housing had room for a 6 terminal connecting plate. Many are a bit tight. Should I ever get round to doing the one on my Pollard drill I shall use an external box. Not an urgent job as I have a440 V three phase magic box and they 6 geared speeds cover a decent range.

Concerning the labels the modern Demo / Brother prints on self adhesive flexi tape ones do a most excellent job. My handwriting re-defines crap!

As you suspect the relative polarity of the windings is important and they must always be connected the same way round. So decent labels are good idea. Conventionally the ends that got to the star point are no 2 so you have got the labelling right.

If it is suspected that one, or more windings are the wrong way round here is method of checking polarity with improvised equipment.

——

Determining 3 Phase Motor Winding Polarity

A simple way to determine the polarity i.e. start and finish of the windings in a 3 phase motor is the inductive kick method.

Tools needed

Analogue multimeter e.g. AVO (UK), Simpson (USA), a 6 volt lantern battery or DC power supply, some crocodile clip jumper leads and some numbered sleeves or tape on numbers.

Procedure

01) Identify the pairs of leads for each winding using the resistance function of the meter and mark the pairs so you don’t get lost.

02) Connect the negative of the battery or source to one lead from one pair.

03) Set the meter to the 60 volt DC range across one of the remaining pairs.

04) Connect a wire to the positive terminal of the battery or power source and briefly touch the second lead of the first pair. The power source is across the first winding so current will flow.

05) When you connect the battery or power source you should see the meter needle deflect. Either forward or reverse.

06) When you disconnect the battery or power source you should see the meter deflect in the opposite direction.

07) We are looking for a forwards deflection of the meter needle when the battery or power source is disconnected. Reverse the meter connections if need be.

08) Identify the first pair of winding leads “Battery +” = T1, “Battery -“ = T4

09) Identify the second pair of winding leads “Meter +” = T2, “Meter -“ = T5

10) Connect the meter to the third pair of windings leaving the battery or power source connected to the first pair and repeat the procedure of touching the positive battery or power supply terminal to the second lead of the first winding. As before we are looking for a forwards deflection on disconnection. Reverse the meter leads if need be.

11) Mark the third pair of leads “Meter +” = T3, “Meter -“ = T6.

12) For Delta, 220 V, connection tie T1 to T6, T2 to T4 and T3 to T5. Input power goes to the three joins between windings.

13) For Y, 440 V, connection tie T4, T5 & T6 together as the star point. Input power goes to T1, T2 & T3.

—–

Obviously convert the temporary T numbers applied to the wire ends to proper U, V, W convention. It doesn't matter which way the coils are conneted so long as they are all the same way round.

Clive

[noel shelleyParticipant @noelshelley55608]

Thanks to Rob and Clive ! 2 Points I will reiterate, Be VERY careful when removing the end casings NOT to let the pry bar enter to far and damage the winding, or the motor will be scrap. Second, like Clive I would NOT want to use cable ties instead of lacing to resecure the windings, they are to hard and run the risk of damaging the varnish insulation. The varnish serves 2 purposes. when wound the wire has a coating to insulate each turn, the whole winding is then dipped or coated as a second insulating layer BUT also as a mechanical fixing of the wires so that vibration does not cause chaffing that will remove the varnish and result in a short circuit – and ultimate failure ! Noel.

[Ro80RobParticipant @ro80rob]

Clive and Noel, thanks for your comments

Clive, your method to easily identify which way round the windings are with a meter and torch battery is simple and clever. Fortunately in this case, I knew which end the windings were U1,V1 and W1 as they were connected originally so the others must have been at the star point.

There's another YouTube video I watched of a guy doing a demo in a pseudo class room environment. He had the 6 wires there on the motor and just identified which wires were pairs and just connected them up to the terminal board on the motor. It was most confusing as he never went through any procedeure to figure them out. I'm glad that you confirmed that it doesn't matter which winding is which. Just which way round is important.

WRT the cable ties, I'm happy with them. The ones I got are not hard nylon, but some other softer material. Also being only 3.5 mm wide, they are stretchy too. If I'd used wider ones, they'd have been too hard/strong and not given the effect I wanted.

The lacing cord I had was flat but not elastic and the original stuff I cut off was the same. A challenge to get it onto the windings as the loops of the windings to pass through would require long nosed pliers to poke the material through. Keeping it tight enough to secure the windings would have proved difficult for a novice cable lacer like myself, the original technique seems to have been to just loop over and over, with only one start and one end knot. I would have probably never got the job anywhere near as secure as I did and would have concluded I'd have to paint/spray some varnish in that case.

Finally, the only part of the varnishing affected was holding the sleeved joins on the tops of the coils, and even that only in small overall areas. The other 95% of the varnished coils are still as the factory did them. All varnished up. I could I guess have found something to paint onto the tops of the coils to provide some additional retention after I'd clipped them back into their same places, but I was happy with the job I'd done and it all seemed secure to me.

So, I'm confident that with what I did, and the useage I'll give this drill, the motor will outlast me, and if it doesn't I can always buy another modern motor and fit that.

I've just got to get round to getting the motor back onto the drill along with the VFD and controls and safety stop, then I've got to figure out how I'm going to get the heavy thing on the bench top safely.

Cheers… Rob.

[john fletcher 1Participant @johnfletcher1]

Good for you Rob, you must feel very pleased its been a success, and also, you have shown others how to do it. It did make me cringe when you said you laid the windings down on bubble wrap, there just might been one wire sticking out a bit and you might trap it and formed a short. I felt like rapping your knuckles and saying don't do that again. Years ago, for that very purpose I made an open top type box, using 4 pieces of thick ply wood about 6" wide X 15 long, each has a slot along its length making it adjustable, for various size stators, the windings dangle down wards. The whole thing is held together with 4 screws and wing nuts making it quick to assemble and useful. John

[Mike PooleParticipant @mikepoole82104]

Providing the three windings are all the same polarity the order of connection will only result in the motor running forwards or backwards. Companies who make motors ensure that they run in a standard direction so that quick change plugs can be used and the motor direction guaranteed. The star delta bridging is also included in the plug top to save installing links in the motor terminal box. Obviously in a home workshop it is easy to swap the direction if necessary and fitting links for star or delta will usually be a one off job. When you aim to build a car every 60 seconds the few minutes saved on a motor change if the failed unit can be unplugged and the replacement can be just plugged in can save thousands of pounds.

Mike

[noel shelleyParticipant @noelshelley55608]

I have a little box of tricks to check phase/motor rotation. Makes life a bit easier. Noel.

[Author]

Posts