What's new
What's new

Full tear down and Rebuild of a 10EE Round Dial

GrantGunderson

Cast Iron
Joined
Jun 4, 2021
Location
Bellingham, WA
I ordered a new Run capacitor from Newark as they are the only one that had one with the exact specs I was looking for in a continuous duty. Its 300uF and 440V. Only issue is that Newark's shipping division is completely incompetent and they shipped it in just a padded envelope instead of boxing it.
image.jpeg
Notice the terminals are bent! After dealing with their horrible customer service for over a week, they finally agreed to refund my card and I ordered a replacement. That one showed up a week latter, with the same packaging issues and the same bent terminals. I contacted them again, and still waiting on a response another week latter... I wont be doing business with them. They could at least use customer service reps based on this continent.

I'm hoping they will send a replacement in a box, third time is the charm, but we will see. I did test the last one after I bent the terminals back and it does test good.
image.jpeg
While waiting for Newark to resolve the issue, I decided to clean up the motor's name plate.
image.jpeg
I used a black Lacquer stick to fill in the engravings,
image.jpeg
The excess gets ripped off, then I hit it with a couple coats of clear coat. This then gets installed with drive screws.
image.jpeg
Turns out the motor uses a #5 drive screw, which is the only size I dont have on hand and is the only size McMaster doesn't stock... odd. I then drilled the motors mount holes with a #29 drill and installed the name plate using a #6 drive screw.
image.jpeg
The name plate came out ok, the engraving wasn't that deep, so really hard to get the lacquer to stick in it.

Moving along, I didnt like how I had setup the terminal box previously on the M/G, so I decided redo it with a dedicated terminal strip for the motor leads... this will allow me to test various configurations, more on that latter.
image.jpeg
Previous box, with the new one on the right.
First step is to use my transfer punches to locate the screw mounting holes for the box.
image.jpeg
Those made a quick job of accurately locating the screw holes. For small holes in sheep metal, I find a step drill works best. I use a sharpie to mark it so I dont drill too large of a hole.
image.jpeg
For larger holes I like to use my knockout punches.
image.jpeg
I have a Klien set that serves me well for this.
 

GrantGunderson

Cast Iron
Joined
Jun 4, 2021
Location
Bellingham, WA
The knockouts use a set of dies for each size that require you to first drill a smaller pilot hole.
image.jpeg
You insert them into the pilot hole and tighten with a wrench.
image.jpeg
This results in a much cleaner hole, that unlike a hole saw hole, does not need to be deburred.
image.jpeg
The chips are also a hell of a lot easier to clean up too.
image.jpeg
Thats a big plus if you are adding conduit to a box already in service as it keeps the box clean from metal chips.
Here is the new box installed and wired up with the motor leads.
image.jpeg
Thats way cleaner and more organized than my previous attempt. I'm glad I re-did it!
Next I tested each pair of windings with my Fluke meter
image.jpeg
I got the following results for each pair of motor windings:
1~4 = 0.9Ω
2~5 = 0.8Ω
3~6 = 0.9Ω
7~10 = 0.8Ω
8~11 = 0.9Ω
9~12 = 0.8Ω
Ok, resistance across each individual coil looks great!

Next, I tested motors insulation by testing each lead to ground with my Klein Megger
image.jpeg
I started off testing at 250V and got >4000MΩ for each lead to ground, I then redid the test at both 500V and again at 1KV and got the same >4000MΩ reading for each. Finally I tested between each motor Terminal and still got >4000MΩ with the exception of the coil pairs. So the motor is sound electrically. Not bad for a garage motor rebuild.
 

GrantGunderson

Cast Iron
Joined
Jun 4, 2021
Location
Bellingham, WA
Cal Haines was kind enough to draw up the various wire diagrams for connection the M/G motor to AC power. I hope he doesn't mind me sharing these here.
Here is the standard 3 Phase High-Voltage wire Diagram
image.png
Here is the standard 3Phase Low Voltage diagram
image.png
Notice, the High Voltage Diagram have the motor's coils in SERIES with 4 coils between any of the Terminals. The Low Voltage diagram has them in PARALLEL .

Now looking at the Steelman connection diagram for converting the motor from 3Phase to single:
image.png
The T1 and T2 circuits are also in PARALLEL, however the T3 circuit is in SERIES. This is interesting. It brings up the question, can you use a start / run capacitor setup with the standard Low-Voltage connection to get a 2 Phase motor?
image.png
Both the Steelman and the Low-Voltage diagrams share the following jumpers
1~7
2~8
4~5
10~11

Noticing this, I then wired those jumpers and placed them behind the Generator wires as these connections will be used regardless, and they are all part of the original factory configuration.
image.jpeg
I now have a good test bed to do some testing and see if it is possible to use the Low-voltage connection with a run capacitor to run on single phase, and see how it compares to the Steelman method of converting to single phase.

Testing any two coils in Series gives me 1.6Ω. This is good, as they all tested the same. This makes sense as testing each of the individual coils gave me readings between 0.8 and 0.9Ω so 0.8Ω plus 0.8Ω = 1.6Ω. Along those same lines, if we then wire two of the series of coils in parallel we would take the 1.6Ω divide it by 2 and we end back up with 0.8Ω same as what each individual coil tested at. The Electrical Theories that I "learned" back in the few EE classes I took in college are quite foggy, but this is at least making a bit of sense to me.

Ok, lets wire it up for the standard 3Phase Low Voltage connection and see what we get.
image.jpeg
For this configuration we add the following jumpers
3~9
5~6
11~12
The motor is now in the 3 Phase low-voltage connection identical to the name plate.
Lets see what we get for the following connections
T1 ~ T2 = 0.9Ω
T1~T3 = 0.9Ω
T2~T3 = 0.9Ω
Ok this is what we would expect based upon the calculations above.

Now, lets remove the 3 sets of jumpers for the Low-Voltage connection and then wire it for the Steelman conversion:
image.jpeg
For this we add only two sets of jumpers
1~12
9~6

Now lets test our Terminal combinations again
T1 ~ T2 = 0.9Ω. Same as the standard low voltage connection above, since it is the same connections.
T1 ~ T3 = 1.6Ω Same as any two coils in series
T2 ~ T3 = 2.3Ω Interesting. Ok, looking back at when I tested the motor previously with the Steelman conversion, before the start cap blew, I was getting a voltage reading across T2 ~T3 of 400V. Interesting.. .thats high voltage territory. Humm.

Lets move the jumpers around and see what we get with the motor in the High-Voltage setting. In this setting, each combination of T1~T2, T1~T3 and T2~T3 is identical, in theory I can just test one of these and see what we get for resistance.
image.jpeg
Turns out it is 2.3Ω! That is exactly the same as I got with the T2~T3 connection in the Steelman configuration.

So this is making a bit more sense, as to why I got different voltages in my previous testing.

So in summary with the Steelman configuration

238.5V T1 ~ T2 = 0.9Ω. Same as the standard low voltage connection above, since it is the same connections.
296.8V T1 ~ T3 = 1.6Ω (two coils in series)
399.9V T2 ~ T3 = 2.3Ω

Since T2 ~T3 also has the start / run caps connected across them, I believe the higher voltages here make sense...(If I am thinking correctly, the capacitor is what provides the phase shift so the motor can start, and because of the phase shift, the meter is reading it as high voltage.

Next step is to add the start / run capacitor to the system and see where we are at. I really wish I had an oscilloscope for the next round of testing, just to better understand what exactly is going on.
 

thermite

Diamond
I really wish I had an oscilloscope for the next round of testing, just to better understand what exactly is going on.
Haven't been WITHOUT anywhere from one to four 'scopes since 1954. And refuse to be without.

All that happens here on a 10EE is within the "Audio" pass-band. The least-capable 'scopes ever built cover that. No need for 50 MHz or 100 MHz bandwidth.

So would the "music" apps built into phones and laptops cover the 25 KHz and below 'audio' band - the spectrum analyzer & waveform graphical display.

The PROBLEM is... that a "MG" 10EE can easily generate steady-state DC up to around 300 Volts, and braking & contactor switching spikes to around 1500-plus Volts. Modular can go higher - even if only by ACCIDENT.
Solid State DC Drives I mess-with... far higher again.

Cheap scopes and leads might stand the 'basic' DC power, but are typically rated for only 600 or even only 300 Volts.

Cheap meters have the same limitations, BTW.

IIRC, my four-trace "Asian made" Rigol digital 'scope was under $400.

Four sets of US made 'multiplier' leads for safer high Voltage measurements? Closer to $1,200 !!

Four-trace is handy. VERY! So is the ability to store and play-back or transfer a captured file.

I can watch DC output and AC input side all at once, same screen. Or three Phases of AC off an RPC, plus the DC out.. 'etc.'

But if you are totally "blind" with NO scope?

Even a single-trace unit - looking at one variable at a time, then move the leads and take notes....is golden.

Just mind those Voltage hazards!
 
Last edited:

GrantGunderson

Cast Iron
Joined
Jun 4, 2021
Location
Bellingham, WA
Haven't been WITHOUT anywhere from one to four 'scopes since 1954. And refuse to be without.
I’ve been trying to find a suitable and reasonably priced scope. Seems there are a ton out there in the $100-200 problem is finding one capable of the voltage range found in the 10Ee and then the probes are stupid expensive. Hell even the fluke probe set for my amp meters was $200.

Open to suggestions on a scope and lead combo.
 

thermite

Diamond
I’ve been trying to find a suitable and reasonably priced scope. Seems there are a ton out there in the $100-200 problem is finding one capable of the voltage range found in the 10Ee and then the probes are stupid expensive. Hell even the fluke probe set for my amp meters was $200.

Open to suggestions on a scope and lead combo.
I have an older Fluke dual-trace "scopemeter", and this newer one that is light and handy... and from these folks:


Check for promotions and clearances and if "membership" can save you a few bucks on shipping or a discount.

Several PM members have these or similar, or other brands they prefer. There is a TON of discussion online about 'scopes. "(E)-magazine" reviews a-plenty as well.

Get AT LEAST TWO sets of High-Voltage "multiplier" leads.

"Multiplier" means "attentuator", actually, as the scope itself must be protected. A 10:1 will attentuate a 1,500 V spike to 150 V at the 'scope's terminals, etc. I generally use 100:1 for this work,

Used 1000:1 with clumsy but most-welcome shields back when doing TV and O'Scope Cathode Ray drive circuit work @ 30 Kilo Volt, "hollow state" days.

Read-up:


Go ogle will find what TE doesn't have. Ex:


You will probably be money-ahead to buy a probe "kit" in a pouch that covers many needs, 'clip-on' included.

Back when I bought my first Fluke, I needed TWO such for full coverage, as I might 'parachute' in to get a switch off its knees somewhere in Asia ...and have to look at high power as well as low-level signaling.

My Fluke leads also fit the Rigol, but are 25 years old, plus I didn't want to be cross-robbing them, so bought the Rigol four NEW "multiplier" sets of its own.

"YMMV"
 
Last edited:

GrantGunderson

Cast Iron
Joined
Jun 4, 2021
Location
Bellingham, WA
Ok, thinking about this some more,

After measuring resistance in each configuration, I dont think the Low-voltage with the caps across T2T3 is going to work.
image.png
In the standard, Low-Voltage configuration we get:
The motor is now in the 3 Phase low-voltage connection identical to the name plate.
Lets see what we get for the following connections
T1 ~ T2 = 0.9Ω
T1~T3 = 0.9Ω
T2~T3 = 0.9Ω
Thats with each terminal set seeing 220/240V across it.

With the Steelman configuration
image.png


T1 ~ T2 = 0.9Ω. Same as the standard low voltage connection above, since it is the same connections.
T1 ~ T3 = 1.6Ω Same as any two coils in series
T2 ~ T3 = 2.3Ω Interesting. Ok, looking back at when I tested the motor previously with the Steelman conversion, before the start cap blew, I was getting a voltage reading across T2 ~T3 of 400V. Interesting.. .thats high voltage territory. Humm.

in the High-Voltage setting it is 2.3Ω between any two combinations of T1,2,3

So based upon the voltages I got from my attempt prior to the start cap blowing in my previous tests, I got
So in summary with the Steelman configuration it is wired to handle the high voltage across T2~T3

238.5V T1 ~ T2 = 0.9Ω. Same as the standard low voltage connection above, since it is the same connections.
296.8V T1 ~ T3 = 1.6Ω (two coils in series)
399.9V T2 ~ T3 = 2.3Ω Same as the standard 3 Phase HV connection.

The low-Voltage option with the run capacitor would not be, if I’m not mistaken.

Going back to the first thing in the Steelman manual: "it is important to remember that T1 and T2 two lines will be carrying more amperage than the nameplate of a three phase motor will indicate. This is true because it will be carrying the same total power on two lines that it would be carrying on three lines when operating on three phase. The current required from single phase lines times 1.73 delivers the same power as three phase provided that the system efficiency and power factors are the same.”

Thoughts?
 

Mr_CNC_guy

Aluminum
Joined
Jul 29, 2018
Location
New England
The thing about static and rotary phase converters (basically the same thing) is
that they are trying to generate the 3rd phase by using capacitors. This scheme
requires a balancing of the phase angle and voltage by of the 3rd phase by
varying the value of the capacitors. Unfortunately, a well-balanced converter
is only balanced at a certain load. This requires a trade off.

I built a rotary phase converter and did a lot of tuning. In the end I got it
pretty well-balanced but only with my lathe running but not cutting anything.
When I would cut something, the balance would shift. This shift and the
corresponding loss of horsepower is the price you pay for this method of
generating 3 phase.

Another thing, you don't mention a start mechanism. That generally
requires another capacitor and a current sensor or timer.
 

thermite

Diamond
Ok, thinking about this some more,

............

The low-Voltage option with the run capacitor would not be, if I’m not mistaken.
I'm not as "up to date" as I should be, but with the required re-connection to GET to Steelman, I thought one LOST the dual-Voltage capability.

I could be wrong. I've been wrong before.

Going back to the first thing in the Steelman manual: "it is important to remember that T1 and T2 two lines will be carrying more amperage than the nameplate of a three phase motor will indicate. This is true because it will be carrying the same total power on two lines that it would be carrying on three lines when operating on three phase. The current required from single phase lines times 1.73 delivers the same power as three phase provided that the system efficiency and power factors are the same.”

Thoughts?

"Anecdotal evidence".. but from memory...

Peter Haas once posted that a Steelman-Haas conversion could actualy deliver MORE than the converted motor's OEM nameplate power. ISTR 125%?

Contrast this with an RPC limit of around 91%, a "static" limit of 31% (smoothly...) A skosh more can be had, but at progressively "raggedier" running 'til it falls on its nose - hard - at around 60%. Statics are also hard on motors. VERY!

..and VFD ability to also deliver more than 100% of OEM .... at least at 'some' speed-ranges.
VFD // "inverters" as a class are reasonably efficient, but far less so than 100%. A Phase-Perfect, for example, is only bareley more efficient than an RPC, 'if even'. Note that P-P and VFD both have cooling fans to shed the heat from conversion losses.

Thyristor-class DC Drive conversion efficiency, BTW, is enough higher that they do NOT ordinarily need - or even HAVE - cooling fans.

DOWNSIDE.. was that the Steelman-Haas ALSO demanded more net INPUT POWER than that percentage gain.

IOW, even at but 70 % of nameplate, where 3-Phase motors are usually designed to run when folk choose their sizes, it would draw more than 70% of what an ordinary 3-Phase motor would draw on a properly "balanced" 3-phase supply.

"TANSTAAFL" effect.

If it were truly "free"? 3-Phase motors could ship ex-factory already strappable for dual 3-P // 1-P use! But they do not.

So... yer on the right track.. but...

....the 'classical' 1.73 might be closer to 2.0 with effiiciency penalty?

Peter would have had the actual maths. It was within his career field, and he was good at it.
He had added the Power Factor correction to Steelman to improve on it.

I don't recall if he published the specific calculations though?

No fear. You are on the right track. "What will be, will be".
The careful and methodical approach as to documenting it all is contributing soilid value.

"Page Two":

AFAICS, trying for 300 MFD AND non-polar, high-ripple-capable, AC 'run" capacitor in a single, compact, package is economically challenging. "Start" caps yes. "Run" caps ...not so much.

Suggest a larger housing - remote if need be - with an array of, for example. multiple ~ 60 MFD @ 460 VAC units, paralleled.

Waaaay easier to source competitively, should be enough longer-lasting that life-cycle cost ends-up lower by a year .... ten.. or even or twenty years down the road.

Bill
 
Last edited:

Cal Haines

Diamond
Joined
Sep 19, 2002
Location
Tucson, AZ
Ok, thinking about this some more,

After measuring resistance in each configuration, I dont think the Low-voltage with the caps across T2T3 is going to work.
This connection:
image-png.368263

is the way the most rotary phase converters are connected. So yes, it will work. The diagrams show only the connection at run time, not during startup. The typical startup circuit simply connects a larger capacitor in parallel with the run capacitor long enough to get the motor turning.

Cal
 

GrantGunderson

Cast Iron
Joined
Jun 4, 2021
Location
Bellingham, WA
Ok, Time for a bit of a recap. When I purchased this 10EE, it Legitimately had a rats nest in the Motor / Generator.
image.jpeg
The Rodent had literally eaten the insulations on the wire leads for the motor / Generator.

So I completely replaced all of the wiring for the Motor / Generator, the Exciter and the DC motor and did a full rebuild on all of it in my home garage.
image.jpeg
Since I already had the 3Phase AC motor apart, I figured it only made since to split the internal star point apart, and bring those wire leads out, so instead of having a 9 lead motor (common) I know have a 12 lead motor. This put me in an interesting position, to do a Steelman method of converting the lathe from #phase to single phase, or allowed me to do a Low-voltgage conversion using a a Starting circuit. Turns out both methods use the same starting circuit.
image.jpeg
The main difference is the Steelman method specs out a 60uF run capacitor, and the Low Voltage method (common on a rotor phase convertor) uses a 100 uF run capacitor.
image.jpeg
The smaller 60uF run cap has a 8mm mounting thread, where as the larger 100uF cap uses 12mm. So I used an 8 to 12mm thread adapter, so I can easily interchange both caps, to test both methods.

To better facilitate testing both methods and taken measurements, I set up a third terminal strip and ran jumper wires.
image.jpeg
I then set up two multi meters, one is reading voltage across the coil of the relay that takes the start cap out of the run circuit. The second is measuring resistance across the relays's switched contacts.
image.jpeg
The idea is when the motor comes up to speed, the relay should drop the start cap out of the circuit and the reistance across the relay's switched contacts should read as an open circuit. Until it comes up to speed it should show some resistance.

It's now time to fire up the lathe for a test.

It F'n works! It fired right up on first try!
As you can see in the video the Relay's switch opens up very quickly as the motor comes up to full speed almost instantly. Note the 400+V across the relay's coil. This is proving to me the original Supco APR5 relay recommend for the Steelman method is undersized, as that relay is only rated for 250V.

To better evaluate the two methods of converting the 3Phase motor to single phase I purchased a Picoscope 2204A Oscilloscope. The nice thing about this unit is it works with both PC's and Mac's which is goos as I only run Apple computers and have for the last 20+ years. It also has some pretty sophisticated software, that I have a lot to learn about.
image.jpeg
Here is the test setup
image.jpeg
Cal Haines was kind enough to come up with a series of test perimeters for me to evaluate, consisting of 8 runs on the Steelman conversion and 11 run on the Low-Voltage singe phase conversion setup. Since the scope only has two Chanels, we are leaving the first challenge to take the T1 voltages, and then using the second Chanel to do the other measurements, this way we can sue the T1 channel as a reference to compare each. This will allow us to look as each set of of windings in the motor separately for both voltage and amperage measurements. Cal's help with this has been immense.

For example, here is an overlay of the voltages of the T1, T2 and T3 winding sets in the Steelman conversion. Hopefully I have the overlays properly aligned.
PastedGraphic-3.jpeg
PastedGraphic-2.jpeg
Notice how much stronger the T3 set is. I think this makes sense as its in series while the T1 and T2 sets in the configuration have the motor coils in parallel.

I haven't used a scope since I had a few EE classes in college and thats 20+ years ago at this point, so I kinda fell like I am learning it all again from scratch. So always open to any insights.

I was also going to setup a set of 3 thermal couples to take measurements of how hot the motor windings got, but I totally forgot. So I will have to implement that in the next series of testing.
 
Last edited:

Mr_CNC_guy

Aluminum
Joined
Jul 29, 2018
Location
New England
I am having some trouble understanding your measurements. In the first chart, it seems
that the red trace is way out of its proper phase position. It should be at 120 degrees
from the other two phases. This is something that you adjust by varying your capacitor
bank.

Do you have a link to the Steelman method that you mention?

One of the problems with running a very unbalanced 3 phase motor is rotor heating.
The unbalanced currents in the stator cause excessive currents in the rotor. You
should check this. This heating is not related to any heating due to normal loading
of the motor resulting from rotor slip.
 

thermite

Diamond
I was also going to setup a set of 3 thermal couples to take measurements of how hot the motor windings got, but I totally forgot. So I will have to implement that in the next series of testing.

Nice work!

'Hopefully'... the Steelman reconfiguration has fairly evenly distributed heating.

It "appears to.." function as if each phase-winding set had become one of three 1-Phase 'motors' each at 1/3 of the OEM 3-Phase power annnnd distributed 120 degrees apart.

Temperature measurements would be generally useful, but even MORE so if there is a sensor clearly associated with each set of windings.
 

GrantGunderson

Cast Iron
Joined
Jun 4, 2021
Location
Bellingham, WA
I am having some trouble understanding your measurements. In the first chart, it seems
that the red trace is way out of its proper phase position. It should be at 120 degrees
from the other two phases. This is something that you adjust by varying your capacitor
bank.

Do you have a link to the Steelman method that you mention?

One of the problems with running a very unbalanced 3 phase motor is rotor heating.
The unbalanced currents in the stator cause excessive currents in the rotor. You
should check this. This heating is not related to any heating due to normal loading
of the motor resulting from rotor slip.
At this point, I am learning as I go, and its proving to be a good cheap education (as long as I dont fry the motor). The motor does appear to be unbalanced, but I also think there is some issues with the data collection with the scope (user error most likely) as well as keeping the phases in perspective with their relationship between one and another in the over lay...

so next step is to verify the data from the scope with my Fluke meters, so I know what needs to get adjusted in the data collection. Then look at how it gets combined in the software.

Once I know I have good data, I can move on to the next step of trying to figure out how to balance it.

I've linked to this above in the thread, but here are the links again.

The Steelman method and literature.

Here is a link to it being applied to a 10EE here on PM.
Converting Monarch 10ee motor-generator 3-phase to single phase, Steelman method

And another one on PM
http://www.practicalmachinist.com/vb/monarch-lathes/single-phase-power-motor-generator-10ees-153348/.

Here is a schematic of the Steelman method drawn up by Cal Haines
1657986099968.png
And of the low-voltage single phase conversion
image.png
Notice the main difference is in the Steelman config, the T3 leg is in series with the other legs being in parralell.

Problem is both of those threads end without much documentation, other than others saying they got it to work. Since it fires right up, I think most others conclude everything is fine. However I am finding few issues with the previous threads on PM as the relay suggested in those, Supco APR5 is under spec for the coil voltage it actually sees, so I used a higher rated one. Also without looking at it with a scope, it would be really hard to know what is actually going on between the phases.
 

GrantGunderson

Cast Iron
Joined
Jun 4, 2021
Location
Bellingham, WA
I am having some trouble understanding your measurements. In the first chart, it seems
that the red trace is way out of its proper phase position. It should be at 120 degrees
from the other two phases.
3Phases_Firstset.jpg
I think this overlay is better, but the T3 voltage (that leg has the run cap on it and its in series instead of parallel) is way too high compared to the rest. I need to varsity that voltage is in fact that high, and if so, I need to find out why and figure out how to reduce it.
 

thermite

Diamond
View attachment 369083
I think this overlay is better, but the T3 voltage (that leg has the run cap on it and its in series instead of parallel) is way too high compared to the rest. I need to varsity that voltage is in fact that high, and if so, I need to find out why and figure out how to reduce it.
We tend to address these challenges predominently - even exclusively - with application of lumped capacitance as the smaller, lighter, and arguably cheaper solutions.

Lumped Inductance also exists.

:)
 

thermite

Diamond
<Lots of great information deleted>

You got a "like" from Thermite!
Cut me a break, guys! I ain't stingey!

Software thing.

The Old VB website had some sort of off-site active tracking link hard-coded to the "like" button!

My OpenBSD plus 'privoxy' plus browser security features de-fanged the bugger expertly, but I was not able to figure out WHICH of 60 thousand-plus defenses to grant an "exception" to.

For MANY YEARS I have been UNABLE to grant a "like".

Email still worked. Lots of folk 'heard from me'.
That seems to have been cut-off?

Swings and round-abouts.
 

GrantGunderson

Cast Iron
Joined
Jun 4, 2021
Location
Bellingham, WA
We tend to address these challenges predominently - even exclusively - with application of lumped capacitance as the smaller, lighter, and arguably cheaper solutions.

Lumped Inductance also exists.

:)
Never heard of this... I need to do some research.
<Lots of great information deleted>

You got a "like" from Thermite! :)
Ha, I'll take it!

I spent a few hours today verifying the data I collected with the scope this weekend, using my multimeters. I attempted to use the Thermocouple set I bought off of Amazon to also get some temp data from the M/G casing during these runs, but that was futile as the unit's auto off time is way too short.

First I read the voltage across the Start cap. I am getting 511V there.
image.jpeg
Next, I used both of my Fluke Multi Meters and my Klein Megger to verify the rest of the data collected with the scope.
image.jpeg
T1 to ground read 115VAC
T2 to ground read 115 VAC
T3 to ground read 503VAC
T1 to T2 read 241 V
T2 to T3 read 404.1V
T1-T3 read 299.8V
T1 43 Amps on immediate startup that quickly dropped down to 9.3 Amps.
T2 0.1Amps
T3 0.1 Amps

Terminal set #12~#3 (T1~T3) has the two coils in series instead of parallel and its connected to the start / run caps and it read 512VAC RMS

Terminal Set #7~#8 (T1~T2) has the two coils in Parallel and it read 241 VAC RMS

Terminal set #1~#2 (T1~T2) has the two coils in Parallel and it read also read 241 VAC RMS

Ok, next I checked the individual coil voltages.

Terminal #12~#3 contains coils #12-#9 and #3-#6
they read 255VAC and 256.9VAC

Terminal Set #7~#8 contains coils #7-#10 and #8-#11
they read 133.6V and 144.2VAC

Terminal set #1~#2 contains coils #1-#4 and #2-#5
they read 133.9V and 144.5V

Next, I started to run another set of test runs with the Oscilloscope. This time I set the height to +/- 500V and the time base to 2 ms/div wit the goal of capturing 1 full cycle. The scope has an auto trigger function that I was just starting to get figured out by run 4, that will allow us to look athlete the waveforms upon startup, and then the scope captures 64 consecutive waveforms, so we can also see how it looks once it stabelizes.

Here is a combined look of T1 (Yellow) T2 (RED) T3 (Blue) and Node 45 (Green) and Node 1011 (brown)
image.png
1658216193981.png
Before I could finish this set of runs, I ran into some issues. Hitting the Stop button the lathe wouldn't turn off!!! I had to kill the power to it via the mains breaker. Looking at the Contractor, it was stuck closed! I popped it back open and tried another run. This time the lathe would turn off, but even when it was off, I was still showing voltages at the motor terminals! Humm. I go look at the main contactor
image.jpeg
And it was full stuck closed. Ok, this is no good and its a pretty major safety issue. Time to figure out what the hell is going on with it.

The front of the main contactor has a black bar that is held on by two screws.
image.jpeg
Ok, I think I have found the problem! Lets look at the contacts on the other side.
image.jpeg
image.jpeg
Ok, two of the 4 sets are completely worn out, and apparently have started to weld themselves shut!
image.jpeg
A third set is not far behind, and the 4th set, look like they just need a good cleaning.

The contactor is a Cutler Hammer 9586 and is a Size 1 contactor.

Searching Ebay for a contacts for a 9586 gave me a ton of options but none looked correct. I then searched for a Cutler Hammer contacts #1 and found that a Cutler Hammer replacement set # 6-106 appears to be correct, so I ordered two full sets, so fingers crossed when they arrive they are correct. I dont think there is much left of the electronics of the lathe that I haven't rebuilt now!

And that is officially 2 steps forward, one step back.
 
Last edited:

Mr_CNC_guy

Aluminum
Joined
Jul 29, 2018
Location
New England
I have been pondering the Steelman design and I think I understand what it is
trying to accomplish. In essence, you are trading off capacitance value with
the voltage rating of the capacitor. You are doubling the required capacitor
voltage rating requirement for only needing one quarter of the capacitance
value. This is pretty clever.

Obviously, the voltage measurement between T1 and T2 will be the line
voltage. The measurement between T1 and T3 should be double the line
voltage. The measurement between T2 and T3 should be the same.

The current in the T1 windings and the T2 windings should be the same.
The T3 winding current should be one half the other readings. Of course,
the phase angle should be 120 degrees.

In the real world this is impossible to achieve since any change in load will change
the required capacitance necessary for perfect balance.

My take on your measurements is that you need more capacitance, your T2 to T3
voltage is too high and your T1 to T3 is too low.
 








 
Top