• Spring 2024 meetup in Calgary - date Saturday, April 20/2024. discussion Please RSVP Here to confirm and get your invitation and the location details. RSVP NOW so organizers can plan to get sufficient food etc. It's Tomorrow Saturday! you can still RSVP until I stop checking my phone tomorrow More info and agenda
  • We are having email/registration problems again. Diagnosis is underway. New users sorry if you are having trouble getting registered. We are exploring different options to get registered. Contact the forum via another member or on facebook if you're stuck. Update -> we think it is fixed. Let us know if not.
  • Spring meet up in Ontario, April 6/2024. NEW LOCATION See Post #31 Discussion AND THE NEW LOCATION

G3616 Conversion.

jcdammeyer

John
Premium Member
Oh and it won't be a permanent solution. The horizontal shaft turns a bevelled pulley which turns the lead screw.
1689895413315.png
I modified the original to have a bearing in a holder and tweak the mesh. Can't see it really well in this picture.

1689895452674.png

In the long run the ball screw upgrade will fix the ball screw shaft and turn the ball nut. The motor will sit vertically under the knee and drive the ball nut pulley. After that the backlash will only be the ball screw and I'll likely want to add some sort of brake to it to prevent it dropping when power is off.
1689895609304.png
 

jcdammeyer

John
Premium Member
Using a 6mm end mill and 1mm depth of cut it shouldn't take too long. Probably longer to turn down the 2.25" blank to 25mm OD to fit inside the pulley. Then set it up on the rotary table so the square hole will be concentric with the outside.


1689901733075.png

The other approach is to first cut the square hole to make sure it fits correctly. Then mill a piece of steel to have the same size square shank to hold in the lathe and then turn the outside down.
 

whydontu

I Tried, It Broke
Premium Member
If it’s any use, 14mm square socket is a very common dimension for butterfly valve shafts. The standard is ISO-5211 that most valve manufacturers adhere. It’s very common to use a large round keyed bushing into the valve gear operator, broached to 14mm square. If you can wait until next Wednesday, I can see if my old employer has a bushing on the shelf.


I make adapters all the time, always putting ABC-brand valves onto XYZ-brand actuators.

 

jcdammeyer

John
Premium Member
If it’s any use, 14mm square socket is a very common dimension for butterfly valve shafts. The standard is ISO-5211 that most valve manufacturers adhere. It’s very common to use a large round keyed bushing into the valve gear operator, broached to 14mm square. If you can wait until next Wednesday, I can see if my old employer has a bushing on the shelf.


I make adapters all the time, always putting ABC-brand valves onto XYZ-brand actuators.

Thanks for the info. As always doing this stuff is a learning experience. For example I'm finding that it's likely backlash on the X is still not compensated quite right. At least at the position of the vise on the table. In the Y direction I get exactly 14.16mm as per the drawing. I'm 0.24mm short on the X axis which is about 0.009".

Also the 6mm tool bit is really to large. I'll compensate by placing a slight 6mm arc on each corner and make the X direction .24mm longer. It's a scrap piece of metal anyway.

1689906239972.png

The scariest part of doing all this is after the knee brings the tool setter up to touch the tool, measures it and then goes down, on goes the spindle and then at full 150 ipm over to the work. I've specified 10mm clearance which is more than enough but it's definitely butt clenching.

Just an air blast for cooling.
 

slow-poke

Ultra Member
Very impressive build requiring a range of skills.

I would like to increase the torque of my lathe at low speeds presently 1HP DC. was thinking 3-ph and VFD but will now consider AC servo.
 

Susquatch

Ultra Member
Administrator
Moderator
Premium Member
I would like to increase the torque of my lathe at low speeds presently 1HP DC. was thinking 3-ph and VFD but will now consider AC servo.

I'd still be going 3ph VFD if I were you but only because I can't really relate to the AC Servo requirements and performance. I know the 3ph VFD would yield lots of low speed torque, but I don't have any first hand experience with what the Servo will do.

I'm personally staying on track to do a 3ph VFD conversion on my lathe with the same objective as yours - low speed torque. My lowest speed is currently 70 rpm. With the 20:1 turndown ratio of the 3ph motor I bought, I should be able to easily do 6 rpm.

But I like the fact that you want to go a different way. We can both be following along to see how our different plans work out. Then we can compare results when we are done and both learn something.
 

jcdammeyer

John
Premium Member
This 24 second video shows a simple G-Code program turning the spindle at 10 RPM so it aligns with the spindle encoder index.
The AC Servo can develop maximum torque when stopped although it's not locked like a stepper motor. But what I have found very handy while playing with my probe, installing and removing tools, is that for test purposes the spindle stays locked and I only need the one open end wrench.
Mine is still V-Belt driven so I can use a wrench and make the belt slip but I have to really reef on the draw bar nut to do that.
I like that the AC servos are very quiet. No hum and 0 to 3000 RPM with no belt change. Using it with step/dir interface although the drive can also accept 0-10V instead.
 

slow-poke

Ultra Member
John, if I go ac servo what should I expect to pay for a suitable motor and drive?

Sounds like the Bergerda motor drives are a good bet?

Perhaps something around 1kW might be suitable to upgrade the 1HP dc motor.
 
Last edited:

jcdammeyer

John
Premium Member
John, if I go ac servo what should I expect to pay for a suitable motor and drive?

Sounds like the Bergerda motor drives are a good bet?

Perhaps something around 1kW might be suitable to upgrade the 1HP dc motor.
As I recall 1 HP is about 755W. So you have several options depending on physical size.
110SM-M0420MAL is 800W with 0 to 3000 RPM.
80SM-M0230MAL is 730W with 0 to 3000 RPM

I'm running the 110SM-M0630MAL on my spindle and I settled for the the 730W 80SM-M0320MAL for the knee since 2000 RPM was more than enough. As it is with 4:1 I get 150 IPM which is scary fast.

I've asked Donald Chen at Bergerda for some prices.

When I went looking for 3 phase and my mill this is the sort of pricing I found.
https://nexusgrinders.ca/product/1hp-3-phase-motor/ $380
Same company for VFD
https://nexusgrinders.ca/product/1hp-3-phase-motor/ $450

Upside of those is they appear to run 110VAC input. The Bergerda is 220VAC although that can be fixed with an autotransformer to create 220VAC from 110VAC. But that adds to the price if 110VAC is a critical value.

The 3 phase and VFD prices I think are on the high side but when I went looking for 2HP the Bergerda were cheaper, especially when I added motors to the order for the XY axis
 

slow-poke

Ultra Member
Upside of those is they appear to run 110VAC input. The Bergerda is 220VAC.
I have 240V adjacent to the lathe, so as long as they can tolerate the extra 20V, (table in link does not specify max voltage or tolerance). I would be inclined to use that. Are you powering them with 240V?

I do like that you don't need a separate power supply as required with a stepper drive. Just one less thing to get and mount, simple is good. For my CNC mill (steppers), I just used an old transformer from my stash, added a big bridge and cap and called it done, no issues after many years.
 

Susquatch

Ultra Member
Administrator
Moderator
Premium Member
I have 240V adjacent to the lathe, so as long as they can tolerate the extra 20V, (table in link does not specify max voltage or tolerance). I would be inclined to use that. Are you powering them with 240V?

This whole 240 and 230 and 220 thing is a red herring.

For the most part they are the result of the measuring method and slow increases in voltage by the supply grid to reduce line losses - not a deliberate design voltage difference.

Line voltages also vary over short periods of time as a result of changes in demand or supply so it isn't unusual to see variations like that on any given system even within a given day.

At any rate, it's not something to fuss over with normal system variations. 220/230/240 are all basically the same.

The same goes for 110/115/120.

The only time it's really important is when you compare utility voltages from other parts of the world. With different distribution systems. But in North America, just forget about it.

There is ONE very rare exception usually only seen in industrial settings. When you see systems designed for what is commonly called 208, you need to be more careful. 208 Volt systems are those designed to use two legs of a three phase supply system. Because the three phases of a three phase system are 120 degrees apart, the resulting voltage doesn't get as high as it does when only one phase is reduced by a transformer with a neutral center tap.

Bottom line - don't worry about 220 230 & 240. They are the same.
 

jcdammeyer

John
Premium Member
I have 240V adjacent to the lathe, so as long as they can tolerate the extra 20V, (table in link does not specify max voltage or tolerance). I would be inclined to use that. Are you powering them with 240V?

I do like that you don't need a separate power supply as required with a stepper drive. Just one less thing to get and mount, simple is good. For my CNC mill (steppers), I just used an old transformer from my stash, added a big bridge and cap and called it done, no issues after many years.
As @Susquatch said. It's not a big deal. I have 220-240VAC running into the house. From the breaker panel I have 110-120VAC circuits and 220-240VAC circuits. The mill runs on 220VAC because that's what the House Of Tools Nameplate said I needed. When I converted over to steppers and DC servos I built a power supply (Toroid transformer) with caps etc. for the 105VDC for the DC Servos.

The stepper for the knee ran from a different power supply. Now I only need about 105VDC for the Harmonic Drive 4th Axis driven by a STMBL driver. The other 3 axis and spindle are all plugged into the 220VAC circuit on the ESTOP controlled side of the power relay.

The AC Servos (and the STMBL) and even VFDs all use the same approach. They take a high voltage DC and create 3 phases and use the encoder feedback and step/dir input to adjust current through the windings to move the motor.

The difference is the VFDs don't use encoders but instead measure the voltage and current both during transistor ON times and OFF times to determine where the armature is. There are various really clever ways of doing this.

So the AC Servos like the Bergerda take (220VAC * 1.414)-DiodeDrop ~= 310VDC and use that filtered value as input to the three phase drivers that are pulse width modulated to create the appropriate voltage and current across each winding.

That's the simple explanation.
 

slow-poke

Ultra Member
Thanks for your input regarding 220V vs. 240V. I actually worked as a power system engineer for two utilities and in my opinion most of the time you can ignore the difference but certainly not all the time. Agreed mains voltage does move around depending on a variety of factors. I also agree that in Canada the nominal voltage has crept up over the years, however at this point there is a 20V (nominal) difference between China and Canada, and a 40V (nominal) difference between Canada and Japan. It's not at all uncommon for good quality older Japanese equipment rated for 100/200V to overheat or fail when exposed to 120 or 240V. Nothing wrong with the equipment it just was not designed to be run with 20% higher voltage continuously.

Very common nowadays for equipment to have a "universal" SMPS up front and agreed most of the time that SMPS is designed to be just that "universal" so it is designed to work over a wide range of voltages, typically 90-240V +/-6%.

Unfortunately there is equipment out there that was designed to the 220V nominal voltage specification that is iffy when run at 240V. Normally a 400V cap would be used for the bulk supply and that would be okay (240V * 1.44) + 10% = 380V, however with all the corner cutting nowadays, I have seen 350V capacitors used for nominal 220V equipment. Will a 350V cap work when exposed to nominal 240V mains?

Answer yes for a while, that while may be a short while if your nominal voltage is on the higher side of 240V, or if there are a lot of transients for example from the VFD that is close by or if the capacitors are budget based like Leon vs. a quality cap like Nichicon.

We were discussing servo drivers and considering the industrial environment that these typically operate in (more transients) one would hope that not too many corners have been cut, but you never know especially with budget priced equipment designed for the nominal 220V China power system.
 
Last edited:

Susquatch

Ultra Member
Administrator
Moderator
Premium Member
We were discussing servo drivers and considering the industrial environment that these typically operate in (more transients) one would hope that not too many corners have been cut, but you never know especially with budget priced equipment designed for the nominal 220V China power system.

I did mention off shore differences. But I deliberately decided not to go into too much depth.

My experience is that for devices where it matters, it's hard to get off shore hardware. So no biggie to users. The stuff we do mostly buy offshore isn't sensitive to the difference, so again it doesn't matter. As you point out, most servos are inherently robust so again no biggie.

Not many of us are gunna be checking or fussing about what brand of capacitor is in our stuff.

I used to fuss over it, but time and experience has taught me to just ignore it. Hence my advice to others. There are a million other issues that are much hungrier that wanna bite me much more seriously than this one!

My own experience with utilities predates wind and solar. Every time I look at one of those energy farms, I wanna puke. I can only imagine what they do to the grid! You are much younger than me. So I'm curious about your experience with the impact of these systems on the grid?
 
Last edited:

Johnwa

Ultra Member
There is ONE very rare exception usually only seen in industrial settings. When you see systems designed for what is commonly called 208, you need to be more careful. 208 Volt systems are those designed to use two legs of a three phase supply system. Because the three phases of a three phase system are 120 degrees apart, the resulting voltage doesn't get as high as it does when only one phase is reduced by a transformer with a neutral center tap.
At least here in Alberta 120/208 uses all 3 phases. It is 120V on each leg with the center grounded. 240 open delta does use 2 phases and line to line voltage is 240. There is no center ground. One of the line to line supplies is grounded in the middle to give 120/240 single phase which is mostly used for 120v receptacles. Its mostly used for small commercial buildings.
 

Susquatch

Ultra Member
Administrator
Moderator
Premium Member
At least here in Alberta 120/208 uses all 3 phases. It is 120V on each leg with the center grounded. 240 open delta does use 2 phases and line to line voltage is 240. There is no center ground. One of the line to line supplies is grounded in the middle to give 120/240 single phase which is mostly used for 120v receptacles. Its mostly used for small commercial buildings.

We are using a few different terms and a few minor different explanations. But I believe we are talking about the same wierd animal.

The nuance for the 208 beast is that two of the three phases are used which results in a 120 degree phase difference which drives a lower voltage than one full phase through a transformer (240 with a neutral tap) does.

As you note, I've never seen this beast anyplace but in commercial/industrial buildings.
 

slow-poke

Ultra Member
My own experience with utilities predates wind and solar. Every time I look at one of those energy farms, I wanna puke. I can only imagine what they do to the grid! You are much younger than me. So I'm curious about your experience with the impact of these systems on the grid?
That is a great question. Hopefully all my blabbering won't be something you already know.......As often the answer is complicated. I imagine you already know that in Ontario we are just a small part of a very large interconnected grid of distributed generation and loads. All of the ac generators in this grid are synchronized with only small phase angle differences between say a generator in Niagara Falls and one in say New York. The flow of power across the grid is controlled by excitation of the individual generators. The protection schemes for the high voltage transmission lines that transfer power across the grid are quite sophisticated and have to be to maintain stability of such a large control system.

Maintaining grid stability when a transmission lines delivering 1000MW to the grid suddenly (1-2 cycles) has to be isolated because lightning strikes the line is tricky, now imagine two or more occurring in a portion of the grid at near the same time!

One of many bad situations is a power-swing condition where large amounts of power start flowing back and forth in an uncontrolled fashion driven by the massive flywheel effect of the grid along the remaining uninterrupted lines. The entire grid can start to oscillate. This occurred in 2003. https://en.wikipedia.org/wiki/Northeast_blackout_of_2003

One thing that makes maintaining stability tricky is that the generation portion of the grid tends to be in big chunks that are physically separated creating the opportunity for the described power swings to occur. So having a bunch of distributed wind or solar farms can really help with stability.

When the concept of NUGS (Non Utility Generation) being connected to the grid was in its infancy the general opinion of the protection engineers that designed and maintained the grid was " keep your crap off of our grid" it's sensitive to being disturbed and your useless little 1MW generator has the potential to bring down the entire grid because you 1MW guys have no clue about anything and your protection systems are next to useless.
The politicians overruled the protection engineers, so they did their best to tweak the protections to accommodate whatever the little generators would throw at the grid.

I worked extensively on many of those protection schemes both the really sophisticated ones for 500kV all the way down to some of the bottom of the bucket simplistic ones for the windmills you mentioned. I imagine we will have another big "event" that is started by one of these small generators, hopefully the more distributed nature of the grid will help enough to minimize how far the next big outage stretches.

I expect we will see a lot more wind farms because the wind generator price has dropped so much with scale, that nowadays it is often the lowest cost solution, cheaper than all other forms unless you happen to have a large untapped river where you need the generation, and it seems we're all out of those.

Overall I think they are the best solution available at this time. However I think that many have inadequate protection schemes and there is no excuse to not better regulate the protections being used on many of these generators.

On a side note as I mentioned that power flow is controlled by manipulating the excitation at each generator. The excitation systems have become sophisticated with built in PSS Power System Stabilizers. The concept is that the power system operators have to react in human time when trying to balance load and generation as the various automatic protection systems isolate portions of the grid in an effort to keep it from collapsing during an event, PSSs see these events and do their best to manipulate the excitation QUICKLY to counter the building instability. We did some pretty cool testing with 300MW generators where we introduced small step changes into the excitation feedback loop while observing with a strobe the massive generator first accelerate, then slow down with the classical ringing sinewave type response. We did this starting with really small steps and no PSS, then with the PSS and then slowly tweaked the steps larger and larger while tweaking the feedback control parameters. Really cool stuff with so much at stake.
 
Last edited:

Susquatch

Ultra Member
Administrator
Moderator
Premium Member
The entire grid can start to oscillate. This occurred in 2003. https://en.wikipedia.org/wiki/Northeast_blackout_of_2003

One of my two best friends was a member of the Ontario Hydro team tasked with assessing that blackout and making recommendations to prevent it from happening again.

He told me that the board wouldn't accept what they found or their recommendations because it flew against the political wind that favoured NUGs. The report ended up pretty much watered down to the point of being useless and swept under the carpet.

I shall ask him if he knows about the Wikipedia version and what he thinks of it.

At any rate, that isn't really what I was asking. My interest is much more mundane. What have all the wind turbines and solar farms done to daily max and min voltages and outage rates?
 
Top