New-to-me mystery lathe

[Ian Paterson]

Hi Everyone,

I picked up an unusual "front ways" lathe and quite a bit of accessories and tooling off usedvictoria last Monday. I've only just barely started to clean it up and investigate what I bought, so I'll post more updates later on as I learn more about it.

In addition to a bearing surface on the top of the bed, it also has dovetail ways on the front, which makes it sort of resemble a Rivett 8" Precision. According to lathes.co.uk, there were only a handful of manufacturers in the late 19th and early 20th centuries that used this design, but I haven't been able to match my lathe to any of those on the lathes.co.uk website.

It's got an 8" swing, with about 20-21" between centres, a threaded spindle with a #3 morse taper, #2 morse taper in a tailstock that cannot be offset for taper turning, and a hidden carriage drive shaft tucked up underneath the bed. The feet under the bed appear to be home-made, and I wouldn't be surprised if other parts of the lathe have also been modified by one or more of its previous owners. A number has been scratched into the paint on the headstock, but I don't know what it means. It came with quite a bit of tooling; HSS cutters, back plate, home-made high speed spindle adapter, tailstock chucks and even a Starrett surface gauge.

All in all, it's a pretty complete kit, but the limited free time I have means this will be a pretty slow restoration and investigation project. I also need to learn about the care and feeding of a lathe with plain bearings. I guess at some point I may try sending photos to Tony at lathes.co.uk to see if I can get help identifying it.

Ian

[Rory Brown]

Hi Ian, I had what looks like a more modern or perhaps shop adapted similar lathe.  The tailstock and the headstock casting look very similar.  Mine came with 2 chucks, one was a Craftsman and the other an Atlas.  The guy I got it from claimed that the lathe was an Atlas, but I am pretty sure that it was not.  There was no other branding on it.  I used that lathe for many years and actually just sold it.  Check out the picture, especially the tailstock and casting for the headstock

[Ian Paterson]

Hi Rory, thanks for sharing that photo. It's a really nice looking lathe that appears to have been solidly built, particularly the carriage assembly. The tailstock on yours looks slightly "bowling pin" shaped, whereas mine is more cylindrical. The overall heft and proportions of your headstock are similar to mine, but the cone pulley is reversed and it bolts to the bed from the top rather than underneath. I don't recognize it, but the reversed headstock cone pulley as well as any other unusual features might be helpful in identifying it

My lathe bears an uncanny resemblance to a Rivett 8" Precision however and I'm wondering if it might be a lesser-known model by the same company, or more likely, some kind of copycat model from another manufacturer. Check out the Rivett and you'll see what I mean: http://www.lathes.co.uk/rivettearly608/index.html

I've emailed Tony at lathes.co.uk to see if he knows anything about it.

Ian

[Rory Brown]

Wow!  Other than the extra screw cutting lead screw, it sure looks like the actual photos of the Rivett.  Nice find!

[Ian Paterson]

On the Rivett, you can see the leadscrew for threading, and a slotted drive shaft for carriage feed on the front of the lathe bed. The reason for this is so that the leadscrew isn't unnecessarily worn by regular carriage feeding. Also, the separate carriage feed shaft opens up the possibility for a powered cross feed for facing. On my lathe, I still have that carriage feed drive shaft, but it's hidden under the lathe bed and drives a worm gear in the carriage. You can see it in the photos of my first posting.

Ian

[Admin]

What a great find Ian! Could potentially be a very old lathe .. very cool :)

[Ian Paterson]

I'm thinking of replacing some of the worn gears in my carriage and I'm trying to determine their diametral pitch. I've tried various online formulas on several of the gears including the rack, and my results are consistently between 22.3 and 22.5. Only problem is that doesn't appear to be a standard diametral pitch (20 and 23 are the closest values that McMaster-Carr carries). Does anyone know if a different system was used for gears 100 years ago? Okay, I know that must sound funny, but I'm actually serious! ...Or could I be making an error in measurement/calculation?

On a different note, Tony at lathes.co.uk was not able to identify my lathe, but he did show quite a bit of interest in it. Right away, he prepared a draft web page for "unknown lathe #129". I still have to provide some more notes and photos though.

Ian

[Admin]

You made lathes.co.uk!!! What an achievement! :D

[Ian Paterson]

Yeah that was very cool. I really enjoyed chatting with Tony too. He was very observant about the photos and generous with his time as we speculated on the lathe's origin and design.

Ian

[Ian Paterson]

I've got my mystery lathe "#129" mostly cleaned up and back together. Aside from making a new shaft for one of the gears in the apron/carriage, all I've done is a bit of cleanup, some paint and lubrication. I've tested the motor and it seems to work fine, but I haven't tried to drive the spindle yet because I need to figure out what oil to use first. It uses plain bearings with a double taper at each end. I haven't tested the bearings with a magnet, but there appears to be some rust on out the ouside (non-bearing surfaces) which leads me to think they may be made of iron?? Anyhow, I've read a bit about avoiding oils that contain sulphur and detergents, as well as modern engine oils not being suitable because they're designed for pressurized, filtered oil systems that run at higher speeds. Besides that however, I'm not really sure how to proceed and was wondering if anyone in this group has some advice on what I should use and where I can get it.

Once I get the lathe running and make some test cuts, I have a feeling I'm going to want variable speed control and that leads me to my next question: Can I use my existing single phase motor with a VFD? All I know so far is that I've got a single phase, 120V, 1/2 horsepower motor that can be reversed with a switch that the previous owner wired in. When the motor spins up, I can hear the centrifugal switch that's used for starting, but I don't see a capacitor anywhere - perhaps it's internal? Anyhow, I'm wondering if a VFD can drive this with variable speed and in both directions, or do I have to get a 3 phase motor?

For those who are interested, I have a Google photo album of #129 here:

https://photos.app.goo.gl/vSziLtaPg3qXmv1L9

...I've added comments to several of the photos and I invite anyone to add their own.

Ian

[Admin]

Hey Ian, way to go with the restoration! Nice work .. I have a feeling she'll work a treat when she's making chips .. I just wanted to quickly answer your motor question and unfortunately you will be needing a 3ph motor to work with a VFD .. you'll feed the phases (termed single phase) of your household into the VFD (2x110V) and the VFD will make up the 3rd phase to run a 3ph motor .. by varying the frequency, you will be able to change the RPM of the motor .. our native frequency here in North America is 60Hz (50Hz in Europe) and your max RPM indicated on your motor plate would be achieved when setting 60Hz on the VFD .. The beauty of a VFD is that it can modulate the output frequency (to the motor) by means of a simple rheostat and the math is pretty simple too to figure out what Hz to set to get what RPM ... For fear of explaining the complete obvious, you can think of Hz (1/s) as the power supply turning on and off 60 times / s (at 60Hz) .. because of the fact you have 3 phases, their on / off are shifted by 120 degrees which gives the motor a direction (and instant reverse for that matter!) .. with single phase (it's a bit confusing because you have 2 single phases coming to your house, but they are 180 degrees apart and are considered 1ph not 2ph) if you output that to the motor it would just pull equally in both directions and not turn at all .. anywho, that's my understanding (and no deeper!! LOL) .. hope it helps somewhat .. you can get pretty decent cheap-ish 3ph motor on usedvic .. last one I picked up was a 1HP Baldor (essentially brand new) for 90$) .. keep us posted!

ps: if someone smarter than me catches me explaining cra** please set the record straight!

[roryn...@gmail.com]

Great job Ian and thanks for the explanation Matt.  I use VFD's all the time and have only been disappointed once.  On that one I discovered that the top end frequency was only 50 htz and that's probably why it was only $60.   Typically I like to program them to have a top end of around 70 or 75 htz.  I haven't encountered any problems with the motors and the difference in the speed range is significant.

The other option, if you can find it, is a permanent magnet motor.  They used to be commonly available for free from unwanted treadmills.  With permanent magnet motors voltage can be used to control speed using specialized control modules.  KB electronics makes a good one but it is expensive.  Some treadmills have MC 60 control boards and they work almost as well.

[Ian Paterson]

Thanks Mat & Rory.

Since I have a working motor already, maybe I'll focus on getting the lathe running and I can take my time deciding on how I want to proceed with the motor. I'm also considering the somewhat more unconventional route of using a brushless DC motor with a rotary encoder that can allow it to run in a closed loop configuration. A company called Odrive is making this a more affordable prospect than it used to be. Although the concept has been around in industry for many years, it's pretty new technology for hobby machinists.

Odrive controllers: https://odriverobotics.com/

Brushless motor for eBikes, etc: https://www.ebay.ca/itm/1000W-48V-DC-Brushless-motor-BY1000-Speed-controller-GoKart-Scooter-eBike-DIY/392766304326?_trksid=p2485497.m4902.l9144

Marco Reps on Youtube talking about Odrive and other options for closed loop operation: https://youtu.be/p4ltHDpxrbI

[roryn...@gmail.com]

Wow!  That Odrive system looks interesting, although I must admit that I don't understand most of it.  The motors look so small.  The thing I have noticed using treadmill DC motors is that mass seems to be important.  Treadmills seem to come in 2 varieties, ones with puny motors mass wise and big ones.  And the big ones work much better than the little ones.  I will be very interested to see how well the odrive system works if you choose to go that route.

[mickeyf]

"if someone smarter than me catches me explaining cra** please set the record straight!"

Not smarter, but perhaps a bit more knowledgeable.

Much in that explanation is misleading. Basically ignore everything in it other than "you put single phase North American mains power in, and get variable frequency 3 phase out, possibly also at a more convenient voltage". This is all you really need to know to use it in any case.

However, if you do care about the details, the VFD does NOT generate a third phase, for practical purposes it synthesizes all three phases - there is no 2nd phase that already exists. The nominal 240V line coming into your house is NOT 2 phase - it is a "single split phase", consisting of 2 hot legs that are 240V relative to each other, and a neutral that is 1/2 that value relative to either of them. Your heavy appliances (Oven, welder, baseboard heaters) are typically connected across the two hot legs, your regular lighting and outlets are connected between the neutral and one or the other of the hot legs.

Fig. 1 on this Wikipedia page illustrates this:

https://en.wikipedia.org/wiki/Split-phase_electric_power

Here is a link to a page with a graphic illustrating the difference between single and 3 phase:

https://www.electronicshub.org/difference-between-single-phase-and-three-phase/

[Ian Paterson]

For those of you who have a lathe with plain bearings, what kind of spindle oil do you use and where do you get it? Looks like I can get a gallon of Mobil Velocite #6 spindle oil from eBay for about $110 delivered, but I wouldn't mind finding something cheaper and more local if possible. Also, how important is it that I use a specially formulated spindle oil without detergents and sulphur? I've heard of some people using hydraulic oil mixed with grease to increase the viscosity, but is that advisable?

Ian

[David Wrate]

I bought a package of four oils from bluechipmachineshop on EBay. 

It included all the specified South Bend oils. 

David

On Nov 29, 2020, at 2:29 PM, Ian Paterson <ipat...@dominionblue.com> wrote:

For those of you who have a lathe with plain bearings, what kind of spindle oil do you use and where do you get it? Looks like I can get a gallon of Mobil Velocite #6 spindle oil from eBay for about $110 delivered, but I wouldn't mind finding something cheaper and more local if possible. Also, how important is it that I use a specially formulated spindle oil without detergents and sulphur? I've heard of some people using hydraulic oil mixed with grease to increase the viscosity, but is that advisable?

--
You received this message because you are subscribed to the Google Groups "Victoria Hobby Machinists" group.
To unsubscribe from this group and stop receiving emails from it, send an email to victoria-hobby-mac...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/victoria-hobby-machinists/4fcccfaf-c049-4b0c-baf9-a05c07122285n%40googlegroups.com.

[Admin]

I followed the directions in my Colchester Student manual and called Shell directly for equivalent (modern) products .. 

The manual called for: 

Tellus 27 which is now S2 MX32 (Part Number 550045429)
and Tellus 68 which is now S2 MX68 (Part number 550045415)

[Admin]

I had to buy 20GA pails .. but the headstock is using quite a bit and I'm planning on doing a couple of changes .. the 68 is considered the way oil for colchesters according to the manual (the Tellus 68 equivalent) and I have plenty .. I could easily part with .5l or 1l bottles if people were interested ..

[Ian Paterson]

Thanks David & Mat, I ended up buying a 5 litre jug of ISO 32 hydraulic oil. I still haven't run the spindle yet because the oil leaks out of the rear bearing just about as fast as I can pour it in. I think the main culprit is a lubrication channel drilled lengthwise through the top of the bearing. On one end, it exits into the steep-taper section of the bearing, where it's covered up by the bearing's rotating counterpart on the spindle. On the other end, the channel just ends in open air and this is where the oil seems to be pouring out. The front bearing has a similar oil channel, but the exposed end has a metal plug inserted. When I disassembled the headstock, I don't recall seeing a metal plug in the rear bearing, but it might have been plugged up with crud instead. There also seems to be a small amount of oil leaking from between the bearing and the headstock casting and in one of my gallery photos, I'm pointing to it with a scriber. The front bearing leak isn't nearly as bad as the rear, but it's enough to drain all the oil from the reservoir in a minute or two.

So my question is should I try to seal up these leak points and if so, how? I'm thinking of making a metal plug for the oil channel in the rear bearing and possibly using some kind of caulking in four "rings" to seal the bearings to the headstock casting. Will caulking even work? If anyone has comments or suggestions, I'd love to hear them.

Here's a link to the part of my photo album that shows the bearings. There's additional info in the comments next to each photo.

https://photos.google.com/share/AF1QipNPymVAoC7Sbaw0wM73pLoLD6ybI3z1dT4E4wsEBJw22lAtXiB_RoqJt1CX4PztzA/photo/AF1QipPlm0i4Hg81ExOY4lOtw1C1Lmw2e6QSVquHaIxj?key=SWdRWVJndWdCR3J3X3RHRnp4clFPdDR1QnFodmZR

Thanks,

Ian

[David Wrate]

Is there any possibility that those holes are for felts?

To view this discussion on the web visit https://groups.google.com/d/msgid/victoria-hobby-machinists/e8f38ac3-5580-447b-ae5a-ea760580d152n%40googlegroups.com.
<PXL_20201209_065445477.jpg><PXL_20201209_065403966.jpg><PXL_20201209_065636258.jpg><PXL_20201209_065412435.jpg>

[Ian Paterson]

David, that's a very good question. As I've worked on this old lathe, I've discovered many unconventional aspects of its design, as well as modifications done by previous owners with reasoning that's not totally clear. In light of that, I really shouldn't discount the possibility of it being designed for use with a wick or felt. If my bearings are supposed to be filled with a wick-type material, it would be horizontal, about 3mm thick and only one end of it would be in contact with the moving bearing surface, while the other end would be open to the air. Does anyone have an opinion on whether such a system would work? David, can the terms "wick" and "felt" be used interchangeably, or are they two different things?

I think the oilers that came with my lathe are "cup-type", but I'm not sure what that means. My oilers are just simple reservoirs with a 1-2mm hole in the bottom that lets the oil drain out quite rapidly. Is that how a cup oiler is supposed to work? I think "drip-type" oilers have some kind of valve that lets you adjust the flow rate, so maybe I should consider making a pair of those? Either way, oil is going to pour out that hole in the bearing if I don't put something in there to stop it or at least slow it down.

Finally, does anyone else have an opinion on whether I should try to use some kind of caulking or sealant to cover the transition between the fixed bearing and the headstock casting? (see photo in my previous post where I'm pointing with a scriber)

Xynudu's wick oilers for old Schaublin lathe:

https://youtu.be/EE32YUn77gw

Rolingmetal's wick/drip hybrid oiler:

https://youtu.be/UgSueO7R7wg

Steve Jordan's wick/drip oiler:

https://youtu.be/nmRKxqnqz0I

Thanks,

Ian

[David Wrate]

Me thinks wick and felt is interchangeable. Caveat being that my experience is about as deep as a puddle…. 

To view this discussion on the web visit https://groups.google.com/d/msgid/victoria-hobby-machinists/f0051e24-5b55-4118-bae8-fbbdbe1cf02dn%40googlegroups.com.

[Ian Paterson]

Just a quick update on the spindle oil situation. I plugged the hole in the spindle bearing on the change gear side with a little piece of aluminum. Now the oil doesn't dribble out like before, but the chuck-side bearing seems to weep oil from between the bearing and the headstock casting and I'm not sure what to do about that. I purchased a couple of rather large drip oilers which should at least allow me to regulate the flow of oil, but I wish it would leak a little less.

For now, I've moved on to troubleshooting vibration issues in the drivetrain....

Ian

[Ian Paterson]

I've run the lathe for a bit and tried making some test cuts. No great results so far; I got a lot of chatter which could be due to several factors. I know the 4 jaw chuck is bell-mouthed and I've only tried using the hand-ground HSS cutters that came with the lathe which are probably dull. I'm not too concerned about cutting performance right now though, because I got quite a bit of vibration from the drivetrain which I want to troubleshoot first.

My first job was static balancing the V-belt pulley and cone pulleys on the countershaft. I made a static balancing jig by mounting two parallel steel plates with honed edges in a vise and allowing the countershaft axle to rest on the top edge. the V-bely pulley only needed an 8g weight which I held in place with epoxy. The cone pulley on the other hand, was more severely out of balance. The casting was really lop-sided with 3-4mm ticker walls on one side than the other. To correct this, I bored out the inside so that it's concentric with the outside on both ends of the cone pulley which helped a lot. I was then able to epoxy a copper weight to the inside of the cone pulley which gave it almost perfect static balance.

When I put everything back together, vibration was significantly reduced, but not absent. I'm now turning my attention to the v-belt and the runout in the v-belt pulleys. the V-belt doesn't appear to have taken a set, although it is a rather conventional, old belt so I may replace it with a new one. Any suggestions on what kind of belt to get?

The pulleys have some runout which may also be contributing to vibration. The big pulley has about 0.002" of radial runout, while the small pulley on the motor has about 0.009". Axial runout on both pulleys is even greater still, but I assume the radial runout has the greatest impact on vibration. Both pulleys appear to be cheap die-cast metal and I'm wondering if a more true-running pulley can be obtained. If I can't find a better off-the-shelf pulley, then I'm considering either trying to true these ones up myself or making my own pulleys. I also happen to have an interesting pulley which uses a tapered, slotted hub which clamps onto the shaft when you tighten 3 bolts which draw the tapered hub into a matching taper in the pulley (see attached photo)

If anyone has thoughts on troubleshooting vibration issues in a v-belt system, I'd love to hear them.

[Ian Paterson]

I got the vibration issues mostly sorted out and "Unknown Lathe No. 129" is finally ready to be used.

Ian

[roryn...@gmail.com]

Nice work Ian!  It looks great.  I see that you found some different oilers or are those plugs?

[Ian Paterson]

Those are just the oilers that the lathe came with. No regulation of oil flow, just a hole that leads to the bearing. As it is now, I have to add a drop of oil every minute or two, but I plan to get the drip oilers installed soon. First I have to make a riser so the left oiler clears the backgear.

[Ian Paterson]

Ok, so I've come to the point where I want variable speed on my old #129 lathe. A while ago, Rory recommended DC motors and I was thinking of a more exotic brushless servo motor solution. Well, I've decided that's expensive and overkill for what I need, so I'm going with a 500W, 48V DC brushed motor from Banggood. The single phase AC motor that came with the lathe is only 1/2 HP, so I figure 500W is a slight improvement. The new motor comes with a bicycle chain sprocket, so I'm thinking of rigging up something so I can drive the countershaft with a bike chain. Depending on the torque and speed range I get, I may even try driving the lathe spindle directly. Thanks to Rory for planting the seed in my mind, I'll keep you all posted on how it goes (but don't hold your breath, I'm pretty slow these days!)

Ian

[Alex Kunadze]

Hi Ian,

I'd recommend against a DC motor. They're nice on paper, but the
torque they produce is dependent on speed, so you end up with very low
torque at low speeds. I have the ubiquitous Weiss 10x22 lathe and
G0704 mill, both with DC motors from the factory, and it's sometimes a
chore at low speeds. If you're set on DC, I would go oversize so that
that torque loss is less noticeable and definitely keep the pulleys
you have now so that you can fine tune the workable speed ranges.

The ideal motor (barring the super-expensive servos) would be a
3-phase AC with a VFD for speed control. They have constant torque and
are smoother than DC motors, resulting in better cut quality. Might be
hard to find in less than 1HP though.

I would also advise against a chain drive or direct drive. One of the
reasons belts are commonly used is that they dampen the
vibrations/pulsing from the motor. A chain would transmit those much
more, it'll show in the cut quality.

Cheers,
Alex.

> --
> You received this message because you are subscribed to the Google Groups "Victoria Hobby Machinists" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to victoria-hobby-mac...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/victoria-hobby-machinists/6d459c25-5f68-4e70-88ae-03649cf202f8n%40googlegroups.com.

[Ian Paterson]

Those sound like very good points, Alex. Since I've already acquired the hardware, I'm going to push forward, but won't keep my heart set on any particular arrangement.

Ian

[Rory Brown]

Hey Ian. Since Covid my source of inexpensive DC motors (free treadmills) has dried up and I’ve gone the 3 phase with VFD route. So far I’ve been more than impressed with this.  There is better low speed torque and simpler forward/reverse setup.   

On Thu, Jul 22, 2021 at 9:39 AM Ian Paterson <ipat...@dominionblue.com> wrote:

--
You received this message because you are subscribed to a topic in the Google Groups "Victoria Hobby Machinists" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/victoria-hobby-machinists/QrWMUn8WuqQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to victoria-hobby-mac...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/victoria-hobby-machinists/6d459c25-5f68-4e70-88ae-03649cf202f8n%40googlegroups.com.

[Ian Paterson]

Hey Everyone,

Thanks for all your advice regarding a new motor for my #129 lathe. I realize a DC motor may not be ideal and the reasons you've all given make sense. However.... I found a treadmill on the side of the road and the price was too good to pass up. It's got about twice the power of the e-bike motor I bought, it's 90V so it doesn't require a huge transformer and it drives a belt instead of a chain. I figured I have nothing to lose, so I decided to try installing it on the lathe. So far, I've machined an adapter for the countershaft pulley and I made a mounting bracket. Today I tested it with all the treadmill's original electronics and was surprised to find it all still works. My next challenge will be to find out how to replace the treadmill control panel with something more appropriate for a lathe - otherwise, I'll have to add a "calories" column to my feeds & speeds table ;-)

Attached are some photos of what I've done so far. Some of you might be amused by my rather sketchy Mini Lathe face plate work because I didn't want the hassle of setting it up on the CNC mill. For those of you who are considering a CNC conversion for your mill, you may be surprised how hard you try to avoid using it when you're done.

Ian

[Admin]

nice work .. it'll work great for you Ian .. with respect to the control panel .. you should be able to download a wiring diagram for the DC control board .. the multi-wire connector to the board should have three wires that essentially serve as a potentiometer for the speed control .. I've never been able to test this theory as I always got lucky to pick up older treadmills that still have a analog pot on them ..

[roryn...@gmail.com]

Hi Ian, If you can find an older treadmill with a dial or slide controller, it will probably have an MC 60 control board.  These are great and will run just about any DC motor. At least any that I have found.  Failing that there is a DC 51 control box.  I have used these with varying success.  They are only 400 kwt but some produce a lot more torque than others.  They are available on Amazon and eBay.  If you use Amazon and it doesn't work, you can always return it.  

https://www.amazon.ca/Controller-Modulation-Regulator-24V-90V-Adjustable/dp/B08SCDY4SY/ref=sr_1_5?crid=1B4G9QY9WXIMT&keywords=dc+51+controller&qid=1645988036&sprefix=dc+51+controller%2Caps%2C127&sr=8-5

I probably have one here that you could test run if you want.

Rory

[Ian Paterson]

Rory, thanks for the link and for the loaner offer. I may take you up on it, but first I'm going to see what I can do with the existing controller.

So far, I've learned the control panel sends a PWM signal to the power board, but unexpectedly, that signal doesn't directly control the motor. Instead, it seems to act like up/down volume buttons on your TV remote. A 50% duty cycle simply keeps the motor running at whatever RPM it's set for, but if you want to change speed, the control panel momentarily makes the duty cycle less than or greater than 50% which serves to decrement or increment the motor speed. When the new desired speed is reached, the duty cycle reverts back to 50%. If I manually try to slow down the motor, I can see the PWM duty cycle momentarily increase. If I try to maintain that load on the motor, I can see the duty cycle drop back down to 50% while my hand starts to burn. When I let go, the duty cycle momentarily drops to readjust to the lighter load, then returns once again to 50%. This behavior may address Alex's comment from July 2021, where he said that DC motors typically don't have much torque at the low end. The system I have seems to use closed-loop feedback to maintain a constant RPM under varying load, so I'm hopeful it will provide lots of low-end torque for threading, parting, form tools, etc.

I haven't yet put the scope directly on the motor because it's powered by rectified AC straight off the mains, so I had to cobble together an isolation transformer to make sure I don't blow anything up. To make the isolation transformer, I added some windings to the secondary of the transformer I was going to use with the ebike motor. See attached. 

My next step will be to look at the PWM signal right near the gate of the mosfet that drives the motor and see if I can find a way to simulate it with my own circuit. So far, my half-baked idea is to see if I can make some kind of H-bridge out of mosfets salvaged from old PC power supplies. I don't know yet if PC power supplies have the right kind of mosfets though. I guess I'm going down a rabbit hole, but I have a budget of $0 and no deadlines, so not much to lose. If any of our group members have experience with this kind of thing, I'd love to hear from them.

Thanks again Rory and Mat, I'll post again when I learn a bit more.

Ian

[Admin]

After your initial post I thought you were brand new to this .. so I thought my basically zero knowledge would help .. seeing your latest post I'll just watch and learn .. thanks for sharing!

[roryn...@gmail.com]

Hi Ian,  You clearly know a lot more about these controllers than I do.  My experience was that it was generally awkward to use the control panels form modern treadmills and I was unable to use the circuitry without them, unless I used an MC 60 from an older machine or the DC 51.  Low end torque was the greatest limitation. Now I have gone the 3 phase - VFD route for any restorations that I do that require variable speed.  I am quite happy with this, although I still have a couple of belt sanders, a wood lathe  and a wet sanding wheel that run with DC motors and DC 51 controllers.

[Ian Paterson]

Hi Rory,

I checked out the link for the DC 51 you posted on Feb 27. I can't see any connection for a speed sensor, so the controller must be open-loop and I would expect it to stall easily at low RPM.

...Right after I wrote the above sentence, I came across some info on using commutation spikes to measure motor RPM:

https://www.precisionmicrodrives.com/using-dc-motor-commutation-spikes-measure-motor-speed-rpm

So that's kind of cool, but I have no idea which controllers use that. Maybe the DC 51 does? Heck, maybe my import mini lathe and mill do. ...I should look into that.

Cheers,

Ian

[Ian Paterson]

I've been slowly working on the motor speed control problem and I think I've made some progress, although I've still got a few things to figure out.

In my first February 28 post, I said the control panel sends increment/decrement signals to the motor driver board similar to how the volume buttons on a TV remote work. It also used a reed switch and magnet to measure the speed of the treadmill belt pulley which allowed it to perform in a kind of slow closed-loop feedback mode, providing extra torque when needed under varying loads at low RPM. This pseudo closed-loop mode seemed great for low-end torque, but the need to vary the speed with increment/decrement signals didn't lend itself well to being controlled by a simple knob. Also, the treadmill control panel is big, bulky and its ergonomics are in no way relevant to machining, so I wanted to eliminate it altogether.

After some more tinkering, I discovered a new mode of operation on the treadmill motor driver board. If I disconnect the reed switch, the control panel enters what appears to be a kind of fail-safe mode where the PWM frequency drops from 240 Hz down to 40 Hz and in this mode, the PWM signal to the mosfet that drives the motor is more or less directly proportional to the duty cycle of this 40Hz signal. This means a simple circuit could be built to simulate this 40 Hz PWM signal and the motor speed could be controlled via a regular potentiometer. See attached photos.

Now I can control the motor speed with a knob, plus I've eliminated the control panel, but I've lost that pseudo closed-loop operation that I described earlier. There still appears to be some form of feedback built into the motor driver board however. In my second February 28 post, I mentioned using motor commutation spikes to measure motor speed, but after some more reading, I don't think that's commonly used. Instead, it looks like some (many?) DC motor driver circuits have a low-value power resistor in series with the motor and by measuring the voltage drop across that resistor, they can infer the motor's back-EMF, which should be proportional to the motor's speed. My motor driver board does indeed have such a power resistor and I can see a trace running from it back to the "brain" of the driver board, so presumably it's measuring the voltage across that resistor. The motor certainly doesn't feel like a closed-loop AC servo motor though. At very low RPM, I can stall the motor with my hand, so I'm not really sure what that system is designed to accomplish. I suspect my mini lathe and mini mill both use a similar system though.

Although I can control the motor speed with a knob, the response is quite slow. After turning the knob, it takes 4-5 seconds for the motor speed to reach its new value. I think there are two reasons for this. First of all, I'm not controlling the motor's PWM signal directly. Instead, I'm talking to a microcontroller that's reading the duty cycle of my signal, then commanding the driver circuit to send a much higher frequency PWM signal to the motor. Secondly, I think there must be some kind of "maximum delta RPM" feature built into the firmware that's designed to prevent people from being thrown off the treadmill because of a sudden change in speed.

My hacked treadmill motor speed control ought to be usable for machining, but ideally I'd like to improve the response time. Also, it remains to be seen how well it performs at low speed under load. I'm going to take what I've got so far and mount it on the lathe to see what happens….

Ian

[Alex Kunadze]

Hi Ian,

I'm not sure I'm understanding this correctly. Is it the control panel
that handles the feedback loop (with the magnet on the pulley)? Or
does it only handle the buttons and send up/down signals to the
driver?
If it's the latter it should be fairly simple to replace those signals
with buttons or a rotary encoder knob and a small MCU board. If it's
the former, then it's the actual motor controller and will be more
difficult to replace. Still possible with an MCU.

As for lack of torque at low RPM, that's the downside of DC motors. At
10% PWM it only gets 10% power, so... AFAIK the only way to fix this
is to gear it down, apart from switching to a 3-phase VFD-driven AC
motor.

Cheers,
Alex.

On Wed, Apr 13, 2022 at 10:27 AM Ian Paterson

> You received this message because you are subscribed to the Google Groups "Victoria Hobby Machinists" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to victoria-hobby-mac...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/victoria-hobby-machinists/e39e961e-1c40-4186-b8f5-716b7d9a4091n%40googlegroups.com.

[Ian Paterson]

Hi Alex,

It's the former. The control panel reads pulses from the belt pulley via a reed switch and sends a "speed increment" signal to the motor driver board if it senses a drop in RPM and it does the opposite if it senses an increase in RPM. If I disconnect the reed switch, the control panel enters a fail-safe mode and simply sends a constant 40 Hz stream of pulses to the motor driver board and the duty cycle of those pulses appears to be proportional to the PWM that drives the motor. The circuit I built exploits this behavior and that's how I was able to eliminate the control panel.

There's a PIC 16F684 chip on a daughterboard on the motor driver board and right next to it is a 5 pin header which might be for in-circuit programming (ICSP) but I'm not sure yet. Even if it is for programming, I have no idea if the manufacturer has left the chip in a readable state, but if so, I'll try to download and decompile the firmware. But for now, I'm just going to install it on the lathe and see how usable it is.

In the above photo, the header is labelled "JP2"

In the above photo, the PIC 16F684 is the 14 pin chip on the right (U2).

The above photo shows the numbers I read off the other two chips, but I haven't been able to find datasheets for them.

Folks, please forgive me for going down an electronics rabbit hole on a forum for machinists. It's still relevant if it's for a lathe, right?

Ian

[roryn...@gmail.com]

Don't stop Ian.  I find this all very interesting as I struggled trying to figure out how to use those treadmill motors without having to mount the whole control panel onto the machine.  Eventually I gave up and went the VFD 3 phase route.  Most of what you are saying is pretty much Greek to me, but I certainly appreciate trying to understand it.  Thanks for the details.

Rory

[Ian Paterson]

Thanks for your encouraging words Rory. I'm not fast, but I'll definitely try to uncover as much as possible and share it with the group. Also, I'm not an expert, so if anyone wants to chime in like Alex did, I'll gladly welcome the input. ...And if I stubbornly choose to ignore good advice, please don't take it personally. Sometimes I have to fail for myself before I learn. ;-)

Ian

[Ian Paterson]

Ok, so I'm at a point where I'm just going to use the speed controller as-is and focus on other stuff for now.

I'm able to control the speed with a potentiometer, but the control is still going through the treadmill's firmware which is introducing a kind of latency between turning the knob and having the motor reach its final speed. If you start at low speed and suddenly crank the knob all the way up, it takes about 5 seconds to reach full speed. I probed several test points in the circuit and confirmed  this delay is being introduced by the PIC microcontroller. With a scope, I can clearly see my 40Hz PWM signal going into the microcontroller and a 14KHz signal coming out, but although the output signal varies proportionally with my input, its rate of change is regulated and limited.

I was also able to confirm that JP2 on the motor driver's daughter board is for in-circuit programming of the microcontroller. However when I connected my PIC programmer to it, I could see that the code protection bits had been set, so I have no way to read the firmware. I thought about bypassing the microcontroller and generating the 14KHz PWM output signal myself, but I decided not to for a few reasons:

1. The daughter board is not isolated from mains power, so I would have to hack into the existing opto-isolators, or come up with my own to keep dangerous mains voltages safely away from the controls.
2. I'm not sure, but the microcontroller may play a role in overload protection and I wouldn't want to disturb that. Perhaps in the future I might find some kind of status signal that I could wire into an overload light.
3. Because I'm not using the treadmill's control panel, I'm not able to take advantage of the closed-loop speed feedback mechanism. This means the motor can be stalled if the RPM is low enough. Also, it doesn't have reverse and that would be a nice feature to have. I don't want to install a polarity switch on the motor because I think Bad Things might happen if the switch got flipped while the motor was running at full speed.

This all makes me feel that any more time and effort spent would be better put towards a completely new controller that can properly handle overloads, closed-loop speed control and reverse. ...Maybe some kind of H-bridge circuit? I need to learn more about those.

From a practical perspective, I've now got basic variable speed control and that's what I wanted. I'm going to use it like that for a while and figure out what I want to focus on next.

Ian

[roryn...@gmail.com]

Hi Ian, That doesn't look anywhere near as bad as a couple of lathes that I converted and actually sold as variable speed machines.  Mine had a lot more of the treadmill controls :). I routinely used DPDT switches to achieve forward/reverse functionality and never had any noticeable issues with the motors. I also used 4 way light switches and drum switches again with no noticeable issues.  

[Ian Paterson]

Rory, are you saying you've used polarity reversing switches and actually flipped one while the motor was running, or just that you've used them without problems so far? My concern is if you flip the switch with the lathe running, the inertia of all the spinning parts will turn the motor into a generator and send a large reverse-voltage back into the motor driver circuit.

...And a question for everyone: My lathe has tapered spindle bearings that are made out of some kind of ferrous metal. What's a reasonably safe maximum speed that I can expect from these bearings? With the pulleys I've currently got, the spindle tops out at a little over 2000RPM and I'm worried that might be too fast. Here's a photo of the headstock part of the bearings: 

Since there's no published info on my lathe, I'd be happy to hear from anyone with a similar-sized machine.

Thanks,

Ian

[roryn...@gmail.com]

I can not recall ever flipping from one rotation to the reverse on the fly.  Only that I've used these switches to facilitate forward and reverse.  Sorry I wasn't clear on that. 

[Ian Paterson]

Hi Everyone,

When I installed the treadmill motor on this lathe, it came with a plastic poly-v belt pulley that I adapted to the countershaft. Only problem was after I made a new hub for it, there was a millimeter or more of runout which created a lot of vibration when running. Because of this, I decided to make a new pulley out of aluminum, being sure to machine the bore and the grooves in the same fixture. It's made from 2 pieces of scrap aluminum bolted and glued together because I didn't have a single piece that was big enough. I also had to make a 40 degree, neutral rake cutter, which seemed to do a decent job as long as I kept the speed low and used ultra light mineral oil for coolant. I haven't installed it yet, but I'm optimistic that it'll be better than the plastic one, and I'm also hoping the 2-piece construction won't come apart on me!

Ian