[asuratman]

dfro,
Let me know your block's dimension so I can imagine the lathe's size.

[BobWarfield]

Re the problems adhering, I did not see any mention of degreasing the metal surfaces. Maybe I just missed it, but my experience filling my mill base with E/G is that any kind of grease or oil (and probably a lot of other contaminants) are problematic.

I'd look for some solvent-type degreaser that will evaporate completely. Epoxy, after all, can be used to join metal and should not delaminate so easily.

Cheers,

BW

[dfro]

Bill,
Thanks. I purposely chose a small project, so that my big mistakes are not that big - start slow and small, they say.
I have an Atlas 12" lathe bed that is 52" long. It is also waiting to be turned into an e/g-iron lathe. I am very glad I did not have to drill out the e/g and chip off the epoxy surface plate on that one!

Asuratman,
My workbench is 72" (6') long and 16" wide. The lathe bed is from a standard Atlas 6", which I think is 36" long. The headstock is a standard cheap import 7x12 headstock. I have increased its LxWxH with e/g.

Bob,
I did forget to mention that I went over both parts repeatedly with naphtha solvent. On the lathe bed, I ground away all of the surface that touched the mold with the angle grinder, anyway. I also went over the bottom of the headstock with some sandpaper. I will edit my earlier post to add that.

Thanks,
Dave

[miro]

It is a shame you didnt succed to make an epoxy surface plate since it looked like it leveled realy flat. I belive epoxy to metal bond failure could be easily overcomed. At least i had positive experiences bonding steel especialy with West System but a couple of things should be kept in mind.
First is Amine blush - a thin layer of amine formed on the surface after the epoxy enters the gel - faze(especialy developed in a cold or humid environment).New layers of epoxy cannot stick to old ones covered with Amine layer. Since epoxy makes no chemical bond (except with its self and that before gel-faze) with its substrate it relies entirely on mechanical bond. So a good surface preparation is a must. First one should rub a hardened epoxy with scotch bright and water (plenty of it), sand for good adhesion, and then clean with water again. Also degreasing ( clean industrial aceton or similar) should take place especialy if steel surface was soked with oil previously (used lathe ways..).
There are couple of tips from MOGLICE Handbook:


Sandblasting has been proven to generate the best adhesion surface, althoughit is not practical for largercomponents.
Devitt Machinery also has a deoiling powder used in rebuilding, primarily to act
like oil dry; it draws excess oil out of old oil soaked cast iron. The deoiling powder
will change color when it absorbs oil and should be reapplied several times until
it stops changing color. Gentle heat can also be used to sweat oil out of a casting.




As for mechanical bond - even a very corse sandpaper (80 or 60 grit) rotating on a high RPMs will not produce deep enoug cuts to ensure good holding key. One should make a series of angle cuts on the surface to ensure good adhesion. Also When aplying you can rub a first layer of clear epoxy with a steel wire on the surface to ensure it entered all the small cuts and holes.



Miro

[dfro]

Miro,

That is very good advice. I think the surface plate idea can definitely work.

The owner of the grinding shop talked to me about how porous cast iron is - that old cast iron machine tools have soaked up oil like a sponge.

Thanks,
Dave

[RotarySMP]

Thanks for posting all this detail of what is working and what not. The E/G thread is great, but highly theoretical. Your practical examples are a big help to understand E/G for us hobbyists.

That lathe bed looks outstanding cosmetically. Nice work.

What was the plan for the epoxy levels lathe ways, if the adhesion had worked? Were you then going to mount a ground steel way on top for a saddle to run on, just run the saddle on the epoxy, or are you mounting linear guides?

If a most machine shops would will willing to do a grinding job like this for only $60, then this method of machine manufacture is easily in reach of hobbyists.

My local hardware store sells granite blocks for steps, for about $20. Two blocks epoxied together with steel plates glued on, and then ground perpendicular and flat would make a killer small mill frame.

[dfro]

My next step was to attach and adjust the linear rails. That took a couple of hours. I tested the adjustment of the rails against the ground outside surface of the ways with a dial test indicator. I think I got within a couple ten thousandths.



Linear rail - headstock end.



Linear rail - tailstock end.



Top view of the bed with the attached linear rails and headstock



Here is the machine with everything bolted together. It is one solid mass of e/g and iron (except for where the spindle passes through the headstock). I will weigh it to see how heavy it is at this point.

I am going to paint it green.

***

The next step is the machine base. I wanted it to be massive and cheap. So, I thought I would experiment with cinder block. I have made two columns of cinder block - one for each foot of the lathe bed. The two columns are also tied together with two 1/2" threaded rods that span the gap. The two rods could support a shelf, if I want.

Each column has three stacked cinder blocks and a cap block. The two hollow cells in the blocks have rebar and poured concrete in them. This bonds them together into one solid mass.

Each block cost $1 and I used about $7.50 in poured concrete. This is a cheap way to create a massive, vibration damping base to hold up your machine tools. I have almost finished making them. I will post pictures, soon.

Thanks,
Dave

[dfro]

Mark,

Thanks, for the kind words.

Thanks for posting all this detail of what is working and what not. The E/G thread is great, but highly theoretical. Your practical examples are a big help to understand E/G for us hobbyists.

I want to share all of the small details of what I am trying, because I think the details are what are going to snag people. You really want everything figured out before you cast your parts, otherwise you might end up with a very heavy paper weight! I am also giving a lot of details, because I want advice from others on how I can do better.

My advice is to start slow and small - make big mistakes on small parts.

What was the plan for the epoxy levels lathe ways, if the adhesion had worked? Were you then going to mount a ground steel way on top for a saddle to run on, just run the saddle on the epoxy, or are you mounting linear guides?

My intention was to mount linear rails to the epoxy surface plate on the lathe bed. I do not think table top epoxy would have the durability to withstand direct contact as a bearing surface. Also, the bottom sides of the ways have no precise reference to the top surface plate. So, there would not be a way of using the bottom surface to accurately clamp a carriage.

If a most machine shops would will willing to do a grinding job like this for only $60, then this method of machine manufacture is easily in reach of hobbyists.

I think E/G is wonderful stuff that will allow anyone, if they are willing to learn and practice, make highly accurate machines at a relatively low cost. I also like the idea of locally sourced materials.

My local hardware store sells granite blocks for steps, for about $20. Two blocks epoxied together with steel plates glued on, and then ground perpendicular and flat would make a killer small mill frame.

I think you have a good idea. If you are uncertain about the flatness of the granite surface, you could rough it up with a grinder, level it, and cast an epoxy surface plate. Before doing that you could mark the attachment points for linear rails, drill over-sized holes, and glue in threaded inserts with toughened epoxy. Plug up the holes with some modeling clay before casting the surface plate. You could do some great experiments for very little money with these granite blocks.

I think I am going to cast my benchtop mill base in e/g, so that I can give it a more complex shape - make recesses for the leadscrews, for example. I also like the idea of casting the threaded inserts into the part rather that drilling holes in granite. Drilling rock is not fun! And I want the protruding steel bars (the foundation for linear rails) to be keyed into the e/g base with bolts - like how I cast the headstock.

There are a lot of great possibilities.

Thanks,
Dave

[Brenck]

Hi Dave,

Great work keep up the updates.

You where concerned about clamping the carriage and tailstock, did you ever check this thing out? http://www.automation4less.com/linearbrake.htm

Its what you need, fits linear rails and it wont happen the deforming of the carriage itself when you clamp everything, plus no need to worry about the unprescise reference of the bottom within the top since it holds by the linear rail...

Hope this is a solution

[dfro]

Brenck,

Thank you for the information! That is a great idea. I have been scratching my head on this for a long time, but I never even considered clamping onto the linear rails. I really like the feature that the clamp puts no stress on the bearing blocks.

The clamps are $90+ dollars a piece, which seems a bit expensive to me. I am thinking that it would not be very hard to make a pair of these out of a soft metal, like brass. I would want to be careful about where on the rail they are clamping, because I am concerned about wear.

Thanks again,
Dave

[dfro]

Here are some pictures of me building the base for the lathe. It is made of concrete blocks, concrete caps, rebar, steel tubing, and 1/2" threaded rod. The lathe feet will bolt down into coupling nuts, which are glued into the concrete caps with toughened epoxy - 'jb weld'. I cast the columns upside down, and turned them over when the concrete cured. They are very heavy; but, I was able to turn them over without any help from a hoist (or a chiropractor).



Dry stacking everything, and testing that the rebar is the correct length and that the steel tubes fit through the holes I drilled in the block.



The rebar keys into the two holes in the cap.



Mid-way through casting one of the columns.



The two finished columns.

The threaded rod will pass through the steel tubes in each column, and will be secured with nuts and washers. This will tie the two columns together. I am going to do some cosmetic plastering of the outside, paint them, and pour an epoxy surface plate on top of each column.

Thanks,
Dave

[tootalew]

I might have an answer to why the epoxy did not stick, it could be the naphtha. The naphtha may be leaving an oily residue behind. Next time you try this, you may want to use something different, or maybe pass your blow torch over the area once the naphtha dries. Did you know that you can use naphtha in your zippo lighter, much cheaper than buying lighter fluid.

[dfro]

One of the next steps I am puzzling over is how I am going to power the lathe spindle. Little Machine Shop has a 250 W motor and controller for around $150. For $300 I could have a 350 W motor and controller.

But, I think I would like more power - maybe 500 W (2/3 hp). I would also like to be able to control the speed from the computer. Here is what I might purchase:

I am thinking of driving the headstock with a 24VDC electric scooter motor. It is capable of handling 500 Watts:
http://www.monsterscooterparts.com/24vo500wamow.html

I could power it with a 24V@20A unregulated power supply:
http://www.mpja.com/prodinfo.asp?number=6645+PS

And, I am thinking of using this controller:
http://www.robotpower.com/products/simple-h_info.html

The computer or a microcontroller will feed the PWM signal to the motor controller board.

I am also wondering what size timing belt to use to drive the spindle - T5, 15mm wide? I will buy from here: http://www.sdp-si.com.

I am also scratching my head on what my top spindle speed should be? I see metalworking lathes usually advertised in the range 1,800 to 2,000 for the top speed. The motor does 2600 RPM, so I am wondering if I should gear down the spindle speed.

Any advice or ideas would be great.

Thanks,
Dave

[Thazul]

Hey Dave,
Many people have used this combo:

Controller:
https://www.surpluscenter.com/item.a...8&item=11-2269

There are a lot of different treadmill motors on ebay, might want to shop around, but here are a couple:
http://cgi.ebay.com/Nordic-Track-Tre...item20af344d46
http://cgi.ebay.com/18A-TREADMILL-PE...item3efc570345

-

Andy

[RotarySMP]

My CNC'd 7x12 uses a 550W 3Ph 2 pole motor driven by a VFD.

The motor drives the spindle through about a 1:3 9mm HTD pulley set.

I have the VFD set up for about 10Hz through 100Hz, which gives a range of spindle speeds from about 250 to 2500.

I am pretty happy with this in general, although there is not much torque at lower speeds (this is relative - it still has way more grunt than the original 250 chinese Watt motor in low gear).

As an example of the limits, turning an M40x1.5mm thread in steel was not successful, even when taking dozens of passes. This was due to the combination of needing a deep enough cut to not rub, slow enough spindle speed for my slowish Z axis to keep up, and limited torque at low speed.

On my set up, 2500 rpm top sped is enough, as my Z is not fast enough to practically use higher speeds. If it had a faster Z, and more spindle speed, then aluminium could be turned down to swarf even faster.

VFD/3Ph motor combinations are really nice to use, as they have large speed ranges, are quiet and smooth, the motors are sealed, designed for industrial duty, and easy to control from a CNC with an interface like Homans.

The ideal for me, would be a choice of two ratios between motor and spindle, 1:9 will give you a speed ratio of about 80-800 with oodles of grunt, while a near 1:1 will give you 600-6000. A 3/4 HP set up like mine would be plenty.

I have considered making a lay shaft setup to do this, but have limited space, and no driving need at present.

[dfro]

Andy and Mark,
Thanks for the feedback.

Mark,
Thanks for the rpm numbers on you lathe. That gives me more of a feel of what I should be shooting for. My shop has 3 phase wiring at the breaker box, but I do not think I want to deal with it. I would like to be able to just plug the lathe into a 110 VAC outlet. I also would like to hang the motor off of the headstock, and those 3 phase motors look way too heavy for this. I may change my mind after studying it some more, though. I also have an Atlas 12", 52" long lathe bed that is waiting to be turned into an e/g-cnc lathe. Is there a way to control the speed of your machine from the computer?

Andy,
The treadmill motor/controller idea looks like a possibility. There is such a wide variety of voltages and speeds. I would like the ability to buy new motors easily, becuase if I fry it, I would like to find the same motor with the same rmp's. But, it is frustrating to try and get specs on new motors - the sellers on the web do not tell you the voltage rating, amps, rmp's. All they want to say is, "motor for the Nordic Track X1000". On ebay, the two main voltage ranges seem to be 90VDC and 130VDC. I am wondering if I buy a 130VDC motor and run it with the Minarik controller at 90VCD, whether I will get a big drop in torque.

Also, I wonder if the motors are sensitive to being driven by a higher voltage. What would it do to a 130VDC motor if I ran the controller at 180VDC?
I see that the Minarik is controlled with a 10K pot. Do you know what the voltage is on that? It seems that to control from the computer, I am going to need one of those PWM to 10V conversion boards.

I am still considering using the lower voltage scooter motors - either 24VDC or 48VDC. The motors cost from $70-180, and they go up to 1000 Watts of power. On Jameco's website I found a lot of switching power supplies that can give me up to 1000W of power at 24VDC and 48VDC. The scooter motor controllers are relatively inexpensive, too. I would just need a PWM to 5V conversion board to run the motor from the computer.

I like that the parts for the scooter motors are easy to find and the specs are straightforward.

Thanks,
Dave

[dfro]

I ended up getting a 90VDC motor and the Minarik controller.

Thanks,
Dave

[Thazul]

I did a quick search of Minarik on the board here, and there seems to be plenty of people using it, so support here should be good
If I'm not mistaken it looks like LeeWay used it on both his Mill and Lathe.

Looking forward to seeing yours in action

Best Regards,

Andy

[RotarySMP]

My Lathe VFD runs off single phase. I provide three phase variable frequency to the motor. That photo I posted distorts the size relationship, as it was shot with a wide angle lens pretty close up. The motor is not that big or heavy.

[dfro]

Mark,
Could you tell me what VFD unit you are using? And, is there a way to control it from the computer? I would like to learn more and possibly use a VFD on the next cnc lathe.

Thanks for the help guys,
Dave