Moray's Harrison Retrofit

[m_c]

After much procrastinating, I've finally decided to physically do something about getting some CNC functionality in my workshop.

The first candidate, is a Harrison Knee Mill, that I've had stood doing nothing for the past couple of years. It's originally an Auto Cycle machine, which means it already has a ballscrew fitted to the X-axis. I don't yet know if this will be useable or not, as I do know that there is play in the X-axis, but I'm not sure if it's in the screw, or the nut mounting.
The mill is essentially a horizontal knee mill, which came with the option of a vertical milling attachment. I've only got the one vertical head to share between two mills, so it'll be remaining on the working mill until this mill has been cnc'd.

At the current stage, I've removed the table, and knee. The knee/table assembly will be getting weighed, and a suitable air cylinder + regulator fitted to act as a counterbalance. The knee itself is going to be the Z-axis.

At the moment, the current plan is to use servo's, but this may change depending on what motors I find.

I've attached some pictures of the mill. The first is the mill before any work commenced. The secon and third are the mill in it's current state.

I'll post up a exploded view of the z-axis nut design, with explanation later.

[Mcgyver]

that looks like a very nice size mill for conversion.

how does the air cylinder counter balance work, does it keep a slight upwards pressure on the screw, and if so since air is compressible, what keeps it rigid? i'm afraid the ball is slow to drop on this one for me.

[m_c]

Mcgyver - The air cylinder counterbalance works by pushing the knee upwards, reducing the load placed on the ball screw. The knee height will still be precisely controlled by a ballscrew.


Here's the cad drawings of the z-axis set-up. Due to the design of the knee, and the lack of room to mount a motor/belt, I've had to opt to have the ballscrew fixed to the knee, and the motor spinning the nut.
Blue = ballscrew + nut (No thread shown as I couldn't figure that feature of Alibre Design out - anybody care to tell me in simplified terms???)
Red = Toothed belt pulley (dimensions are for a 80t)
Yellow = Bearings (in this case 7210 singular angular contact bearings)
Grey = parts to be machined (except the bolts obviously)

The mounting plate (which will bolt to the machine and also have the motor mounted on it), will be clamped between the two bearings. The adjustment of the bearings will be done via shims between the spacer and the machined tube that the ball screw nut is mounted into.

[m_c]

This retrofit is off to a bad start

Went out this afternoon to put the knee back on, so I could measure it up and get the ball screw and nut ordered. However, after putting the knee back on, I realised there wasn't going to be enough room to mount the ball nut the way I wanted.
I had been looking at my good mill, where there is plenty room when the knee is down, and thought this one would of been the same. I knew the two mills had different bases (good one is a newer welded steel base version, whereas the other one is an older one with a cast iron base), and the old mill had an inch less knee travel, but I didn't think they were that drastically different.

I could of hacked the cast iron boss of the base, and then bolted a steel plate on to mount the nut, but I've decided I'm going to build the old mill up into a useable manual mill, and then retrofit the newer mill. This gives me the benefits of a steel base which will be easier to modify/add bits to if need be, and the cnc mill will have the extra inch of clearance under the spindle.

I'll post up some new photos tomorrow to highlight the differences.

[m_c]

Here's the photos showing the difference in the bases.

The old cast iron base has two raised bosses, one where the mill body bolts on, and the other where the z-axis nut bolts on. When fully lowered the knee comes down below these bosses.

The new welded steel base has only the one plate steel boss (looks about an inch thick), which the mill body and z-axis nut bolts onto, and the knee rests on the same plate when fully lowered. Also the mill body is considerably higher, which gives over an inch extra clearance under the spindle.

I'm now of to rebuild various bits and get the old mill fully functional, so I can start retrofitting the newer mill.

[m_c]

Not had much time to play with the milling machines the past week, but I have got the old mill partly re-assembled.

I did get a pnuematic cylinder for the counterbalance of ebay, and have been looking at ways of fitting it in. Due to the size, I think it's going to have to go on the front of the knee, with a section cut out the front of the base for it to fit in close enough.

[m_c]

2 and a bit years on, and this project hasn't progressed much, although it has done a couple laps of the workshop while trying to make space for other stuff, so I figured it's about time to do something about it. That, and all the other projects seem to of cleared up at last!

I've spent the past few months thinking about how to find a workable solution for the Z-axis, and have finally got a design that should fit quite comfortably in the space available (drawing attached)
As the plans stand, I'll be using a 25x5mm ballscrew, with a RSB style nut, running on a 3208 double row angular contact bearing. Not the most ideal solution, as the nut is going to be exposed, but it should fit in the available space without too much hacking of the original metal.

Drive for the Z-axis is going to be a stepper. It's a 6Nm motor from Arc Euro, which I've been tripping over for the past few months.
I've also gathered up the makings of a powersupply (transformer, bridge rectifier, big capacitor).

The original plan to counterbalance the knee with one big cylinder at the front, has been changed to using two smaller cylinders mounted on each side of the knee. This way everything is kept a bit more compact, and I won't have to worry about fitting the Y-axis motor around the air cylinder.

I also ordered a SmoothStepper and C23 BOB from CNC4PC last week, so hopefully it should turn up sometime this week.

My aim for this week is to get the required 3208 bearing ordered, and start machining the housing for it, as I've already got the metal waiting.

So this conversion should see some good progress over the next month.

[Al_The_Man]

Personally I am not in favour of Z axi'ing the Knee, I do have an Excello where the Knee is positioned in this manner for initial positioning only.
The air system uses a pneumatic balance valve, that keeps the pressure constant whether traveling up or down.
The lift ball screw is compact as it closely resembles the Nook Ball Screw Jacks, which have a right angle worm to attach the motor to.
When using these items in a vertical direction pushing up, the backlash is usually taken out.
I did also successfully use the Nook product on a Huge Avey Mill/Drill Z axis that had about 400lbs of lead counterweight.
The BS Jack replaced about 8 Hydraulic valves that operated the Z feed cylinder using a binary code to fire them for different feeds.
Check the Nook site for info.
Al.

[m_c]

Z'ing the knee is the only option with this mill.
I agree it's not the ideal solution, but it's the only one available.

I'm needing to get the knee/table of so I can get a better idea of the weight that will need counterbalanced. I guess the table weighs around 50-60kg going by the amount of effort required to lift it, but I have no idea how much weight is in the knee itself.
Going by a rough estimate, I'm looking at two 40mm air cylinders, which would give me about 150kg of counterbalance at 6bar.


I've so far managed to get the bearing housing machined, and it now just needs some bolt holes drilled and countersunk into it.
I'm holding of starting to machine the ballnut carrier until I have the ballnut, and decided how I'm going to secure it in the bearing. The drawing above uses a hollow bolt approach, but I'm considering buying a suitably sized nut instead.


Also got a customs charge card through the post today, so the Smoothstepper and BOB have finally arrived in the country. Hopefully get them picked up tomorrow.

[m_c]

Not much has been happening with this over the past week.

I ordered the Z-axis ballscrew and nut last week, so they should hopefully be here sometime this week. Then I can get on with machining the ball nut holder, which has so far seen the chunk of 90mm bar stuck in the lathe and faced. I don't want to start machining it too much, as I've got another job to do in the lathe before I tie it up for a couple of days.

I've also ordered a Gecko G203V for the Z-axis.
I was going to order 3, but given the speed that this retro is likely to go at, I'm hoping the new Stepper PID drive is out before I need drives for the other axis. If not, I'll just order some more G203Vs, as I've got ideas in my head for the next cnc project.

I have spent the weekend sitting looking at the C23 BOB and SS with a blank look on my face, trying to figure out what I need, and what goes where. The CNC4PC manuals aren't exactly comprehensive, so you need to piece all the bits of info together from the various parts of their site.

However, I've now got a rough layout sketched for the control box, along with lists of possible components to go in it, and how certain parts will be wired up. I now need to start work on a wiring diagram, and tie all those snippets together. Then I can start ordering the remaining parts for the basics of the control box.

[andypugh]

Any progress on this? I am part-way through the exact same conversion:

Harrison Universal Miller Conversion