[charlieslasher]

I am fascinated by horizontals. After building a fixed gantry machine for aluminium. I am thinking of building a new machine.

Can/has a horizontal been built Diy?

Minimum requirements for the rotary part of the pallet are there direct drive options available off the shelf?


Is it really the best idea having the majority of the machine mass in the base and not where the spindle is connected? i guess once you go big enough this is less of a problem.

[hjl4]

One good thing about horizontals is they seem to be much more rigid. Also having gravity help clear the chips away from the cutting area is also a plus. If you decide to build it, it will certainly be interesting to see what plans you have in mind, and following this thread.

[underthetire]

The bigger you go up near the head, the more mass you need to move. This design is similar the the Mori NH series, except they use double screws and motors on Y and Z.

Sent from my G-Tab Quantum using Tapatalk

[RICHARD ZASTROW]

To you question "are there direct drive options available?" Yes there are some. There's tha few questions there. At what price? Is it practical? are there better options?

You can buy direct drive rotary motors from Bosch-Rexroth, Kolmorgen, Etel and others. These are the motors we have used to drive rotary motions.

There are linear motors being used for (obviously) linear motions. We haven't used them because of small steel chips messing up magnets. Vertical applications present safety problems of free-fall in the event of power interruption.

Servos directly attached (bellows type couplings) to good quality, accurate ball screws (no belts) have worked fine. We always use accurate, good quality resolvers as well.

Dick Z

[underthetire]

We have multiple linear motor machines at work, all for diamond turn. Mori did make a large HMC with linear motors, but afik, never actually sold one. Had to have massive hydraulic system to stiffen up the axis. That's the problem with linear motors, they can position quick, and feed very smooth, but have a hard time holding enough stiffness to compensate for heavy cutting, and as posted before, don't do well in heavy contamination environments.

Sent from my G-Tab Quantum using Tapatalk

[wizard]

I've contemplated building one for my little shop. The idea being suitability for wood, composites and aluminum, it is just that it is expensive to put a shop together and other things have priority. In my case I wasn't even thinking about a rotary table, maybe an indexing table but even that isn't a strong need.

I am fascinated by horizontals. After building a fixed gantry machine for aluminium. I am thinking of building a new machine.

Horizontals certainly have interesting capabilities. While a horizontal is a ways off, I've looked into the possibility of a moving column machine, mainly due to the problems of handling large pieces.

Can/has a horizontal been built Diy?

Well obvious no in my case

Minimum requirements for the rotary part of the pallet are there direct drive options available off the shelf?

The minimal requirement is do you really need one? It is an extra complication. The second question is do you need continuous positioning or would indexing be good enough. Indexing 90 degrees at a time is fairly easy and can be done with a mechanism that locks in pretty good.

Is it really the best idea having the majority of the machine mass in the base and not where the spindle is connected?

Roughly yes. The problem with mass is that it does no good at all if it is in the wrong place. This is where I have issues with people trumpeting mass but not thinking about where the mass is going. A good engineer would use various design tools to determine where the weak spots in a machine are. If you aren't an engineer (like me) all you have is experience to go on. As has been noted by others already in this thread horizontals are known, in most cases, for being stiff machines.

i guess once you go big enough this is less of a problem.

I might disagree with that, the bigger the machine is, the more problems you have with mass. You can't make the machine out of giant slabs of cast-iron or what ever because it ends up too heavy to be a realistic solution much less cost effective. So with the larger machines you have to carefully design them to put the right amount of material in the right place. The picture that accompanied this posting (not sure if it was yours or the system adding it) is a good example of a large machine with some engineering time put into it as you can see extensive use of cores to minimize iron usage.

If you are doing a small DIY mill for a home shop you might use whatever steel tubing or plate steel you can come up with. Mass may still be important but due to size and materials you can only put so much into the machine.

[cyclestart]

Can/has a horizontal been built Diy?

I don't recall seeing a diy hmc posted in these forums. It would be awesome to see someone try !

It's a design that may be hard to scale down.

[charlieslasher]

Ok good points so far.

I only require it to machine aluminium at fastest rate possible. Is a smaller spindel doing 24k rpm ever going to compeat with a larger slower spindel 15krpm or less with a bigger tool profiling and roughing 80% of the material.

I dont see the point in not using 4th axis positioning. This is my main requirement. Some kind of lock and indexing required. Continuous 4th not essential and softwere is $$$$



Machine frame from EG cast on surface plates for ways.

What controller would handel this. I love the price of mach 3 but I still havnt found a setup that close to industrial. Second hand 840d
Simens controller?


All this for under 20k? Dream? Accuracy ideal if it could make injection moulds.

[wizard]

Ok good points so far.

I only require it to machine aluminium at fastest rate possible. Is a smaller spindel doing 24k rpm ever going to compeat with a larger slower spindel 15krpm or less with a bigger tool profiling and roughing 80% of the material.

15K is plenty fast. However the real answer is probably more involved, it have to wonder if your items to machine are big enough that you can achieve reasons able axis velocities to actually make use of the high RPM. I would think that a very high RPM spindle would add more cost to the machine than you might be willing to pay.

I dont see the point in not using 4th axis positioning. This is my main requirement. Some kind of lock and indexing required. Continuous 4th not essential and softwere is $$$$

If it is important to you the obviously it has to be designed in. I just wanted to point out that plenty of horizontal type machines exist without such tables.

Machine frame from EG cast on surface plates for ways.

I'm not sure about this from the strength standpoint. In my mind it would be easier to use epoxy granite as a fill for steel tubing.

Speaking of machine that are homebuilt horizontals I've seen at least a couple of machines in my lifetime. At the extremely small end you Have the Gingery machines built out of aluminum. These certainly aren't what you are looking for.

A few years ago I was at NAMES and a guy there had a home built CNC set up that was if very reasonable size. I'd call it more than a benchtop but not really a floor model. I was actually very impressed with this machine but I never did make contact with the owner. Whomever it was had a good sense of quality and economics. The machine appeared to be made mostly out of steel tubing and plate. The machine was big enough to machine a small engine block ( I think it was an engine block) this was several years ago so everything is fuzzy.

What controller would handel this. I love the price of mach 3 but I still havnt found a setup that close to industrial.

Mach 4 is in beta and appears to be a big step forward. If you want a fourth axis though I'd seriously take some time to spend on the LinuxCNC site as there controller has been implemented on a number of novel machine designs. In fact I beleive somebody did a conversion on a huge horizontal with a big tool changer. Realize though that with LinuxCNC you will likely need some programming skills if your design falls of the beaten path.

Second hand 840d
Simens controller?

I'd shop around. Siemens does make nice stuff but I think it is a bit early to latch on to a controller and frankly there are lots of good choices out there.

All this for under 20k? Dream? Accuracy ideal if it could make injection moulds.

For under 20K, that would depend entirely upon the size of the machine. I honestly doubt it can be done at that price for a largish machine.

As for an injection mold making machine I don't see a lot of horizontals doing mold work but that might be due to limited exposure. I'd be seriously concerned about stiffness with the mold base parts mounted on a tomb stone. This also makes me wonder why you would have to have an indexing third axis, is it to support drilling and porting operations for water and so forth? When you first mentioned the extra axis I was thinking ganged up parts for production machining which makes sense. The problem is I would suspect that the fourth axis will add a substantial expense to your machine which may be a problem with a 20K budget. You can but all sorts of off the shelf indexing dials but few of them are really designed for use on a machining center. As such you dial may end up being a DIY design that would need to address the stiffness required in a machining center. I just don't see it as being a trivial part of your machines design.

So in a nut shel what you want to do isn't impossible. Everything from tiny manual horizontals to rebuilt industrial machines have been tackled by the DIY crowd. Over the years I've come across other horizontal projects so maybe some Internet searching will turn up something. In the end I see feasibility constrained by size. The more you blow out the size of the machine the more likely that you will end up needing the services of a machine shop which of course adds to the cost. I'd strongly suggest looking towards the used market for a clapped out horizontal that can be had at scrap metal prices. You may end up replacing a lot of parts but it will save you a lot of time and you end up with a substantial hopefully well designed frame. A rebuild allows you to focus on the controls and linear motion components.

[charlieslasher]

Sorry it is for nesting lots of the same part. But the odd injection mould would be needed.

The main goal of the machine is to bridge the gap between startup manufacturing to being able to buy a machine new.

I like the idea of buying an old machine and refitting thus most of the hard work is done. Finding the right machine might take a while though.

I currently use a 24k 3hp spindel and can machine at 3-5m/min but still have long cycle times. Alot of parts are small with 3mm holes 6mm roughing finishing tools. But i dont know if i need more HP and a slower spindel, better programing or stock water jeted pre machining.

Could a harmonic drive be suitable for driving the 4th axis and some kind of brake.

I really wish i could just buy a new/real machine but have to start some where. And building machines is some what fun once its cranking out parts.

[cyclestart]

Sorry it is for nesting lots of the same part

Alot of parts are small with 3mm holes 6mm roughing finishing tools

I'm picturing these parts nested vertical on a tombstone where the parts can be machined on more than one side(angle) in one set-up.
If there is a different reason for the rotary table I'm lacking the imagination atm.

A horizontal mill and a horizontal machining center (hmc) are different animals. A linuxcnc hmc retrofit can be found by searching cnczone member samco or samcoinc on youtube.

A converted horizontal mill could get a bit cramped on the z axis for operations like drilling and tapping. As always it depends on your needs.