Additive and Subtractive manufacturing

Hello to everyone reading. First time poster from northern California.
Iv bin reading for years and have to say thank you all for contributing to my cnc addiction. I have bin collecting a lot of robot guts for multiple builds. wile only 1/3 threw the treacherously long EG thread so I am very undecided still. I have contacted RAMPF about details on there epuself and it sounds promising and i may well have to give that a try if its not to pricey. Along with my x2 and my 3040 I have got 4 of the new 2015 series dyn4 with 750w 16bit encoders and 2 of the dyn2 with 400w 16bit encoders. I plan on buying 2 more of the 750w with brakes installed and 2 more of the 400 with brake and 2 without. So that will take some time to save up. Ill worry about rails and such after that. Its hard to argue with 16bit encoders and DMM is amazing to work with and has some of the best customer service iv ever dealt with. I tried to get a price quote from Parker for some nema 32 steppers for a plasma table some months ago and after 2 months of referring me through 4 different reps I cant say anything good. I could have already had steppers from china well before they ever got back to me lol.

So after a long time of contemplating different approaches I have found myself with questions I can't find answers to.

I plan on using 6 of the 750w servos. Going for the largest work envelope wile still getting high speed rapids pushing for 800 ipm. leaning toward a bridge mill using 25mm rails with 5mm pitch screws with double ball nuts maybe rack and pinion but id rather do some sort of anti-whip on the screws. But I still have no idea how big of a work envelope could be achieved with 6 servos. I know its overkill unless you build it right!
maybe 1 on x and 1 on z and 4 on y? or go vertical and use 2 on each?

I had contacted tormock to see if they would sell me a frame and to no surprise they rudely said no and that they don't sell frames to hobbyists. Novakon might have gone for it. while the Torus Pro has better specs in every way and has financing and servos as an option I don't think id be happy with a bench top machine in the end. So aside from bolting a bunch of dura-bar together and filling with EG which would probably cost an arm and a leg just to ship. An interesting thought I had on EG filled steel beams gathering that the biggest issue is that the steel structure is not as repeatable as iron do to thermal expansion being much grater with just a few degrees. What if you were to imbed some sort of piping for coolant in close contact with the steel or the steel shell actually being the coolant piping to prevent this?

Also looking more into tool and die methods from the view point of additive manufacturing of metal parts I feel as if plasma arc welding is superior to the current Direct metal laser sintering methods in a lot of ways. mainly PAW with wire feed won't leave powder inside a part. I have contacted a plasma arc welder manufacturer about this very subject. But what I'm trying to understand is why I can't get one of those cheapo 3 in1 plasma cutter/stick and tig welders and leave the plasma torch on in tig mode and add an outer jacket for shielding gas? lol like a hose with a valve? there are some companies that make an adapting black box to turn your tig into a plasma arc welder. If I were making money with a business doing this Id probably buy a plasma arc welder but for the purpose of testing and proof of concept a $200-300 piece of crap 3in1 is ok. I'm very surprised not to have found any topics on plasma arc welding on the zone given that it wood be one of the best ways to weld up a frame. plasma arc welding is also a lot more affordable than lasers and a lot safer.

The DMM dyn2 and 400w servos are going to be for a mach3 automated welder/additive manufacturing of some sort. But what im still trying to wrap my mind around is how to get modbus to control the welder and wire feed at the same time? Im not even shore modbus would be the best way to do it. Maybe something as simple as a reed switch attached near the poles on the wire feed servo might work. but then i think it would have to be off instead of holding position

Also thinking hot wire vs cold would have some advantages over passing the current threw the work piece since AM is much more like brazing and you don't need full penetration. Although I don't think it would work as well with aluminum since you seem to need alternating dcen-dcep to clean as you go. it also seems to make sense to me using very thin wire with micro plasma arc welding should help minimize heat input.

Sorry for the rambling.
Any input is appreciated.

Fwiw, laser sintering will finish to close or at net shape. Plasma welding is faster but much more crude. Laser sintering can also compensate for thermal stress while in build. So it all depends on what your looking for in the end.

So how many machines are you planning to build? If you started a thread for each one, it would be easier to understand what you're asking about. If you want a large mill for cutting steel, what most people here do is retrofit an old cast-iron mill with new electronics. That's a lot easier than starting from scratch. I don't understand why you'd want to use 4 servo motors for the Y axis, though.

If you want to build a machine that will additively produce metal objects, that's a whole other topic. I'd suggest you get a MIG welder and try to build something freehand with it, if you think that's a good way to go. Or try aiming a TIG at a pile of metal powder and see what happens, before getting involved in a project to automate these processes.

Quote:

---

Originally Posted by underthetire

Fwiw, laser sintering will finish to close or at net shape. Plasma welding is faster but much more crude. Laser sintering can also compensate for thermal stress while in build. So it all depends on what your looking for in the end.

---

Thank you for your response underthetire
Generally yes it does seem more crude. I think you could get grate results with a plasma welder like this Pro-Fusion: Micro Welding: TIG / Micro-TIG this welder will go to as low amps as 0.1. It also has a programmable soft start and soft stop that seems to be independent of the workpiece making me think it has some sort of pilot arc start or a non-transfer arc. I think if you were to use something like 2mm to .8mm wire you could get really close to a near net shape.
With DMLS the outcome is quite porous. From what I understand this is do to the powder not fully forming a melt pool. You could just as easily use wire with a laser but I just simply think there too dangerous and 3-4 times more expensive than plasma arc welding.

I don't know where you get that info. SLM is not porous at all if done correctly, in fact can be over 90% of billet strength.. Plasma would have splatter holes.

Quote:

---

Originally Posted by awerby

So how many machines are you planning to build? If you started a thread for each one, it would be easier to understand what you're asking about. If you want a large mill for cutting steel, what most people here do is retrofit an old cast-iron mill with new electronics. That's a lot easier than starting from scratch. I don't understand why you'd want to use 4 servo motors for the Y axis, though.

If you want to build a machine that will additively produce metal objects, that's a whole other topic. I'd suggest you get a MIG welder and try to build something freehand with it, if you think that's a good way to go. Or try aiming a TIG at a pile of metal powder and see what happens, before getting involved in a project to automate these processes.

---

I plan to build 2 machines of the same work area.

In a bridge mill as in a moving bed with fixed gantry design for the y axis is by far moving the heaviest load. While I don't know if I will use 4 servos on y axis on a bridge mill I could see some advantages. I'm not to interested in retrofitting an old machine for a number of reasons. if there was just a supplier of cast iron box tubing that would be a great option. if i can fined local iron ill probably do that but at the moment my local steel yard will only charge $120 for 12' of 2" 3/8 box tubing and its only 3 miles away. thay priced around $160 for the 12' 4" box tubing. cant remember the wall thickness.

To combat the expansion of steel causing loss of repeatability as apposed to iron on a milling machine.
One of the things I am trying to fined an answer to is why you couldn't use 1/2" to 1" box tubing welded together lengthwise to form channels for coolant that is also the outer frame with an epoxy granite core? so in affect the shell of the machine would be chilling itself to become stable.

I don't think Id use a bridge design on the machine for additive manufacturing of metal since there is no need to move the work piece around that is increasing in weight anyways so i will be using the 400w servos for that because its only moving the torch. I don't think id ever spend my money on a MIG. plasma arc welding which uses a similar if not very the same inverter as a TIG. Pro-Fusion even makes an adapting box to convert a TIG. Hear is a patent I found on the subject Patent US20100258534 - Method of converting a gas tungsten arc welding system to a plasma welding ... - Google Patents
Plasma arc welding has bin in practice since the 50s and has amazing results over most alternatives.

Hear is another paper on plasma arc welding being used for additive manufacturing. Haven't red the hole paper yet but I don't think there using a micro plasma arc welder. Looks like there using a thermal dynamic unit. INVESTIGATION IN THE USE OF PLASMA ARC WELDING AND ALTERNATIVE FEEDSTOCK DELIVERY METHOD IN ADDITIVE MANUFACTURE | Abdullah Alhuzaim - Academia.edu

thank you for your input.

Quote:

---

Originally Posted by underthetire

I don't know where you get that info. SLM is not porous at all if done correctly, in fact can be over 90% of billet strength.. Plasma would have splatter holes.

---

I'm shore it can be performed better by some than others but ultimately other than adding drain holes for the powder in your cad model your kind of limited in the internal structure with powder beds regardless of plasma or laser. I don't know how true that is that you will get splatter with plasma arc welding. with plasma cutting absolutely it will happen. but the plasma jet stream is scaled for the needed psi but the flow rate is different. there is 2 sets of argon flow in plasma arc welding. the plasma jet and the shielding gas.

I understand both process quite well. Like I said, it all depends on what your looking for in the final product. SLM is not appropriate for home use anyways, nor hobby use. Not only is it prohibitively expensive, its also a very dangerous operation. Been a few near deaths that I know of from explosions. Most of the larger players of SLM still don't have a real good handle on the safety aspects yet.

Just a little update on progress. I managed to buy the duel arc 80 on the used market as well as the water cooler and cables . I still need to test and make shore its all working. Then buy the wire feeder. there are products at arcraftplasma.com similar in specs that go as low as 0.5 amp while the duel arc 80 can go as low as 0.2 amp and 250 pps. it Is also programmable for automation. The one listed on arcraftplasma can weld at 100 microns. with 0.023"-0.025" grade 2 TI or steel welding wire from mmc I think this could work out well. the proprietary technology of pro-fusion is very clear so I will have to omit certain details. surely I will post any prints when that happens. I think this unit will work out well with DMM ac servos.

Hello again, I created a new account because I lost my password do to my pc bsod and the "help" center will not help... tried everything.

Im trying to find a way to make the wire feeder oscillate as it feeds the wire aside from editing the g-code line by line. one solution that could work is to have the rollers on a pneumatic cylinder so that the wire can still feed at the beginning of a weld and retract at the end. but this might create a lot of vibrations.

I thought about using an arduino to switch the state of the direction pin by using the direction pin itself with a dwell at the beginning of the sketch so it can feed enough to weld and still retract but that looks like it could mess with the timing and fee to much or too little. Also the arduino might not do to well with the HV/HF start.

I was thinking of using an ssr to switch the welder on but I don't know if this is even necessary if my bob is opto isolated. Id think the best way to do it is how some wire in their plasma cutters.

The welder came with the PMB-50 Machine Plasma Torch witch look to be in good condition aside from needing a new cup.
The Dinse style connector on the torch says 16-25 printed on the side but it measures at just under 9mm and way off from 3/8". Is his normal? I came across some 10-25 connectors that said they were 9mm.
I'm planning on getting a ck9F tig for it soon so ill have to figure out what Dinse i should get.
It would be really appreciated if someone with more knowledge on this could help me identify whats going on with this connector and what might be a good one to pick up.
hear are some pictures of the torch and Dinse connector
Attachment 307908
Attachment 307910
Attachment 307912
Attachment 307914